Michael Dürr
A PRELIMINARY RECONSTRUCTION OF THE
PROTO - M I X T E C TONAL SYSTEM*
Un sistema tonal m u y e l a b o r a d o es una de las características
del m i x t e c o . La t o n o l o g i a c o m p a r a t i v a muestra una diversificac i ó n dialectal m u y p r o n u n c i a d a y — a primera vista — p a r e c e
tener p o c a s reglas en c o m ú n . El a r t í c u l o trata de reconstruir el
sistema
tonal
del p r o t o m i x t e c o .
Cada m o r f e m a
tenía
una
c o m b i n a c i ó n de d o s t o n o s : * a l t o - a l t o , * a l t o - b a j o , * b a j o - b a j o y
* b a j o - a l t o . A d e m á s , el cierre glotal final, h o y d e s a p a r e c i d o en
la m a y o r í a de l o s dialectos, se ha m a n i f e s t a d o en c a m b i o s de
t o n o . Basándose en esta r e c o n s t r u c c i ó n se esclarecen m u c h a s
diferencias dialectales.
0.
Introduction
0.1
0.2
The Hypothesis
Phonological Inventory of Proto-Mixtec
1.
Reconstruction o f Tone
1.1
1.2
1.3
1.4
Proto-Mixtec Basic Tonemic Couplets
Proto-Mixtec Non-basic Tonemic Couplets
Proto-Mixtec Tone Sandhi
Problems Unsolved
2.
C o g n a t e Sets
This article is a revised and translated version o f the central parts o f m y M. A . thesis
s u b m i t t e d t o the Free University o f Berlin in spring 1 9 8 2 .
I N D I A N A 11, Berlin 1 9 8 7
INDIANA 11 (1987): 19-61
ISSN 0341-8642
Ibero-Amerikanisches Institut, Stiftung Preußischer Kulturbesitz
19
0. Introduction
The tonal system, especially the complexity o f tone sandhi rules, is one
o f the most characteristic features o f the Mixtee languages.' At first
glance, there seem to exist in each dialect arbitrary distinctions between
morpheme classes causing tone sandhi o f following morphemes and
morpheme classes not causing tone sandhi. On the other hand, there is a
seeming lack o f correspondence between the tonal systems (viz tonemic
couplets and tone sandhi rules) o f the different dialects. Inner-dialectal
splits and dialectal divergence, both loose their arbitrariness in the light o f
comparative and historical linguistics.
ft 1 The Hypothesis
The Mixtee dialects o f Ayutla and Santa Maria Zacatepec preserve a
morpheme-final glottal stop. Concerning this glottal stop, Pankratz and
Pike ( 1 9 6 7 : 287 - 2 8 8 ) assumed:
" I n f a c t , the role o f m o r p h e m e - f i n a l and w o r d - f i n a l glottal s t o p in m o r p l i o t o n e mics is so i m p o r t a n t that A y u t l a M i x t e e m a y b e c o n s i d e r e d t o preserve here an
archaic feature o f c o n s i d e r a b l e i m p o r t a n c e t o the understanding o f the d e v e l o p m e n t o f M i x t e e m o r p h o t o n e m i c s in its m a n i f o l d dialectal variations."
Others - Daly ( 1 9 7 3 a : 103) f o r the dialect o f Peñoles, Pike and Wistrand
( 1 9 7 4 : 84) for the dialect o f Xayacatlán, Bradley and Josserand ( 1 9 8 2 ) in
their reconstruction o f Proto-Mixtec — suggested this connection, t o o . But
up to n o w no comparative work was done to prove it.
It will be shown that this suggestion comes right to the point. There is a
high degree o f correspondence between Ayutla forms having a final glottal
stop and cognate forms from other dialects causing tone sandhi on a
following morpheme. The m o r p h o t o n e m i c realization o f tone sandhi is
similar, too. Therefore it is plausible to connect the existence o f m o r p h o tonemically distinct word classes with the presence or absence o f a final
glottal stop that has been reconstructed for Proto-Mixtec f r o m non-tonal
evidence (Bradley and Josserand 1982).
From the historical point o f view. Mixtee tone sandhi might be explained by "floating t o n e " as in a number o f African languages (Schuh
1978: 253 - 254). Proto-Mixtec morpheme-final glottal stop became a
tone or tonal ghde - such transformation o f a final glottal stop into tone
is well attested for a large number o f tone languages all around the world
(Hombert 1978: 92 - 9 3 ) :
20
" T h e e f f e c t o f a glottal s t o p o n the p i t c h o f the p r e c e d i n g v o w e l is w i d e l y
attested. By the sixth c e n t u r y , glottal s t o p s had disappeared in V i e t n a m e s e and
were replaced b y rising tones ( H a u d r i c o u r t 1 9 5 4 , M a t i s o f f 1 9 7 3 a ) . In the L o l o Burmese f a m i l y , Burmese high t o n e c o r r e s p o n d s t o J i n g p h o glottal s t o p (Maran
1971),
and
Lahu
high
rising
tone
developed
through
glottal
dissimilation
( M a t i s o f f 1 9 7 0 ) . Mei ( 1 9 7 0 ) had s h o w n that M i d d l e Chinese í / z a « ^ s/jen^ (rising
t o n e ) c o m e s f r o m a final glottal s t o p . "
The supernumerary character o f such a third tone in the tonemic couplet
explains its instabiUty. This instable supernumerary tone was phonetically
realized either as overlapping o n t o a following m o r p h e m e as "floating
t o n e " in non-prepause position or, in some dialects or dialectal forms,
pushing away the second tone o f its o w n tonemic couplet.
0.2 Phonological Inventory of
Proto-Mixtec
The canonical structure o f Mixtee morphemes is in most cases bisyllabic. The description given by Kenneth Pike ( 1 9 4 7 : 1 6 6 - 169) for the
dialect o f San Miguel seems to be valid f o r all dialects (cf. Hunter and Pike
1969: 33, note) and bisyllabic forms have been reconstructed f o r all
Proto-Mixtec (independent) morphemes. As to tone, most descriptions o f
tonal systems from modern dialects use Pike's term " t o n e m i c c o u p l e t "
(Pike 1948: 79). T o n e m i c couplets will be reconstructed f o r Proto-Mixtec,
too.
Reconstructing the complete tonemic couplet o f a morpheme and not
the tones syllable per syllable in isolation has first been used in the description o f Proto-Mixtecan tones by Longacre ( 1 9 5 7 : 93 - 112).
The distinction between basic tonemic couplets, i. e. couplets without a
modified tone restricted to tone sandhi, and non-basic couplets with
modified tone is also taken over from Longacre ( 1 9 5 7 ) — tonemic c o u plets with a modified tone are unpredictable and can be explained as tone
sandhi variants o f one o f the basic tonemic couplets o f Proto-Mixtec,
quite the same situation as in Proto-Mixtecan.
Proto-Mixtec has first been reconstructed by Mak and Longacre ( 1 9 6 0 ) .
Recently a new reconstruction based on a large amount o f data collected
all over the Mixteca was done by Bradley and Josserand ( 1 9 8 2 ) and
Josserand ( 1 9 8 3 ) :
21
CHART 1: PROTO-MIXTEC PHONEMIC INVENTORY
(Josserand 1983: 238 - 276)
A. Phonological units
t
k
k"
"d
S
X
i
i
U
e
a
0
1
n
V
y
: Y : V
:
obligatory tone
W
B. Monoraorphemic couplets*
(C)VV
*
/ VV /
The
phonetically
is
treated
occur
(C)VCV
as
feature
in
C
consisting
V»
of
nasalization
final
^
two
i\ >
but
should
be
mora.
)
can
only
syllables.
T o facilitate comparison of the reconstructed forms given in part in
Josserand ( 1 9 8 3 ) and in part done by myself, I will try to follow strictly
this system o f reconstruction - o f course I assume complete responsibility
for any deviations or incorrectly reconstructed forms.
22
1. Reconstruction of Tone
LI Proto-Mixtec Basic Tonemic Couplets
Longacre ( 1 9 5 7 ) reconstructed three basic tonemic couplets f o r an
older stage o f Mixtee: *mid-mid, *mid-low and *low-mid. A s to a possible
couplet *low-low Longacre ( 1 9 5 7 : 106) argued:
" L o w - l o w d o e s n o t o c c u r in M-SM n o r 4 4 in M - S E -
t h e r e f o r e p r e s u m a b l y PM
did n o t have a * l o w - l o w t o n e p a t t e r n . "
But Longacre t o o k into account no more than t w o dialects (San Miguel el
Grande M-SM and San Esteban Atatlahuca M-SE) - therefore, by using a
greater number o f dialects^ a somewhat different and more symmetrical
system seems probable:
CHART 2; PROTO-MIXTEC BASIC TONEMIC COUPLETS
*HIGH-HIGH(')
»LOW-HIGH
»LOW-LOW(')
»HIGH-LOW
Note that *high corresponds to Longacre's *mid and *modified (a tone
restricted to tone sandhi, see b e l o w ) to Longacre's *high. It seems to be
preferable to reconstruct only t w o phonemic tone registers plus a feature
tone modification following Daly's (1973a, 1977) analysis o f Peñoles
tonal system because o f the non-high reflex o f the modified tone in a
number o f dialects (see below). Final glottal stop seems to be restricted to
tonemic couplets with identical tones. It is treated as part o f the tonemic
couplet because it determines tone modification.
The following chart will give the reconstructed tonemic couplets and
their reflexes in modern dialects:
23
FROTO-MIXTEO
Molinos
»HIGH-HIGH'
iiiiiiiiiiiniiiiiiiniin
mid-inid(M)
*HIGH-HIGH
»HIGH-LOW
mid-mid
mid-low
Ocotepec
mid-mid
mid-mid
mid-low
Jamlltepec
mid-raid
mid-mid
mid-low
Atatlahuca
mid-mid
mid-mid
= mid-mid
mid-mid
mid-low
mid-mid
mid-low
mid-mid
mid-mid
mid-mid
= mid-mid
Silacayoapan
(San Jerónimo
Progreso)
Mixtepeo
high-low
mid-mid
mid-mid (kir
ship terms:
low-low ?)
CVCV: fml-rlh
mid-mid
mid-low
CV'V: mid-high
Alaoatlazala
(fml-mid)
mid-low
(low-mid)
mid-mid
Metlatonoo
mid-low-mid
mid-high
low-mid
mid-mid
mid-mid
(-low)
Ayutla
low-low"'
low-low
low-low
mid-low
(mid-mid)
(low-high')
24
Fenoles
low-low(M)
low-low
low-high
Diuxi *
low-low(M)
low-low
low-'low
Jicaltepeo '
low-low
low-low
V: high-high
high-low
mid-mid
Coatzospan *
low-low
high-high !
I low-low
»LOW-LOW
»LOW-HIGH
PROTO-MIXTEC
Molinos
low-mld
Oootepec
low-mid
Jamlltepec
low-mld
Atatlahuoa
low-mid
San M i g u e l
low-mid
Cacaloztepec
low-mid
Xayaoatlán
low-mid
Cahuatache
low-mid
low-low
low-low
Silacayoapan
(San J e r ó n i m o
Progreso)
Mixtepec
low-mid
low-low
low-low
low-mid
CVCV: l o w - r l h
low-low
(rlm-low)
CVV:
low-mid
low-low
Alacatlazala
Metlatonoc
::
low-mid(-low)
low-high
low-low
N (CVCV): l o w - l o w - m i d
(cVV):
mid-low-mid
V:
mid-mid
low-low
mid-low'
mid-low
high-low
high-high(M)
high-high
Diuxi
low-'high
high-'high(H)
high-'high
Jicaltepec
high-high
high-high
high-high
Ayutla
Peñoles
(high-mid)
Coatzospan
N: !fhl-hlgh
V: h i g h - l o w
A: h i g h - h i g h !
high-high
25
CHART 3: DIALECTAL REFLEXES OF PROTO-MIXTEC BASIC TONEMIG
LEGEND
COUPLETS
^final
glottal
stop
^^high-high''
Special
mid-low
^'high -low
->•
mid-mid
^ low - low
->•
mid-low
reflexes
with:
nouns
Num
numerals
V
verbs
CVCV •
CVV
CVV ^
(
Tonal
)
adjectives
couplet
rare
glides:
forms
reflex
h
high
m
mid
tone
low
tone
1
r
rising
tone
f
falling
rlh
glide
£ising
fml
glide
falling
(M)
tone
from
from
l_ow to
high
mid
Vaw
to
modification
* In Diuxi,
Jicaltepec
and Coatzospan
the
between
the reflexes
of Proto-Mixtec
tonemic
with
and without
glottal
stop
is
not
founded
on
data.
Note
that
Diuxi
tonemic
couplets
are given
to
the
reanalysis
of
Daly
(1978).
The
indicates
nonpredictable
word-final
stress,
a feature
of
tone.
The exclamation
mark in Coatzos pan reflexes
a process
phoneme
causing
downstepping
terrace
26
modifi(M)
-Í-
N
A
tone
cation
distinction
couplets
sufficiently
according
apostrophe
which
is
indicates
tone.
As for * low-high, there is a spHt into two distinct couplets in several
dialects. The non-regular reflexes seem to be arbitrary in each dialect and
with no clear correspondence in one o f the other dialects.^ I cannot
explain this split, although a number o f these forms should be analysed as
compounds. The same may be true for some reflexes o f *high-low, e. g.
Mixtepec fml-mid and Alacatlazala mid-mid.''
As for *high-high and *low-low, there is evidence for the influence o f
final glottal stop on the development o f tonemic couplets. Forms without
such glottal stop tend to have a modern reflex o f the Proto-Mixtec tonemic couplet distinct from that o f forms with Proto-Mixtec final glottal
stop. The dialects Mixtepec, Alacatlazala and Metlatonoc mostly show
reflexes o f *high-high'' and *low-low'' having a final high tone or tone
rising to high. Ayutla has at least in part a reflex low-high'' o f * highhigh'' and in other dialects final high tone occurs in rare and non-predictable cases.® Note that this split is clearly evident for *high-high'' but only
in part for *low-low'' .
The distinction o f [ + m o d i f y ] (perturbing power) vs. [ — m o d i f y ] in
Peñoles and Diuxi seems to correspond to the presence or absence o f
Proto-Mixtec final glottal stop.® The split of some couplets into distinct
perturbation classes in Jicaltepec seems to be related to this Proto-Mixtec
feature as well, although the partial merging o f some Proto-Mixtec couplets (basic and non-basic) in this dialect complicates the analysis. Jicaltepec morpheme classes high-high (sandhi class I) and low-low (sandhi
class I) resulting from Proto-Mixtec *low-low'' and *high-high'' respectively contrast with high-high (sandhi class III) and low-low (sandhi
class II) resulting from Proto-Mixtec * low-low and *high-high.^
Coatzospan seems to merge *high-high'' and *high-high as low-low, and
*low-low'' and *low-low as high-high (isolation variant high-rlh).®
A number o f dialectal reflexes should be explained by older subphonemic variants o f tonemic couplets. The reflexes San Miguel and Atatlahuca
mid-low and Metlatonoc mid-low(-mid) o f *low-low show that this ProtoMixtec tonemic couplet had no plain but a falling contour. The same is
true as to the reflexes Xayacatlán and Cacaloztepec mid-low o f * highhigh"' and, in a number o f dialects, mid-mid o f *high-low, the contour in
both cases being [mid-mid-low].
In some dialects there is a tendency to differentiate the reflexes o f
verbal and nominal forms and to reduce the number o f distinctive tonemic
couplets o f verbs. This could be the result o f marking verbal aspect mainly
by tone in these dialects, but there are not enough data to give clear
evidence. The possible influence of morpheme-structure and segmental
phonemes on the tonemic couplet is another unsolved problem. Lexical
fields, too, may be o f some importance, e. g. numerals all have a tonemic
27
couplet *low-low (the number four * l o w - l o w ' ' ) , some animal names and
several kinship terms have either *high-low or *low-high.'
There is a high degree o f correspondence between the reflexes in the
dialects San Miguel, Atatlahuca, O c o t e p e c , Molinos, Cacaloztepec, Xayacatlán, Chayuco, Silacayoapan, Mixtepec, Alacatlazala, Cahuatache and
Metlatonoc. These dialects form dialect area A.
The dialects Peñoles, Jicaltepec, Diuxi and Coatzospan show a clearly
distinct system. The reflex o f * low-low is at a higher level in the register
than that o f * high-high. The couplets changed places. These dialects will
be called area B.
The dialect Ayutla has an intermediate position between area A and B.
This split into two distinct dialect areas with tone couplets reversed in
the register is a problem which will be discussed in 1.4.
1.2 Proto-Mixtec Non-basic Tonemic Couplets
Non-basic tonemic couplets are characterized b y the presence o f at least
one modified tone. Modified tone differs from the unmodified ones as
follows:
a) In a large number o f dialects, modified tone is at a higher level than
every unmodified tone and there is a merger o f modified low and m o d i fied high. This is the case in dialect area A and in Ayutla.
b) In the other dialects (i. e. area B) tone modification lowers high to
mid and mid to low tone. The derived couplets are characterized by at
least one lowered tone and their formation follows more complicated
rules. But this is a historical statement - judging f r o m the published data
it seems to be difficult to distinguish basic tonemic couplets and non-basic
ones phonetically in these dialects.
There will be no reconstruction o f Proto-Mixtec morphemes with a
non-basic tonemic couplet (this seems to be misleading in m y o p i n i o n ) but
only o f forms having a sandhi variant o f one o f the basic tonemic c o u plets. This I practice because I assume that most, if not all non-basic
tonemic couplets in modern dialects have been caused by regular tone
sandhi at an older stage o f Mixtee. This older stage might have been in
some cases, but not necessarily, Proto-Mixtec and, rarely, even ProtoMixtecan. Being made up in most cases o f c o m p o u n d words, they lost or
fused their first element (the one producing tone sandhi) and became, as
result, a seemingly non-composite word with canonical bisyllabic f o r m .
The only distinction to non-composite morphemes is the preserved tone
sandhi caused by the lost or fused first element (Dürr 1984). I o f f e r the
following arguments and facts in favor o f this hypothesis:
28
1) In most dialects o f area A , non-basic tonemic couplets cannot be
perturbed by regular tone sandhi, or at least have restricted or special tone
sandhi. (They cannot be perturbed because they already are perturbed! )
2) The reflexes o f the non-basic tonemic couplet o f a given f o r m are
not h o m o g e n o u s enough to be derived from one peculiar Proto-Mixtec
tonemic couplet reconstructed as non-basic. The postulation o f shortened
c o m p o u n d forms could explain this inconsistency — the individual nonbasic couplet is the result o f any older stage o f dialectal tone sandhi. Such
a f o r m , whether shortened o r not, followed the changes o f dialectal sandhi
rules f o r a long time, up to the m o m e n t it became frozen and o b s o l e t e . ' "
3) A great number o f non-basic tonemic couplets are to be found in
words which might be explained through evident c o m p o u n d s being not or
only partially reduced alternate f o r m s in the same dialect or through
forms o f other dialects (Dürr 1984).
4 ) A t least Ayutla shows a special, optionally progressive or regressive
tone sandhi involving morphemes with non-basic tonemic couplets (Pankratz and Pike 1967: 2 9 7 ) :
" T o n e s e q u e n c e s 3 1 , 3 3 1 , 3 1 3 , and 3 2 1 w h i c h are Class B act m o r p h o t o n e m i c a l ly as t h o u g h p r e c e d e d b y a p r o c l i t i c w i t h t o n e b u t n o segmental p h o n e m e s . "
This proclitic should be explained as the lost or fused first element o f the
compound.
Non-basic tonemic couplets therefore are to be explained by tone
sandhi rules and will be discussed in the following paragraph.
1.3 Proto-Mixtec
Tone Sandhi
Proto-Mixtec tone sandhi is based o n the fact that in non-prepause
position morpheme-final glottal stop became a " f l o a t i n g " tone modifying
its environment. Proto-Mixtec had t w o different forms o f tone sandhi:
1) Following a non-prepause m o r p h e m e ending with a glottal stop there
was a regular perturbation o f basic tonemic couplets changing the couplet
to a derived one having at least one modified tone. In this position forms
without a final glottal stop had tone sandhi variants with their first tone
altered, the second unaltered, i. e. tone modification lapped over o n t o the
first syllable o f the following m o r p h e m e and created a modified t o n e . ' '
Forms with a final glottal stop might have had the same change, although
the more frequent sandhi variant had both basic tones changed to m o d i fied tones, i. e. the modifying power o f the final glottal stop was realized
as modification o f the second tone o f its o w n tonemic couplet, and not as
29
CHART
PROTO-MIXTEC TONE SANDHI
basic couplet
basic couplet
sandhi variant
"modified-modified'
'^modified-modified f-
/
/
/''high-high'
A
fV'high-high
I
L
(
I
I
1
i
1
I
L
1
:
«modified-high'
> »modified-high 4
¡'«low-low'
*'modified-low'
I
I
"modified-low
'high-low
,
/
-I «low-low'
modification o f the first tone of the following morpheme. I cannot explain this ambivalence of *high-high'^ and *low-low'' concerning their
sandhi variants — in some cases there is evidence even of two modified
tones in forms not having final glottal stop.
LEGEND
(CHART
5)
H
M
high
3
intermediate
mid and low
mid
low
tone
tone
30
The analysis
of
Alexander
(1980).
res
pect.
between
(Atatlahuca)
tone
regular
sandhi
special
sandhi
restricted to
«
tone
Atatlahuca
tone
Mak
(1953)
rule
rule
special
sandhi
differs
(e.g.
in
optional
or
constructions)
follows
some
CHART 5: COMPARISON OF DIALECTAL TONE SANDHI TO HIGH (AREA A)
non-perturbable
tonemic couplets
Cacaloztepee
basic tonemic
couplets
basic tonemic
couplets
HH
HM
HL
Xavaeatlán
HH
HM
HL
Ocotepee
Atatlahuca'
HH
MH
HM
MM
HL
ML
HH
(MH)
HM
(MM)
r*
HH
HLH <-
LH
HM
HLM
LM
-> HL
LL
Í
H3 •V' H ^
HL
Hetlatonoc
HH
HH
LH
MH
(LH)
HM
HL
Ayutla
HH
HL
HHL
•»• HL
MH <••
LH
ML
LL
31
2) In special, close-knit modifying constructions (mainly noun + m o d i fying noun, adjective or verb), a special tone perturbation o f the tonemic
couplet o f the modifying m o r p h e m e to *modified-modified (rarely to
*modified-high or * m o d i f i e d - l o w ) t o o k place in contrast to regular tone
sandhi in the same way as described b y Pike ( 1 9 4 8 : 8 2 - 8 7 ) f o r the
dialect San Miguel el Grande. This tone perturbation occured in these
special syntactic constructions whether the first m o r p h e m e ended with a
glottal stop or not. This second form o f tone sandhi might be explained
by deletion o f *xáá'¡' "relative marker" combined with retained tone
sandhi o f the following modifier.
For practical purpose, I will discuss in the following summary o f dialectal sandhi systems areas A and B separately.
AREA A (inclusive Ayutla): Morphemes causing regular tone sandhi to
high in San Miguel, MoUnos, Cacaloztepec, Xayacatlán, Metlatonoc and
Ayutla, or special tone sandhi to high in Atatlahuca and O c o t e p e c have
had final glottal stop in Proto-Mixtec. Dialects and dialectal f o r m s not
having this perturbing power but being also a reflex o f a Proto-Mixtec
morpheme with final glottal stop have in most cases final high tone or a
final tone rising to high. The effect o f the supernumerary tone was realized o n the last syllable o f its o w n tonemic couplet and therefore no
longer could shift o n t o the following morpheme.
As can be seen from chart 5 the sandhi rules being the result o f ProtoMixtec final glottal stop vary only in detail — area A shows but minor
variants o f the Proto-Mixtec tone sandhi system.
Most other sandhi rules may be explained through the tonemic couplets
o f Proto-Mixtec or another older stage o f Mixtee dialect evolution. Mak
( 1 9 5 3 : 9 2 ) suggested that Atatlahuca morphemes having a lowering influence o n t o the tones o f a following m o r p h e m e historically ended on low
tone, e. g. the subclass o f Atatlahuca mid-mid with lowering influence
goes back to Proto-Mixtec *high-low. The tonemic couplet mid-mid(-low)
in Metlatonoc is the regular reflex o f Proto-Mixtec *high-low and, ahke,
causes tone sandhi to low.
Some o f these sandhi rules (and reflexes o f Proto-Mixtec) should be
considered as the result o f older subphonemic variants. In O c o t e p e c (Mak
1958: 70) and in Xayacatlán (Pike and Wistrand 1974: 8 4 ) a low tone
following another low tone in a couplet has a lower allotone. T o suggest
such falling o f tone in couplet *low-low('' ) f o r an older stage might
explain the reflexes San Miguel and Atatlahuca mid-low, Metlatonoc
mid-low(-mid), Ayutla m i d - l o w ^ ) and tone sandhi rules changing the
reflex o f *low-low to mid-low in Metlatonoc and Molinos. The reflexes
Xayacatlán and Cacaloztepec m i d - l o w ( m o d i f y ) o f *high-high'' , the rule
changing Xayacatlán mid-low to high-mid and the regressive change o f
32
Cacaloztepec mid-low (with following perturbation to high) to mid-mid
show that *high-high'' must have had a phonetic realization [mid-midl o w ] at least in the ancestor dialect o f Xayacatlán and Cacaloztepec. The
rule changing mid-low to mid-mid after preceding low tone in Cacaloztepec and the reflex mid-mid o f *high-low in other dialects show that
*high-low at least in some dialects had the same phonetic realization
[mid-mid-low].
Non-basic tonemic couplets high-mid and high-low being reflexes o f
older sandhi variants *modified-high and *modified-low are rare. Couplets
high-high, mid-high, low-high, and, in some dialects, mid-mid and low-mid,
too, are reflexes o f *modified-modified. The lowering o f a number o f
dialectal reflexes o f *modified-modified to low-high and mid-high in some
environments, probably following low t o n e , ' ^ seems to be an old morp h o t o n e m i c rule, maybe even o f Proto-Mixtec. Mid-mid and low-mid have
been lowered in a second step from mid-high and low-high in Molinos and
Xayacatlán.
Mainly a phenomena to be found in non-basic tonemic couplets, but
not restricted to them, is, that a m o r p h e m e may loose its perturbing
power if it has b e c o m e perturbed itself.
AREA B: The suggestions concerning area B are based on only a few
forms per dialect and therefore are somewhat speculative. Peñoles tone
modification at least seems to be a reflex o f Proto-Mixtec final glottal
stop, as Daly ( 1 9 7 3 a : 104) already suggested. There are some forms as
hints that Diuxi and Jicaltepec tone sandhi with lowering influence may
also be a reflex o f Proto-Mixtec final glottal stop and that the other sandhi
classes are to be explained as reflexes o f the Proto-Mixtec basic tonemic
couplets seemingly merged. The merging o f the reflexes o f the ProtoMixtec basic tonemic couplets and the lack o f a clear-cut division between
basic tonemic couplets and non-basic ones, in combination with insufficient data does not allow a correlation o f sandhi classes with reconstructed tonemic couplets. Finally, Coatzospan is the only dialect that
does not seem to fit in with these suggestions.
33
1.4 Problems Unsolved
My reconstruction o f tone and the reconstruction o f final glottal stop
done by Bradley and Josserand ( 1 9 8 2 ) does not fit well in Longacre's and
Rensch's reconstructions o f Proto-Mixtecan and Proto-Otomanguean. But
these reconstructions are not sufficiently founded with data on Mixtee,
and therefore there may be at least some truth in m y suggestions. It would
be o f great use to have a new look at Proto-Mixtecan and Proto-Otomanguean tones, laryngals and tone sandhi when better reconstructions o f the
pecuhar languages and language families b e c o m e available.'^ The f o l l o w ing problems remain unsolved:
Problem 1: There is no clear correspondence between Proto-Mixtec
final glottal stop and the laryngals reconstructed by Longacre ( 1 9 5 7 ) for
Proto-Mixtecan and Rensch ( 1 9 7 6 ) f o r Proto-Otomanguean.
The evolution o f Proto-Otomanguean laryngals into modern Mixtee as
described by Rensch ( 1 9 7 6 : 4 5 ) does not fit into the reconstruction o f
Proto-Mixtec final glottal stop. Rensch explained the morpheme-pattern
* C V V o f Proto-Mixtec as being derived from Proto-Mixtecan * C V h and
Proto-Mixtec *CV'' V f r o m Proto-Mixtecan *CV'? - therefore a ProtoMixtec * C V V ' ' or
V ? would have had t w o Proto-Mixtecan final
laryngals — but see Rensch's distinction between Proto-Otomanguean
* C V H and *HCV and his proposal o f * H C V H (H means laryngal h or
here). A quite reasonable suggestion f o r further investigation is given by
Josserand ( 1 9 8 3 : 4 6 0 ) : ' ^
" S o m e p r e - P r o t o - M i x t e c a n patterns are fairly d e a r , such as the d e v e l o p m e n t o f
o l d e r C V m o r p h e m e s t o a c o u p l e t shape, either b y lengthening the v o w e l t o a
structural C V V c a n o n , o r b y c o m b i n i n g such C V f o r m s w i t h o t h e r m o r p h e m e s ,
t o f o r m C V C V c a n o n s . It seems that m a n y such o l d C V r o o t s , either P r o t o Otomanguean or Proto-Mixtecan, developed into C V C V forms by Proto-Mixtec
times, and the o l d C V r o o t is usually the ultimate syllable o f the c o u p l e t . "
Problem 2: Although there is clear inner-Mixtec evidence f o r the distinction o f Proto-Mixtec *high-low and * l o w - l o w , the embedding o f these
t w o couplet f o r m s into Proto-Mixtecan is not quite clear: Proto-Mixtecan
* 3 3 b e c o m e s in most cases Proto-Mixtec * l o w - l o w , but Proto-Mixtec
*high-low may c o m e from Proto-Mixtecan *mid-3, *mid-low or even
Problem 3: The contrast between the t w o distinct dialect areas leads to
another problem o f this study. Tones have been inverted in the tonal
register in the t w o areas:
34
CHART 6: CORRESPONDENCES OF FROTO-MIXTEO TONEMIC COUPLETS
IN DIALECTS PENOLES AND MOLINOS
*high-high
*high-low
*low-high
*low-low
*modify
Peñoles (area B)
Molinos (area A)
low-low
low-high
mid-mid
mid-low
high-low
high-high
low-mid
low-low
modify: lowering effect
modify; high
This striking development leads to the problem how to explain such an
inversion o f tone by historical tonology. The suggestion o f a rule "feature
reversal" would violate the principle o f non-arbitrariness. But note that
Dwyer ( 1 9 8 1 ) reports a similar diachronic development of tonal inversion
[a high] > [-a high] for Loma, a Southwestern Mande language. ProtoBandi-Loma tones high, non-high and falling correspond to Loma tones
non-high, high and rising. A possible explanation Dwyer refers to would
be via the generalization o f a rule characterizing a particular linguistic
style (e. g. courtship, ritual or play language). For the Mixtee case another, more traditional explanation might be via allophonic glides o f the
Proto-Mixtec tonemic couplets - but such a sophisticated analysis should
be based on more data, especially for area B.
Another problem is that in my reconstruction the innovation is attributed to the dialects of area B, i. e. to those dialects being peripheral and
lacking contact. Coatzospan, the northeastern outpost o f Mixtee, lies
within Mazatec territory. Diuxi and Peñoles are located in the eastern
Alta, Jicaltepec in the western Costa. Therefore one would expect that
35
these distant dialects preserve Proto-Mixtec conditions quite well. Although I consider this as quite probable, I have not used it in the reconstruction o f Proto-Mixtec f o r the following reasons:
1) A n y argument based on geographical distribution is highly speculative f o r there are many more dialects to consider (mainly from the northeastern Alta and from the Costa) than those being the base o f this study.
Tonal isoglosses should be compared with the isoglosses and the dialect
areas given in Josserand ( 1 9 8 3 ) .
2) From the descriptive point o f view, t o o , dialect area B is not well
enough documented — four dialects vs. thirteen o f area A . Moreover the
number o f cognate forms is not sufficient to establish reliable correspondence. They cannot give but a vague idea.
3 ) A reanalysis o f Proto-Mixtec with inverted tones would have further
imphcations as to Proto-Mixtecan tones and even Proto-Otomanguean
tones. Such an enterprise would be far b e y o n d the scope o f this study,
although such work would be useful because Proto-Mixtecan tones are
based only on t w o Mixtee dialects ( o f area A ) and one dialect each o f
Cuicatec and Trique.
Because o f these problems, I preferred to reconstruct Proto-Mixtec
depending heavily on Mixtee area A evidence although I am quite aware
that the reconstruction may not be one o f Proto-Mixtec but o f Proto-Area
A-Mixtec. Therefore this brief study can't give but a glance on ProtoMixtec historical tonology. Its preliminary character should be overcome
by the integration o f data from more dialects.
2. Cognate Sets
Data have been taken exclusively from published sources, transcriptions
have been standardized according to phonetic symbols used in Mesoamerican linguistics.'^ In general, the transcription is phonemic as analysed in
the source, but in a few cases predictable phonetic details, e. g. nasalization o f vowels, have been indicated to facilitate comparison.
The arrangement o f cognate forms is the same as in K. Josserand ( 1 9 8 3 )
and is given in the following list, which also includes the abbreviations
used in the cognate hsts and the works consulted f o r each dialect:
NORTHEASTERN
ALTA
1) Diuxi (Diu): Daly ( 1 9 7 8 ) , Pike and Oram ( 1 9 7 6 )
2) Peñoles (Pe): Daly ( 1 9 7 3 , 1973a, 1977)
3) Coatzospan (Coa): Pike and Small ( 1 9 7 4 )
36
CENTRAL
ALTA
4) Molinos (Mol): Hunter and Pike ( 1 9 6 9 )
5) Ocotepec ( O c o ) : Mak ( 1 9 5 8 )
6) Atatlahuca (Ata): Alexander ( 1 9 8 0 ) , Mak ( 1 9 5 3 )
l)San Miguel el Grande (SM): Dyk and Stoudt ( 1 9 6 5 ) , Pike ( 1 9 4 7 ,
1948)
NORTHERN
BAJA
8) Cacaloztepec (Cac): Pike and Cowan ( 1 9 6 7 )
9) Xayacatlán ( X a y ) : Pike and Wistrand ( 1 9 7 4 )
SOUTHERN
10)
\\)
12)
13)
14)
15)
BAJA
Ayutla ( A y ) : Pankratz and Pike ( 1 9 6 7 )
Alacatlazala (Ala): Zylstra ( 1 9 8 0 )
Metlatonoc (Met): Overholt ( 1 9 6 1 )
Cahuatache (Cah): Schultze ( 1 9 3 8 )
Mixtepec (Mix): Pike and Ibach ( 1 9 7 8 )
Silacayoapan (pueblo San Jerónimo Progreso) (Sil): North and Shields
(1977)
COSTA
16) Jicaltepec (Jic): Bradley ( 1 9 7 0 )
17) Jamiltepec (pueblo Chayuco) (Jam): Pensinger ( 1 9 7 4 )
Additionally, f o r a number o f dialects kinship terms have been taken
f r o m Merrifield ( 1 9 8 1 ) , and, in a few sets, gaps have been filled with those
f o r m s f r o m Josserand ( 1 9 8 3 ) marked f o r tone.
The following list contains 110 cognate sets.''^ O f these, 51 sets have
been reconstructed by myself according to the system used b y Josserand
( 1 9 8 3 ) , the other reconstructed forms are taken from Josserand ( 1 9 8 3 )
and only tone has been added.
37
•'S
>>
#
3
0
e
1 13
1 M
1 13
>ts]
1 '73
1 ü
1
>M
13
A)
13
>
N
13 13
»
13 13
>N >N
1
1
1
1
1
1
1
1
1
11
11
11
1 ':3
1
1 i:3
1 ^3
1
1 V3
>N
1 13
1 M
1 13
>»
00
c^
ft
gj
H)
L,
•u
•'S
>>
Ä ft
t^
c*td ft
•a
"•td c
-d
tí «
ft
I
1
1
•tí
<n
c
m
H)
•Q
ft
'Cd"
B
C
'd
n)
'0)
'3
tí
•tí
tí
1 'H
1 >C0
1 C
'd
A!
"tí 'rti
-p P
A! A!
tí
^cd lid
led ltd
•a •tí
tí
tí
-
10) 10) IdJ
-p -p p
13 13 13
•tí
•tí
•tí
G tí tí
tí
,—,
E
—
>10
ltd
»cd ltd
Al AI
»H
E
—
13 13
B tí
13 13 IH
>N >N >tS3
S
E
—
—
itS las ltd icd
E B E
a c^
C' c^
icS ltd ltd ltd
m Ul m CO
,—,
E
TH —
'H -e- IH
>o
1 >o
1 '3 '3 13
1 •tí •tí •tí
ft
tj
33
<0
O
M »H (11
cc X
1 »cd 0
te Ai ft
u
cc
«
tN
ft
m
«3
W
>J
»r« e
p-, -P •H
»i« C
o A! ftId
o
ü
M
E
W
fe
O E
E-i PL,
38
1 13" 13"
t
1 1 3 13"
>tS] IC
E
—
'td 'td icd
1 c^
c^
1 C
'd C
' d ltd
1 T) •O T3
c tí tí
tí
ft
tí
1
1
1
'tu
'3
c
•tí
1
1 -P
1 >N
ft
«
<11
•c
S
c^
O
m u
ft
»3
,—
•—
tí
tí
E
—1 IH 1 H
1 X
1 lid ltd
rH O
O o
E o
td
A¡
1
1
1
>cd 'td 'td
1 E B E
1 c^
1 red
'td
to
to
1
1
1
,—,
E
—
'H
>o
13
tí
•tí
,—,
E
—
'td
lid
Al
•H
3
Ai
1
1
1 -H 1
1 -P 1
•H
3
A!
1
1 Id
1
1
1
1
»H '3
1 >o >ü
1 '3 13
1 •tí •tí
C tí
'13
1 >o
1 '13
1 •tí
1 ':3
>o
1
1 •tí
1 'H
1 >10
1 'Id
A!
E —
E —^
E
—
1« 'H 'H
p P 10
1,« IH IH
Al Al Al
1
1
1
o
td
o
>>
E
m
1
1
1
E
—
'td •t'd icd icd ltd »cd C
' d 'Cd
1 C'
1 lid C
'Cd
'd C
' d 'td ltd loi
1 Tí •tí •tí •tí •tí •tí •tí •tí
tí tí tí tí tí C tí c
tí
^—.
A;
E
—
»td •'Cd
E c^
E E 1
c^
c^ 1
ltd ltd C
'd 1
tí Tt m
'3"
'3
>i
E
—
tí
lO
1
1 'O
c
-p 1
1
1 13 1 H 1
>ä
—
lid
>o 1
13 1
•tí 1
,—,
E
—'
IH IH
1 >ü >0
1 13 13
1 •tí •tí
E
—
1 IH IH
1 p p p
1 IH IH 141
Al Al Al
3
td
•H <u o
O n. o
E
—
ltd
c^
ltd
T3
C
1
1
1
>>
td
;><
—
tí
1
1
1
tí
'td
'3
-tí
C
IH
>í
1 p
13
1 •tí
tí
1 IH IH
1 >10 >10
1 lid lid
Al Ai
1 »H
'H
1 >to >o
1 ltd »Id
Al Al
1
1
1
1 •H
' IH
1 P -p
1 IH IH
AI
A!
1 'H 'H
p
1
1 »H 'H
Ai Ai
1
P
1 rti
•p
0)
E
td
o
X H ü
•H •H •H
E
1-3
E
Id
td
iH
<<
JS
A!
VH
-P
•SH
«
»o
»o
>>
, ^
<0
o
ft
^
1
1
1
>o
1
1
1
»o
IH
>o
IH
IH
>ü
IH
IH
>o
IH
1H
»O
IH
»O
lO
lO
>m
lO
>N
1
1
1
lO
c^
lO
>N
ft
QJ
Q.
0
ft
1
1
1
«o
lO
>N
—'
1
1
1
»o
lO
>N
ft
>>
C
nj
0
0
ä
ft
1
1
1
1
1
1
1
1
1
«
m
•»et)
>1
•H
-c
o
la!>" I0>"
c^
C'
IOÍ" letf"
IC
IC
I<d
c^
10)
1
1
1
10)
c^
1(0
>tQ
1
1
C'
10)"
IC
<•0
C'
lO
1
1
1
1
1
1
»O
1
1
1
1
1
1
1
1
1
'Cd
»ttf»
10)
IC
10)"
IC
IC
1
1
1
1
•
1
1
1
1
Kd
»Ol
'0!
10)
>»
Id
•»0)
1
1
1
*H
':3
lO
>>
>>
1
1
1
»o
lO
>>
'H
>ü
»H
'H
•p
'H
'O
C"•O <0
>>
IH
>>
IH
lO
c^
lO
>>
E
—
1
1
1
»0)"
10)
c^
10)
>S)
1
1
1
1
1
1
S
liH
1 ^ ^
1 141 liM
>S] >N
1
1
1
10)"
IC
c-
«an
«0"
IC
1
1
1
£
E
^a)
»cö
'Ol
>N
'iS
>m
yiH
^
rM
1
1
1
•—•
ft
ft
1
1
1
^^
ft
OJ
c
0
—-
1
1
1
1
1
1
NH
>o
IH
IH
>o
IH
1
1
1
1
1
1
"tl
>tS) >N
1
1
Irll
—
•—
1 C^
1 »H
1 Jll
'H
^
IH
>>
1
1
1
1
1
1
1
1
'
13
c^
13
'H
Jt
'3
13
13
IH
J.!
IH
>>
T—
r—
tt
ac
u
<u
M
3
X
1 •>•3 0
X «>> fts
o
E
1
^H
3
m
*
ftOl
1 »3
1
1 '3
'3
»3
1
1 '3
1 '3
1
1
1
>ra
E
E s
13 •>3 13 13 '3
13 13 13 13 13
>N >N >N >N >N
.—.
>N
0) iH
3
•H O O O
Q PL, O E
13
>N
13
>N
O 0)
O -p
o «í
>>
'3
13
i;3
•>3
13
'3
i:3
13
#H
ja
13
S
IH
-p
E
JC X rH O a
1Ö •H •H •H «
E CO 1-3 •-5
o
>>
>>
>>
>N
.
E
E
13
>N
13
>tQ
—1
1 13 13 13 '3
1
'3
o
E 0)
CO o
rH
13
>»
>>
0)
|X|
'H
s
'H
>>
<
>>
0)
rH
<<
>>
>H
fs
1
1
1
>>
39
3
»cö
Ja
0
u
K
/H •íH"
1 a B 1
1
c^ 1
1 »ta ycö 1
4«!
13 13 ICD"
1 a a a
1 1
1 t^
C'
1 1
1 ICÜ 10) • cd 1 1
1
1
1
IH
1 a
1 c^
1 ICÜ
o
-Y
•Q
0
IJ
«
K
CV
R\I
K
QO
1
1
1
'3
A)
'CD
1
1
1
1
1
1
13
AI
13
A:
AI
AI
let) icö
13
13
AI
led
>1
AI
C
Ol
0
Ä
K
3
11
•Q
»
-•ca
yoj •p IClj ICÖ icö
-P
-p p -p
1 ycü C
' d icd ICÖ led
1
1
AI
'3
»3
AI
yH
AI
icö leö
P
icd loj
AI
A!
13
13
M AI
1
1 13 13
1 13 13
AI AI
13
C
13
13 13
C C
13 13
Ai
1
1
1
icö
-P
lo!
loJ
13
13
AI
1
1
1
1
1
1
IH
^
13
AI
1
1
1
1
1 13
1 13
AI
O
OJ
G
'3
Ö
'3
•"rl
X
E
O
O
1 /H 1
1 >01 1
1 'H 1
AI
•>3
C
1 IH
1 Kl
1 »H
AI
1
1
1
1 «H
1 >o
1 »H
AJ
«ci.
»o
•'o
pH
«
W
Z
O E
OH
40
IH
IH
X IH >W
IH 1 H IH
AI AI AI
>m
13
C
13
c
to
CO
t-1
Cd
J
o
o
^3
C
•o
c
<0
•c
lO
lO
lO lO
lO lO
-p
13
iß
IH
-p
iC
IH
B
'3
'3" )C
«t-f" yH
>0 -B-
13"
id
IH"
-p
cö
0) o
PL, O
rH o
O o
s
o
cö
p
<
lO »o
lO ^o
-p
-p
lO"
lO" 13"
IC
lO" 13"
IH
>o >o >o >ü
E
lO
ic 13" 13"
IH lr-K> 13"
>o >0 >ü
ü >s
CÖ P
CÖ <d >1 iH 01 cd
o X <c «C E o
X rH o a
•H •H •H td
S m
>-3
t
<11
c
0
K
o
I
ft
Ir-I"
IH"
1
IW i,«" IH" IH" ^rV
IH" IH" ^H"
1
Irí"
I r^" I H" yr-K> I
Ir+n
I rí" I r^n
Ir«"
•o
1-,
d)
»3
#
ON
C\i
>í
C3
Ä
to
<u
'3
'H
ft
1 13 13 13 13
IH IH IH IH
I
,
l A'3!
,
ft
m
«
^01
»0)
Jh J»
O
ft
^
CO
CV
3
Ä
1
IH
o
c::
ft
1
1
1
-p
Jf
I
1
1
Id) 10) 10) 10) 101 10) »0) 101 10) 10) 10) 10)
cc^ cc
i<n Id) 10) 101 101 10)
10) 10 10) 10) 10
Jt
Jt Jt J» 3
'I
3 Jt
10)"
-P
IH
Jü
IC#" 1 lc3 ICÖ IH IH ^r-f" Irl" 11 tí
-P 11 -P
1 B Ö >ü >o
IH I IH 1 H 1 H IH 'H IH 1 IH
4)
4)
Jt Jt 3
3
l<D
41
IH
1 IH ^H" >>-p
-P
1
1 1H
J»
rcv
Ved" Ja
^cö
S
o
AiiJ
^
CV
te1
ÖC
o
I Ioí" la)"
I ictfn la)"
I 3 S
I
1 ^O
lO
1
1 ^O 'O lO
1 lO
1 c^
1 lO
A!
lO
c^
lO
A;
1
1
1
1 lO
1 -p
1 lO
lO
-p
lO
1
1
1
1
i
1
i
1
i
1
i
1
i
1
i
1
3
e
-ü
0
k
»o
^0
Ai
I
^ri" iní"
1 ^0)" ^H" ittf"
o
o
'O
»0
'O
lO
-P
lO
cv
ífl
0 o
OH o
o
o
o
o
td
E
<33 w
o
<d
o
1
1
1
i
1
1
1 fO
1
1 ^o
>s
ta
X
<<
I
1
1
I lO
1 c^
1 lO
1 lO
1 -p
1 lO
Ai
oj
iH
o;
-p
0
E
n)
o
X iH O
•H •H •H ta
E CO >-3 1-5
41
-Í
»cd Cd
C tä
I I I
I I I
I I I
1
1
1
1
1
1
'd
1 C
'd
led C
1 E B E B
1 led led" led icd
ö c C ö
1 C
'd
1 B
1 led
C
1
1
1
1
1
1
1 led C
'd
1 B E
1 led led
C C
CO
C
'd
•»cd
(H
'td
»cd
H
•O
u,
•H
Ja
t
Od
o
CH
ft
»cd
C
'd
1
1 H
1 »cd
C
'd
1
^td
1
1 C
'd C
'd
H -ö-
1
1 C
'd
1 led
to
1
1
1
1
C
'd
led
rH
led C
'd 1
led led 'H
ra CO •P
'Ol
1
1 led
ra
1
1
»oi
'Cd
m
1
1
1
1
laJ ltd
• cd led
H m
ltd
C
Q CO
1
1
1
1
'Id iid
C'
1
1 fed »cd led
ra rH CO
1
1
vO
ft
'H u
CO c:
»H 3
CO Cd
#
ft
ft
í)
'O
-p CJ
»H c
CO 3
Ä ft
Tí »rH
'H Tí
Tt 'H
Tt
—
»H 'H 'H IH
'H
tu ra >ra ra 'H TU
»H IH IH IH IH 1 H
tl ra >m ra >co TU
1 H 'H
TU >m
IH 'H
TD >co
—
'O
-p »O »o 'O 'O lO 'O 'O lO 'O
•p lO
'H
lO lO
TD 'H »H -p IH
-p IH IH 'H
TD Tt ra >m ra ra TU TS >co
1 IH IH
1 >C0 ra
1 IH IH
>C0 ra
1 'H
1
1 'H
IH 1
1
1H 1
>o
>ra
1 'O lO 'O 'O lO
-p •p -p
1 -p
1 IH IH IH 'H IH
>co
>m ra >w
1
1
1
•sf
c^
C
'd
»cd
»
e
I)
S
•M
ft
'd
1 C
1
'd
1 C
1
1
1
1
1
t
1 led C
'd
1 ^
1 led led
A!
1
1
1
1
1
1
1
1
1
1
1
1
1 'd
1 C
1 C
'd
'
1 icd led C
'd
1
ü AS
1 led led led
>!
1 led
1 ^
1 led
1
1
I
c^
ft
•-H
o '3
0
iJ >s 0
1
1
4J
1
-p Ol
1
ÍC Ä
ft
o
w ^
#
cv
m
ft
w
^
dl »3
'3
a. -P
p^
ö »3 C
O —
-p Q. »3
o
-p
ü Ä ft
o ^
M
CS
E C^
CN
W
3
•H
o £
PL,
O
42
»3
C
'D
p
»3
»3 -P '3
-P
-P
'3 »3 13
-P -P -p
C
M
Cd iH
CD o O
CL, o z
1 '3
1 >
N 'H
1 ICD 1(1)
-p -P
1
1
1
1 'H '3 IH
1
led >>
1 led
led
1 13 '3 '3 13 '3
1 -p -p -p
1 13 13 13 13 '3
-p
-P
o
u
o
cd
-p
E
<c m
ü
cd
o
cd
<
1 13
1
-p
1 13
•P
cd
iH
<
-P
O)
E
1
1
1
1 13
1
'H
1 led 10)
-P
1 '3
1
1 13
+=
1:
3 13 '3
-p
1:3 13 13
-P -P
X
-H
E
rH ü
•H •H
m 1-s
^
cd
B
cd
—
t»
0
•Q
1
1 ^td
1 C
e
dl
•o
•H
Ol
C c
•H
#
c
1
1 »H
1 tí
c
»H
«
tt
tt
IH
(1)
CO 5)
•o
k
K
IH
Ol
ä
O
•P 'S
k
-—
fe
TJ 'H 01
(D
3 ^ed 0
>í •c
tí
fe
•H #
-p
c 1—
o
fe
Ü
—
IH
91
«
U
o 'H
•J tí 0
1
1H
a; #
fe
u
—
)1
33
»
O
CO
E-i
fe
M
1-5
c
HH '3" ä
O »3 0
u
O >>
fe
o
o
M O
E
U
a
o
tH Ph
1
1
1 T«
ß
-p
1
1
'IS
ictfn tí
»Id
a IH
C
1
'd
»(Ö IH led C
tí 1 a a 1 'H
IH -p IH IH -P CO
íH IH /H
tí C C C
IH f H IH IH
'•a
'3
Tí CO to
^H IH IH
1
1
1
ra
IH
-—.
1
1 »3
1 '3
te
3
•H
Q
0)
OH
ft
^
1
»Id
1 ^t!) 'd! '(d /cd led C
'd
1 -P H •P -p •P
IH IH IH IH
'H
tí »H
C
' d tí
' d tí
^c C
ftí »(B
1
1
/•H
1 C
'H
1
1
1
13
,—.
1
1 —- '3
i icd led
>tS]
1
1 13
1 •a
IH
1
1
1
1
1 IH
1 c
IH
1
1
1
1
1
1
i:3
CQ
1 i:3
1 1
1 C
V
I
1 i;3
1
i:3
1
1
1
1
1
1
1 13
1 CO
1 13
>>
1
1 C
'Ö
1
'H
1 IH
1 411
1 led
>>
>.
Ití
1
' H> 1 '3
'3 '3 13 '3 C
'3 13 13 13 ICVi 1 '3
IC ítí )C ftí ítí
Ití
cd
o
o
o
H O cd
S
O O
cd
E O •a; CO o
>.
ed
!•<
1
led
1 led
C
'd
1
IH
IH >> IH
1
1 j»
1 icd
1 'H IH 'H
IH 'H
1 tí tí tí ití tí tí
1 led
icd led led led C
'd
ítí ftí )C Ití IG
<tí
>>
<<
led
C led led
1 IH
tí 'Hc led 11
•
1
1 Ití -•v
1 C 1
-p
1H
1
1 IH IH
1 C C
IH IH
Q)
W
IH
0
m
IH
0
-q
ft
>H
1
1 *3 Jt
1 C
'>td
q *>ecdo
1
1
1
1
1 icon
1 C
IH
1 IH 'H 'H
1 s Jf
1 led fed
>N
>>
1 IH IH
1 tí tí
1 led led
>» Ití
lO"
lO"
Ití
ed
iH
<
1
1 led C
'd
-P
1
IH IH
1 'H
1 tí
1 C
'd
Ití
1
1 lO" '3
1 »O" 13
IC )C
•P
O)
E
cd
o
1 'l)
1
1 led
1
1
1
1
1
i
13 13« '3
13 130 13
Ití Ití Ití
X iH 0
•H •H •H
E CO
B
cd
43
T3
c
«m
o
•o
<0
1
1
1
lO
1 c^
1
I -a
c
lO 'O lO
1 c^
1 'O rO »o
•ö •ö •a
c c
c
ICS Id
l
1
1 > s
in) ItÖ
C
'd
s
1 led
1
1
c^
"•O ^O
•d •a
c c
X
0) «
Ol t»
1 c
0) -H
to
O -Q —
1- -H
U tfi
t
K
^C
Kl
>>
<0
'O dl
e
Ä
1
1
1
»C iC
C
'd
1
1
1
1
o
i)
fte
1
1
1
1 13
1 )C
1 '3
13 13 lO"
IC iC IC
'3
>1
1
1
1
lO lO lO lO lO
c1 c^
1 lO »o »o VD 'O
1 Tl T3 •o •o T3
c c c c C
Id
Jt
c13
1
1
1
1 lO"
1 IC IC
1 '3 »3
13"
IC
1
1
1
»cd ICS led ICfl led icd icd C
'd
C
'd C
'd C
'd C
'd C
'd C
'd C
'd C
'd
1 c^
1 C
'd
1 >s
1
1
1
• cd
A
'3
1
1
1
1
i
1
1
1
1
1 C
' d i(d C
'd C
' d led
1 C
'd C
'd C
'd C
' d »cd
AJ M
1
1
1
1
1
1
1
1
ft
o
ca.
CO
'ftt
«
O
'O
-P
Ä
C
O
u
e
ft
•O
0
ä
ft
—
1 'rH
IH
1 -O) TU m
1 •
' r« 1' +1 'H
A:
1 liff 1« IH
Q TU
1 CO C
1 'rH 'r« 'H
1 'H IrH IH
1 CO C
Q CO
1 'H 'H 'H
I '3"
1 c^
1 »3 »3"
-P -p
13"
1
1 13"
1 '3"
a -p
1 '3"
1
1 '3"
1
1
1
1
1
1
1 IH
Q
1 C
1
'H
1 NH
1 CO
1 •»H
4«!
lO"
1 13" c^ 13" 13"
1
lO"
1 'Pn c '3" '3"
P -p -p
1
1
1
13"
'3"
00
Ä
ft
ü
M
>>
c
' D <0
ÍB C
1 p
e
S
i
t
ft
g
ij
»
-sf ft
CO
•c
E-1
u
<0
W
hJ
B
Oh X 0
'H 4J
&
o -p C
O
«
CJ)
o
M vO
s
w
fr,
O >:
fH p-,
44
1
1 '(1)
1
P
1
1 Id) IIB ItU led
1 'tu 'tD '0) C
'd
p >CJ
1 led C
'd
1 C
' d led
1 >5 >>
p •p
1 led
1 C
'd
1 >>
-p
1
1
1
1
1
1
1 IH IH IH 1 H »H
1 >C0 >tn -e- >CQ >o
•
IH IH 'H 'H »H
-P -p
P
1
1
1
1
1 led C
'd
1 C
'd C
'd
>ü •p
.—.
1 -P
1
1 >C0
—'
1
IH IH
IH
1 IH
IH >tQ
1 'H 'H 'H 'H 'H
>ü >o >o >0 p
cd
3
•H <D o
Q O, o
tH O
O ü
E O
O
Cd
-p S
Cd
-a; CO o
1
1
1
>i
cd
X
1
1
1
>>
<i
cd
rH
<
0)
x;
cd
o
X iH ü B
•H •H •H cd
CO 1-3 1-3
^oi
'ed
3
K
•O
0
o
00
c
»ed
»ed >ed led led led led led laj
—
cC'
pc^
c^
'ed ^ed
'ed 'ed ^ed ^gj *cd *cd ^ed
J» J*
J» Jti Jt
M JS
led led led
c^
c
c^
^ed ^qi led
Jt
M
>
led led ^ed led
-5i
o
•O
X •-H
0
u
S
c
»
1 ^ri"
1 >m
1 »H
Jt
>0) Ir-I"
C' X
»H
JI
1 Ir+n Ir^n IH IH"
1 H 1 r-t" IH" IH
1 X X >to >ra >eo >m
>M >m -e1
yH ^H
S
.01 3 J» Jt 3
IH
>m
IH
.ai
1
1
1
O
m
-p
»
k
L,
<1)
•H
L,
1
1
1
-C
0
0
K
1
13"
1 c^ c^ ,30
1
C C
fe
1 l(B Id) lel)
1 -P -P -p
1
'3
>tsi >N >N
>N
1
1
1
1
1
1
1 led
led led »ed
1 >i led >0
1
':3 »3 '•.73"•3
>>
1 13" 13" lO"
C'
1
1 '3" /3" »eD"
C C
1
1
1
1
1
1
1
1
1
1 lO"
1
1
c
1
1
1
1
1
1
1 10 10 10
CL CL
1
1
'O
»0
CO 0) TÍ
1
10
1
1 10 'O
in
to
1 10
1 c^
1
03
1
1
1
1
1 »0
1 »0
to
led
>u
»3
>N
1
1
1
>>
eX)
»0
c
«
-
1
1
1
in
T3
0)
3
C
•H
c
0
Ü
»0 K
IH
'O 51
iH (1)
XÍ ft
—
, ,
>•0
»0
<•0 "•o
•a HI
1
1
(
01
00
c
•H «
K
0 »3 "H H)»3
M c « IK - c '3
^eD •f^ -H
C »3
®
1 X
.»ed »ed C
«
K
>m m
0
hJ
«
in
to
fe
ir>
W ^H •0 *ed
>N
tJ >>
•H
PL,
<•<1) »ed
>w
d
Ö <em
0
U ^ed
0 #
•W Tí ^H
0
M •vt
m
s
w
3
B 2:
ed
0
•H B 0
E-i PL,
Q OL. 0
—-
1
1
1
1
1
1
m
1 13 13 13
1 C s
C
1 'ed
'ed
X X X
1
»3
1 13
c
a
1 ^efl ^ed
0) >0)
I<1> Id) 10) Id) 10) 10) ^0)
c^
.^cl)
»0) ^0) IH
m to 03 to Tí Tí to
rH 0
0 0
E 0
ed
H
t)
E ed
CO 0
>>
ed
X
>>
<i
1
iO
1 C c
1 led led
to >01
1
1 C •C3
1 'td »ed
03 >v
1
1
1
IH
led
\<0
1 10) >i 10)
10)
C'
1 c^
t^
1 10) led
'ed <•0) '0)
m 03 03 J3 03 eD
ed
rH
<
0)
£
ed
0
X tH Ü S
•H •H •H ed
S CO
1-5
45
K
to
'td
M
•H
1-.
0
ft fc
,—.
,—.
S
E
-—
1
.CO
>ca ^co 'td 'Oi •
•p Itli ICO
1
1 »H c
-p
TU
m
to >to
—'
.—,
E
-CO
-p
tH
Tt
>—
,—,
E
—
C
'd y
.cd
-p led •p
y^H
.H
TU >03 Ui
C
. O icO •»cd .cd
-p
.cd
IH IH
'H
>Ü3 to to
^cO .cd
•^H .H
>C0 >05
^
00
vü
»o
'O
•o
c
«
sO
K
D
C
10
KU
»o
yo
•O
' >tq
1
-P
•tí
c
—
1
1
1
1
1
1
—
1 ^o yO 'O
1 lO lO .o
t T3 T)
c C c
e
x
•o
C
T3 to
Ö« eto
1
1
1
1
1
1
1
1
1
1
1 .O .o .» 00
1 .O
•o •ö
c
C C
.cd"
>ra
IH
•n
c
1
1
1
1
i
1
lO
.o
»0
•a
c
1
1
1
^
1
1
XUwi
>10
1
1 •>rH 1
1
1
ß
^tlVi ^cy ^co
1 X X tQ m
1 1H 1H
1 •o 'O •o • a
2 c
c
c
.cd >cd . cd
to -o- to
N •tí
•a
a a C
1
1
1
1
1
1
»0
.0
>H .H •»H
>0
>0 -P
.cd .cd •>cfl
>1
1
1
1
vO
^
»o
'O
AI
#
b
II)
to
c
m
K
1
1
1
s
•^o
>o yo
tO
^
»o
'O
c.o
iO lO
ü >!
.o
.o
1 .H »H
1 >o >o
1 inj C
'O
1 .H
1 >o
1 icd
1
1
1
—
1
1 yO -o
1 lO lO
i
1
1
>•0
. 0
. 0
. 0
. 0
m
c^
vH
ft
e
C
O
u
u
0
tj
e
1
1
1
>H
1 -e>o
1
1
^cd
1
1
1
1 .H
1 >o
1 .cd
vD
.r«
3
ft
>1
<H
to
•o
K
^H
»r«
I
I
I
rM
^
iM
1H
.H .H .H
Ji J» S
yH /H .H
^
Irti
AI
AI
AI
1 .H
1
Ji
1
'H
A!
1
1
1
ym
^
y<ti
M
vO
a
0
9
.rH
j
1 AI
ft
:s
Q
ft
^
rvj
ft
t^
3
(3
tu
ft
ft
O)
c:
0
hJ .cd« •H
DL. •p •a
.cd tu
•p 5
0
ft K
ü
CO
EH
M
0
•H
M
s
^—
VO
0
E
E
»H >Hn —
B B
AI
A!
46
CU
AI
AI
E
. 3 "
13"
AI
AI
A!
A!
B
. 3
AI
l IH
1 B
1 .3
AI
IH >H .H
B
B
B
.0
13
. 3
A:
AÍ
AÍ
NH .H
B
B
»3
.3
AI
AI
E
^cd"
C
-P
—
»cd
P
1
1
1
w
EH
AI
.H
a
.3
E —
E
— .r-f" .H
. 3 "
E
B
13" .0 . 3
E
3
•H
Q
O)
1
1 .cd«
p p
1 .cd .cti"
-p P
Cd
0
u
H
0
E
0
0
0
E
—
.cd .td .cd
fe; C p
icd ltd .td
-p P •p
1
1
1
0
td
0
cd
cd
p
<
E
K)
>>
1
1
1
>>
<
ictf"
—
1 »cd"
1 p
1 .cd
p
cd
H
<
0
E
1
C0
1 >P
1 C
.O
P
Ä
cd
0
X
C
.O
•P
.td
-p
H
•H •H
E m
1 .td
1 -p
1 .cd
-p
ü s
•H td
>-5 1-Í
»o
»o
>1
tt
fe
,—,
c
o
o
Ke
1
1
1
—
>o
»o »o
>K1 >M >N
1
1
•0
1 lO <
10
>IS1 >
[S3
1
1
1
1
1
1
1
1
1 »0
>>
1
1 »0
1 »0
>>
'O VD »0
yO
»0 <•0
>a >N >v >a
m
«
K
0)
c
0
u
ftOl
-—.
-
>3
»3 »3
>N >m
S
1
1
1
'3
13 13
>N >N '3
>N
—•
<3
13
>>
>>
c-
E
1
1 C
1 •>3
C
^3
C
*
1 »3 >3" -3 '3 »3 »3
1 »3
<•3 '3 13 13
C
E C C C c
D
ti-t
K
1
1 tí Q C c
1 '3 »3 13 13
C C a C
1
1
1
1
1 C
1 13
C
1 »3
1 »3
C
1
1
IH
»H
1
1 c
tí
1 '3 <0
G c
1
1
C\J
s
o
ij
1
1
1
»O"
»3 sjn »3
»O" »3 »3 vgn »3
c
tí c C tí
S
Q.
m
0
Ol
ft
-rt
X
1 »oj
1 a
1 »cd
c
E
—
1 »ta
1 B
1
tí
E
—
1 »cd »cd -•td
1 a a E
1 Its ltd C
'd
tí c c
1 -•td
1 B
1 ltd
C
,—,
E
— 1
»0
>N »0 »0 1
IH 1 H »0 1
X X >m
»o
>01
ltd
—
1 »td
1 a
1 »td
c
1 »0
1 >->
1 »H
>m
1 >td »td«
1 B B
1 »td »td"
c
tí
1 »0
1 >>
1 »H
>m
1 »0
1 »0
>K1
1 »td
1 B
1 »td
tí
1 »0
1 >>
1 *H
öl
O
m
H
ft
w
ij »H •0
CL,
ID
c
o »H Ol
Ol •C
o
S
ft
o
S
H
e;
o
1
1
1 ^rfn ^H
1 C C
1
C tí
—
ft
o
1
1
'3
1 »3 <•3
>1 >>
—
e^ e
íH C
C IH
'O 0
c ku
#
3 »0)
Ö c
•H
-P
Ö
O
ü
13
<•3 ':3
>i
—.
1 A!
1
1 TU
E
1 ^H »H »H
1 c C c
1 »H »H IH
>K1 >01 >01
3
td iH
•H Ol 0 0
Q CL. 0 E
0
0
0
td
-p
<
—
, ^
E
»H
C
IH
>01
E
»H
C
»H
TU
E
m
V
td
u
—
IH
C
/H
>m
-H
C
IH
>W
>>
td
X
>>
<
td
»H
<5
-p J3
(B td
E 0
»H "•H
C c
»H •>H
>01
»H
tí
»H
>01
X iH 0 B
•H •H •H td
E m 1-3 1-3
47
fc
i
<u
^r-t"
X
«cri m
m c
»
00
-3
10
fc
K
4)
U
0)
•H
C
fe
>01
>m
•a Tí
-
Tí
NH
Tí
1
1
1
1
1 X X X
1 ^a! ltd Itö
m m ra
1 >rO >ra
1 ^ts nd
TU m
1 >m >m
1 led icd
ta >tti
1
1
1 '3 '3 '3
1
IH IH
>m
>CQ
1 13 '3
A!
1
1 1H IH
•m m
1
1
1
13
ü
IH
»ra
>m
<cö
1
1
1
1 ':3
1 A!
1
^
(M
00
'td
3
»
fe
I-.
í!
U
fe
C
/Ö
1 •c'ö >0
1
NH
1 3 3 3
A! A!
led
C
'd
'td 'id
IH ^H
3 3 3
Ai
'td
IrH
C
'd
1 >.
1 'H
1 3
H)
a
-•Oj" CO
cv
^ca
^
o
4j
I ^oj«
I
I
'O)
•O 3
(J
C
C
'd 0
Ä fe
TS
I c
1
1 '0)
A!
yoj« --ajn --a}" --aj«
^cíí"
I ^cíí" yoj«
yoj^ injo
injo inyi
inyi ^aj"
^aj"
l
ll
/oj«
loí"
laí"
i
ii
lo)« loj«
inj«
inj« lo)«
lo)«
l
ii
i ctj"
yaj«
^ay
o
00
T3
C
»cd
A!
1
1
1
C
'd
'ÍD 'id '0) 'ÍB >>
T3
•0 •a -ö
I C
c
1 c^ C! c c
1 led led 'ÍD C
' d icd
Al A; -p •p A!
laJ
•O
1 c
1 c^
1 icd
A!
C
'd
3
c^
13
Al
IH
>>
c
C
' Ö 0) ^CIÍ
T3 JC TJ
c oc
C^ L, C^
'Oj ^^CS
US
•O
Cc^
la)
A!
CT^
t3
'3
A!
#
IS ^Oj
o
J«!
I
0)
C
O
fe
^
B
O
U
O
J
«
ccco
EH
K
w
o
•J í«" OO
CLh C^
ED^itíO
o
pit; 4-)
o » «
ü
"
M vO
S
W
B
O E
E- Oh
48
^3
—
»3
A!
I
I
I
c
A!
^cö
1 '3 '3 '3 '3 '3 '3
1 c^
1 '3 '3 13 13 '3 '3
A! Ai Al A! Al Ai
I
1
I
I
I
I
0\
I
1
1
^cU
13
Ü
13
^ttf
>!
O
I
I
I
I
I
1
,H
I rí))
I ^H
I
1 I
C
/Ö
4<!
3
I
I
I
^Oj
I
1
1
3
^
1
I
I
o
I ^rtV ><««
íT+i
I
I C C ^ C C ' C ^ C ^
I
I
^fit«
^H« lttK> lf«f« I
3
Cd rH O
•H
O O o
O CL, o E o
cd
<
E
W
O
cd
o
I
^on |
I C ^ C '
1
I l;^ rty^ I
>>
cd
>>
<
cd
iH
,
I
I
-p
O) cd
E o
|
IC'
I
M
X rH O S
-H •H •H Cd
E CO
1-3
yo
>e!
dl
>
•ri
K
«k^n
^^n
I
^^
I
I
I
I
<3
I
I
ÍTH
^Hn
I
^rtjn
I
I^
>C0
1
1
1
i-H
>0]
1H
C-C-C^
s ^ ^310
IC^C^
I
I
I
I
I
I
I
I
I
I
I
lO"
I C ^ C - C ^ C ^
10"
0
O&
0
ä
u
s
<•3
>H
41
13
IH
3
10
I
I
«"tI "^H
r®
00
ft
00
00
(0
ft
k
1>
c
•H
-
I
*
«»H"
«»H"
I
I
I
»H
TB
^H
JC
1
1
1
1
1
1
^H
>m
^H
A
«r-i«
-»r-P
NH
•tu
K
(l>
m
ä
m
K
>r-K>
TU
NH"
Jt
Jt
«•H
>CQ
IH
TU
M
IH
ÍQ
1 H
3
1
1
1
»H
Ü
^
00
»ffl
3
c^
ft
>>
1!
•0
<0
a)
•c
k
^Oj
Jl
»3
>ts) >N >m
1
1
1
1 'ta 'ta 'ta 'ta
1
1 13 13 <•3 '3
>N >N >>
'ta
1 ^
1 c
1 'H
>>
1
1
1
1
1
1
1
1
1
ita
I
1 C'
1 13
>»
1
1
1
^
vO
00
AÍ
»3
ft> >
•K
Ja «
L, w 1
0 3 1 AS
« -Q 1 "•3
>N
K
ft
1 »3 '3
1
1 »3 »3
>N >N
'3
iS
13
>N
'3
A!
13
>ts]
1
1
1
1 '3 10 1 1 ':3
1 A! 0) 1 1 J>!
1 13 "S 1 > ':3
Ol
•O
1 ':3 Ol 1 '3
1
m 1
1 ':3 i 1 '3
ft
^
in
00
0
P-,
0
i=>
0
X
ü ft ft
^ta 1
C^ 1
ta 1
>10 ^m
1 '0) '0) 'ta 'td 1
c- 1
1 c^
1 '0 10 ltd ^
1
Q
X X X C
1 'ta
1 c^
1 'ta
>ra
Itö Ita
C'
'ta Id
>m >m
CO
ta
0
0
H
0
S
0
ü
0
ta
<1:
E
Kl
0
ca
0
>>
td
«Í
ta
rH
<5
0)
ta
0
X
iH
0
5
•H •H •H td
s: Kl 1-5 1-5
49
,
^H
IH I H ' r - t n
1 C C Ö C c
c
a
1 <cö C
' S "•ta »ta ^ta ^td *ta
AJ
AJ
c
c
m 0
M
Ä K
O
c^
cö
m
ct!
M
»
1 'H 'H
1
El
1 "•cta •^
ta
,,
1
1
1
1
1
1
ra 1 Ita
m
1
1
1
1
1
1
1
1
1
1 •^H
1 IC
1 "•ta
M
» ft
00 T3
C
C
• H CO
»-S - Q
^
01
•H
3
m -c
1:
^OJ
—
„
'ta
•o)
1
1
1
»CÖ
TD
'Oj »cfl
AS A!
/•CD
^CS
m
'ta
Ai
1
1
1
1
1
1
1
c
-p
IO"
1 '3" '3"
•P
1
Ita
Ita
Ita M *ta *ta "•ta *ta
>C>3 Ita
1 >5 'ta ^ta 'ta Ita
1 >j
•H
'H
-p U1 •f
•p
^ta
'ca
1 >[SI 1
I
1 1
1 1H 1
p
ra ra 11 T i
ita
^ta
^ta
'ta
'ta
^ta
I
'ta
1 Ita
1
l i e
I
I
I
"•O"
I
1 1 '3" 'O" 1
•p -p
'ta
J«!
»ta
'ta
A!
vo
o
3
»
u
<0
•O
I
I
1 '3"
-H
tn
o
c^
ta
^
ta C
>>
1 C
O
••ti •rH
H
H
Ä
I
I
13"
1
13"
^ta
'ta ^ta
Itö
^
^ta "•ta 'ca ltd
>ü >ü - e - >ü
I
I
1
"3"
-p
^ta
^ta
>tsl
1
-p
—^
1
1 •"ta ^ta
1 'ta »ta
>o
O
c-
3
•fti
•P
Ä
C
D
•ü •
IH C
3 dJ
K
1
-C 1
1
1
1 >3"
1
1
1 ^3n
1 '3" '3"
>ü >ü
-e-
>o
'3
1
IH
-p
\
'3"
>o
'On
'CV
>ü
1 '3"
1 '3"
1 >>
P
»H"
p
1
1 XQO 13"
1 >i
'3"
-P >ü
1
I
t
1
1
1
1
1
1
1
1
1
^
o
m
w
B
•i«
O
EH
#
O
W
OJ
o
ft.
M
n
o
s:
w
H
50
^
1
1 iÖ
1
1
1
1
1
1
1
1
1
1 fö
1
1
1
1
1
1
1
1
1
1
13
1 Nr-K" IC ' H O Ir+n
1 •»rri "•H
-p >ü >ü
B
1
1
1
S
•»H
•
-p
1
1
1
1
lr->0
1
1
P
1
1
t
1
1
NH"
1 'H"
P
1
1 IC
1
m
p
1
>H
(¡1 C
•r+l 00 ^
1
C -rt
•H C
O1
•H
1
XH C
K
*
tn t>
s
(K
O E
6h
K
to
Oh
3
•H
o
1
1
1 'Hwi
cfl
CD o
PH tj
H
O
s
O
ü
O
tö
<
E
m
O
ta
o
>ü >ü
>H
^H
IH"
IH"
>o
P
0)
s
ta
tj
>i
ta
ta >» (H
X <C <
'H
"•H" 'H" fC
<H"
'H
>ü >ü -P
1
1
1
X rH O B
•H •H •H ta
S
t-2 1-3
•»H
J»
1 c^
1 <ca 'IS •>0)
1 Tí •tí •tí
c
ß C
a) 0
T3 0
Q.
C«
fe
•H
'
Jt ^3 •>3
Vd
•o •n
•tí
c
c c
4> 3
1 C' c^
1 /(tí <cd
1 -tí •tí
C c
IH
3
1 c^
1 Id]
1 T)
C
1
1
1
icd
•tí
Ö
K
ä
O
O
^
O
o!
»H
M
yo)*»
0)
5
^isy ^c!ín
<0)" ^a)^
s 3
^OJ" ^td" »CU"
vojn »ajm -a)"
3 3 3 3 3 3
4<¡
M
3
3
(1)
»H >^
>ü
I «
Vi)"
lot"
V
í
a
n
^njn iJ
vta"
3 3 3 -H
J«!
o
«D
®
M
fe
tJ
01
1-.
1
1
1
fe
v®
1 C' 1
yCU 1
-p 3
1
^—
M
1
1
1
yoj »oi
^ca
^nj loJ
»td
1 c^
'i
1 ^ta ycS <•01 •'oí ^td ^0) •O
• 3 3 3 3 3 3 3 3 3
M
CV
O
•H
+5
0)
S
M
3
•a
(D
3
c
•H o
o
c
o
o
3
H
a
o
EH
o
o
o
w
o
o
w
3
c
a)
«
>ü >o
^0)
3 3
J«!
1
1
1
1 ^H <H 1
1 <H »H 1
1 3 3
1
4<l
1
1
1
<0 c
a -M
01
Q)
L,
00
fe
^
^r-i" >ra
>01 O<H
3 3
•c
IH
1 -P >o
1 ltd »a!
1 3 3
1 >o
1 »a)
1 3
1
1 ^r^
1 3 3
tí
Ai
tí CC
^ 11
<0) ycd
Ai A!
1
1 ^H 1
1 ^H 1
1 3
1
1 -P
1 ^Oj
1 3
1
IH ^H
1 ^H <H
1 3 3 3 3
M
4>i
Al
^H
1
1 1 -»•
1
1 1 >.H
1 1 1 3
Ai
X ^H
X
'
l-H ^H •H
»3
3 3 3
4<í Ai
1
1
1
13 13 13
tí tí c^tí 11
c^cd »0J ^Oj 1
Al A! Ai
<•3 0
13 lO 13 13
• tí Q. 1 tí tí tí tí
c1
B 1
1
H 1
lo! ^(d
Ai fe
A! A! Al Al
oo
m
»3« »3
1 H ^H
1 >ü >o 1
1 /tS icd 1
1 3 3
1
4<l
1
1
1
'
1
1
13
^tí
Al
00
CO o
E
CU
o
fe,
oJ iH
O O
o s
o
o
o
Oj
-p E
<
o
Id
o
>»
o!
XI
>>
<
0)
rH
<
+5
(1)
ü S
<0 •H •H •H cd
o S CO
t-2
51
0)
0)
>>
»
(I)
3
00
c
0
K
1
1
" <aj
>N
1
1
1
1
1
1
' d »(d C
lOJ C
'd C
'd
»cd »cd »cd C
' d d »cd
>N >N >N >rsl >> »c>>
iH
1
1 »cd »cd C
'd
1 »cd <cd »cd
>tsi
>>
1
1
1
o
T—
•r«
3
•r«
1
ay
>>
#
K
0)
Q.
0
(U
a.
K
1 >N
1 1
1 ^aí
id
IH
A
IH <H
1 >IS)
1 1 IH
1
la c
Ii«
M
¡3
>N >N
1 1
'td C
'd
ÍC
1
1
1
'H
»H
IH
1 >> IH
1
Jt
»3
1 '0) 'H« 13
c
c
1
1
1
'i
1 IH »H
1 Jt
1 IH »H »H
(C ÍC
O
O
3
•H
iH
#
'H
>o
' P
rH
'H
H
tí
»H
rH
' 3
rH
•O
IH
rH
13
H
»H
iH
13
iH
»H
rH
' 3
rH
»H
rH
»3
rH
13
13
rH
IH
rH
lO
lO
rH
13
13
iH
lO
c^
lO
iH
' 3
' 3
rH
00
O
(U
®
X
!(£
, ,
ä
<U
ftC
1
1
1
C
' d >0)
C
' d »0)
VI >01
1
1 »<D »0
1 lO IS
X X
»Cd C
'd
»cd C
'd
X CO
1
1
1
1
1
1
^tö
ltd
>01
I
I
I ICÖ ^cö '05 ^cd
I >cö 'Cö 'cö icö
-e- m >ü m
I>
O
3"
c^
3
>>
1
•H
CO
»
^
^
vo
o
•H
m
#
0)
3
O«
•H f-i
U C
O
>3
dj
1 sgn
13" vgn
L, 91 1
»3
C3. e 1 »H" >> '3"
'3"
K
Tt TÜ >05 >01 >0) >01
1
1
1
1 »3"
1 c^
1 '3"
>01
I3IO
>:r>n >>
>m
>m
ft
o
u
ft
1 'O
1 >!
1 ..H
TU
1 lO »O
1 M
1 'H 'H
>01 >01
»O »O 'O
A!
»H »H
>01 >01 TC
1
1
1
'O 1
1
^H 1
>01
1 'O
1 /<!
1 ^H
Ol
1 »O 'O
1 ^
1 'H »H
JS >01
m
o
et:
o
pi,
52
PL,
3
•H
a
Q>
Pn
0} r H
O O
O s
o
u
o
cd
+>
<
E
ü
cd
o
>>
cd
X
<e
c8
rH
<<
Ol
E
cd
O
X
•H
s
rH
•H
CO
o
•H
1-D
E
cd
QJ •
u m
cu L,
í-H
Q.-0
C
tj m
L, «
•o
U
(
•o
(O *J ^
i U <J
<1
u
U
(
k
0 to
u
o.
WJ
0
^
K
q
0
u
4J
0
Ss
•o
3
u
u
01
K
c K
L,
3
m
•Q 91
0
0
4J
c
°0
o
u
00
>> 51
m
i >>
V K
K
•srH 1-.
3 >>
C O
•O
m
—.
•ií K
C
•H 0
01 L, 0 K
iJ
•V! •O »H
<0
m
O 0 01
c
01
e
ft
_ «
C-V
'
0
OJ cv 0
T
—
.—,
—.
^—
•vt m
O
o
1-3
1-3 1-3
1-3
a. CM r^
«
— vO vO ITV
O T
.—
r-
NrH NH
-p
X x
C
' d •e'ö »(S NH
M
«
Ä
Nr«
01 -P
yri
c •
•H ^
•o C^
C 00
o »
k
K
.—,
.
c\¡
o
vO
1-3 1-3
O
m
1-5
—
Ih
«
—.
vo
ís a
o 1 L,
»J
01 c
c
o
o
0 --I tt
ij
4J
S
01 3 ^ 3 0
U Q.
e o <0
o o
O O
U c OOM
(h
9)
3 o4
' -1 •-H
0
0 u tJ 0 O
01
M
«
«
—^
o
ft
«
5
O
u
c
3
IH
O
C\í
1-3
,—.
.—>
w
^—.
T—
m
—-
C'
s
-O
c\i
3
'O
«o
C'
•>o von
»3
+3 >>
S
»0 »o 'O
A!
J-í
Ä »
»
«
^
»
01
«
o
—,
—
«
k
,—.
O
C\i
NH
3
«
Q.-I
m
H)
1H -O
c
o í!
U L,
0)
to
E-i
4J O
-I
0
c
W
W
w
H
<•:
z
m 01
'n "
01 01
o
U «
Ol
o
u
[t,
o
X
w
o
L,
L,
01
01 <ü
k
•o
L,
01
4J
01
01
k
k
oo
o
•o
01
•o
•H
0)
c
•H
k
L,
0)
0)
T3
k
01
ä
0
tt
q
IH
0
-Q
k
k
k
•O
ID 01
O
1..
0
o
k
01
c
•H
q
k
01
s
-Ü
to
01
a
01
o
43
k
01
e
•H
0)
^
0
01
k
k
t»
q
0
U
3
U
O
k
00
00
q
•H
01
k
oc
q
•H
01
0
4a
uo
(O u
^
•Q Ih
01
C V,
01
01
1H -
01 t-5
o
Ja hH e
E-c
w a: -Q
m
<
£C Cu
Oh
<
«m
01
3
Jü
•H
1-1
<
•o
c
OI
01
01
E-.
o
a:
,—.
-—s
O
t>
t—
1-3
»
^—.
.—.
—
. ,
—
-—
Ov
Oí •sí
C\¡
^
1-3
»
•sf
—
'•3 'OJ »H »3
01
c
C
VH »•H •>H
- •>-H
*
.—.
Oi
-—^
t—
—-
»H
+3 •P
»H
*
-—,
—
.V
t^
r1-3
_
C\í
r\j
,—.
,—,
,—.
-—
—-
00
'Cd" '11! "3
»td c'ta 'tö »tö
O
9X
m
O
m
o
00
•—
«
#
in
vO
1-3
,—,
—-
t>
00
vD —.
O^ T
—
Oi
<•0)
•o 3 c^
•H c
ta •^to
c
m +3
C cta 'ta Id ta *td
M
i)
»
—
#
o
*
#
53
K
to
•o
c
H)
Ol
K
<u
0
•O
<0
Ii
c
u
(J
l-.
0)
<0
K
Ol
c
!0
0)
Ol B
d) ä e
0) •o
•c
•Cl •-H
o
a •H
M (U
U c
(J
&
•H
iq
t-H
<u
Ol
0
0)
u
(U
•H
C
•o
Í0
Ol
J::
e
u
C
3
S
•H Ol
M
1H u
0
c
3
Ol H)
3
q
0 •H
•H U
O 'H K
m •Q c
^
Ol B)
Q. 6
e
K
•C
u
•H <0
0 B
0 0
u ij
Ol Ol
,—,
CO
(1)
-p
TI
rvj
.—.
1-3
O
o
K
t-
C
C\i
c\¡
1-i
^—,
m
X 'O
'O 'O
-o 13
-o
c
c
ÖÄ
»
#
w
H
W
to
w
E-1
<<
S
O
o
o
(t,
o
o
M
H
W
OD
<C
ÍB
PL,
<<
54
-P
T3
C
•O
Xc CÄ
m
u
01
•H
01
"
01
c
<0
B
0
^
ft
fe fe
>>
3
^
ä
O
--i
0
u
01
ON
00
—
'X
ft in ,—, f—- ^—s
o
O
ir\
NO
,—,
—
—
cv
00
—
ft
00
Nü
—
o
'O
•H
a
X
c
fe
t)
di
,—,
c\¡
m
1-3
^—
nO
í
roi
ft
t--
—
.
1-3
ft .—s
OJ
NO
—
CNH
3
/oyi NO) '3 "•H
>>
>í o '3 'H 'H '(d ' O
-p >a
X
03 -p
1
X
/Ȓ c
'H
No)
'H NH •H 'td 'di N<13 'H
'ta 'td •H
-P
03 m Vi 01 03 03 m 03 CQ 03
-P
#
Ä
Ä
*
•-H
Ä
Vi
Ä
Ä
Ä
fe
-u
c
to
01
<11
(J
m
,—,
NO
O
T—
S
01
01
1
01
01
0
1-.
u
1-3
-—
NO
ITN
fe
oo
c
'H
•-i
K
00
C\i
1-3
ft
NH
-P
.—
o
^—
O
fe
a
tu
fe fe
c
OI
m
•H
ai
to
e
1H
00
fe fe fe fe
QO
•O
to
0
IH
•PH
O
^
to
tu
fe
ti)
fe
M
•M
•-H
to
ä
o
e
to
ft fe fe fe fe fe
a
to
0
01
c
>H
0
o
4J
0
0
u
01
o
to
>H
0
0
ta.
•O
c
10
J3
01
1-.
ä
^
fe fe fe fe
fe
fe fe
fe
fe fe fe fe fe fe fe fe fe fe fe
^
¡«I
w
Ö
s
M
<
CM
1-5
—
c-
Q
(O
3
C
•H
-p
C
O
ü
•
ft
—
. .
•—•
00
^
r-
1-3
•—•
00
c-
—
'H
lí
' 3
—
. - ft
NO
tv
Nojn
Nfl
s
M
#
r^
O
-—^
in
ft
CNi
O
T-
—
s
ta
Ä
'ta
s
O)
c^
03
S
Ai
»
T—
O
T—
-—-
o
o
T—
—
1-3
i—^
Os
ON
—
•H
Oí
4mÍ
«
•rí"
•H
•H
3
>!c
•H
s
M
«
ft
T—
—
,^
t-T—
1-3
<x¡ ft
—
—
'Cfl
>>
'H
'ta
A!
*
H
#
s
Nta
—
a.:
ON
OJ
ft
NO
tn
'ta
3
C'
Nta
rH
*
1-3
ft
NO
m
NO
'O
rH
, V
1-3
ft
00
o
NO
00
1-3
ft
00
ir*
,—,
CO
1-3
ft
^—
r-
t^
'tan
3
'H
C
'O
—
3
c^
3
H
Vf. ft
'Oí«
3
Nta
tí
«
'ta
c
c
ft ft
o
,—,
1-3
ccv
1-3
ft
tm
NQ
'O
c
ft
^—,
in
Oi
' 3
' 3
—
CN
Nta
c^
Ntd
t3
^
O
T—
•H
3
c^
ta
•a
ftd tíft tíft
3
C«
c
o
4J
k k
(1) c
(0
•H a
o
u ^
K
ai
•c
o
-Q k
OJ
«
Ol »H k k
Q. 01 c
Ol ••
O c
c
0
0 <0 II o
o
ca. «
•c
e
k
«
3
k
<u Ja
a. ^
o 91
1H
o
o
,—,
•-3
O
lA O
C^
t—
C\i O
r~
1-3 •-3
1-3
—^
k tk
OJ
T- T— o
T—
Cm
-—
-—•
•<tl
00
-—
S
•rM
sctf«
>>
»0
C^ 3 c
1
'O
(ÜJÍ
»tu \ca
»0
'3
>>
>1 >> >» >» >>
>>
>>
3« »
»
Ä w
^—.
—
-N
o1—
-—•
o)
Id
>.
k
C
•H
aj k
ij tH k k
c
01 01 c
3
01 ä
0
IH o
4Ü
E
—
,^
t^
1-3
OO
- •
»3
¿i
»3
>>
k k
01
u
c
0 3
O
01 e
—^
xO
C^
O c^
C\i T— O T— 1-3
1-3
•-3 1-3
1-3
O
«
00
m
—
• o •>3
N®
-P
<3
»3
•>3 '3 •>3
Ä «>> Ä >>>
!c Ä>í
.—.
k
m
k
e
•O
m ä
01 m
L. IH
«
01
.—^
1-3
vO
00
>
»3
>>
#
,—.
1-3
O
NH
s
•>3
>>
Ä
/—
T3
0)
3
C
+3
tí
O
o
CO
H
U
W
W
E-i
<
s
o
o
o
tu
o
X
M
a
s
M
iJ
<
O
t-t
E-i
k
k
k
H>
c
L,
01
00
c
•ri
k
tO
1-3
a.
T—
<
C
m
«1 o.
k
k
Ai 01 k
k
L« •o •a
J5
3 c L,
U, n 4J
5) 0
O •H m ^
L
,
.c kä
^
k
k
k
k
—
, ,
T—
^
1-3
O-
a.
^—'
T—
O
vD
•H
«
k
C\¡
^
O
<d
•r+n
k
01
00
c
•rH
ln
k
£1
•H
01
k
k
•
^
M
m
w
o.
«5
01
•H
•ta
C
01
1-3
^
n
O
—
(V
o
>.
0)
>»
-p
A
>>
1
•it<
-P
3«
•M
+3
«
—
.^
C\i
vO
1-3
A
3
-—
1
»0
•M
i
+3
»
» 0
k
k
L,
01
a
<0
a
k
J<
O
m
k
-vt
k
•O
0
0
k
OO
—
,,
00 <N
T—
t— T—
OS
1-3
00
0
4J
,—V
-—.
k
k
k
k
-—-
1—
X O" '3
+3
/ o »3
•p + 3
»
'—^
m
os
1-3
K
OS
OO
1-3
«k
sO
c^
»0)
»H
'US
3
+3 '3
ft
ft ft3
k
m
01
3
0
.c
k
k
^
.—,
01
01
ä
01
k
k
k
•O
ä
O 0
c u
k
,—.
sO
1-3
n
00
C\i
>1
ft3
k
ä
01
c=
k
,—,
,—.
•
00
03
ft3
T
— T
—
1-3
1-3
-
«
Os
lA
«•l-P
+3
X
"•H
3
ft ft3
'—^
o
•—
01
<D
X
ft
h
00
k
<0
e
o
u
Os
T—
1-3
-—N
m
<01 »3
tí
«01
X X
ft
k
,—s
Os
T—
T—
1-3
cv
1-3
4J
O
0
ft
_
O
f es
'O
>>
' H
X
ft
55
NOTES
1 What traditionally
Is
unintelligible
dialects
separate
(M-SE)
tend
to
fuse
Silacayoa
tisi,
yatei,
Diuxi
can
established.
be
kumi';
[ +A ]
(Daly
as
used
can
k4ti(M)
but:
high-high
class
in
this
»high-low,
for
Jicaltepec
Cahuatache,
kuñu
or
final
no high
in
the
tone
revised
equally
zikiÍM)
But
note
Daly
nouns
like
úú,
but
conditioner."
analysis
but:
< »uwi,
ú'ú
(sandhi
class
class
III)
of
Diuxi
[+raodify]:
be called
I)
I)
kó'ó{M)
|| <
»i^.
kúmí < » k a w ^ ' ,
yü'u <
una < » o n ä ,
»yú'ú'í
low-low
<
< »slt^',
glottal
of
'^da'á,
< »siki',
»i^.
a basic
class
dialect.
reflex
Ocotepec
díkí(M)
(sandhi
the
and one
e.g.
<
(sandhi
in
and
yiwi.
sü'u < »yú'ú''.
stop
Mix-
sitó
correspondence
i ti^"', y u ' ú '
low-low
II)
t^ayu,
no clear
dialects,
< »kiti',-
Coatzospan
< »wé'yí,
Atatlahuca
yüsü and i n ^ !
Metlatonoc
sá'^má',
high-high
< »sini'',
Possibly
as
except
Metlatonoc
yafei v s . ñ^^,
are
1978)
Jicaltepec
glottal
káá,
t-^na;
kiimi with
síní
8 Cf.
56
'there
like
< »koo',
and
si'^e and t ^ i á ;
úú < » u w i ,
but:
104):
are
(M-SM)
s e ' e and s a n u j
kúmí(M) < » k i w | ' ,
Peñoles
< »yíkí'j
(sandhi
sö'ö,
other
z u ' ú and Cahuatache
7 Cf.
mutual
treated
» l o w - l o w with
and Coatzospan
"dá'á',
Ayutla
»low-low"';
tone
be
be reconstructed
kü'feä,
tayi,
sa'tea o r
näma • For
Diuxi
to
of
taá.
Alacatlazala
(1973:
not
pan and Metlatonoc
Silacayoapan
all
better
Proto-Hixtec
kaä and c á n ü j Ayutla
6 Cf.
groups
data.
Metlatonoc
5 Note
are
San Miguel
» l o w - l o w could
supplementary
tepec
probably
two dialects
therefore
4 Note
Mixtee
languages.
2 Longacre's
3 Note
called
Note
reflex
stop
of
k " í á < »k'^lya,
the
»sxt^'
became
appearance
fei'i
of
a
and other
a medial
glottal
medial
forms.
stop
9 Cf,
animal
names
kinship
terms
"yéni{
numerals
10 There
is
evidence
like
like
like
another
for
compounds
modifier
11 The couplets
complication.
it
from
that
Proto-Mixtec
are
following
low tone
Xayacatlán
and
13 Longacre
and
at
Rensch
hand
and as
in
for
tones
anyone
101):
by
can
on
be
the
Intrude
of
ones,
er
his
(Kuan
1977:
15 Longacre
not
result
stop,
high-high
Ocotepec,
two
of
to
1982),
Ancient
became
Josserand
was
(1983:
presented
will
have
an
Proto-Mixtecan
that
with
a similar
the
to
fourteenth
b{
century
meaning
literary
form
Chinese
Ancient
Chinese
evolution
similar
present-day
er duo,
three
characteristic
which
From the
novels,
to
attitude
be
cf.
of
Chinese.
of
Proto-Mlxtec
mentioning
morphemes
related
as
A critical
seems
corresponding
e.g.
closely
on Longacre
reconstruction
As a result,
"ear"
from
(1978,
Chinese
words
yanjing,
in
two
language.
syllabic
and
perturbing
San Miguel,
on Proto-Mlxtec,
while
into
blsyllabic
and
to
reconstruction
worth
documented
spoken
for
data
and Josserand
may be
nucleus.
perturbing
Proto-Otomanguean.
who works
Bradley
in
an
the
glottal
rules
heavily
reconstructions
on compounds
to
did
"A new reconstruction
effect
14 It
forms
a final
gives
high-mid
these
sandhi
leans
phonemic
these
of
have
to l o w - h i g h
Proto-Mlxtecan
towards
«sísi,
Mixtepec.
had only
dialects
alternation
never
with
recent
(1984)
high-mid
all
forms
similar
^iau;
«sito,
semantics,
h i g h - l o w , Molinos
power
seems
Dürr
identical
showing
San Miguel
and Cacaloztepec
12 There
having
but
high-low
-
«k^á'^á, * x e n ú a n d
«uwi,
still
Identical
^ í s o and
" í n á , '^láá,
»sa'yi,
yan
Chinese
"eye'
began
next
to
has
many
mono-
became
"nose"
became
phonemic
tone
biz i
6).
(1957)
Proto-Mixtecan.
reconstructed
Tone
four
3 is
a level
tone
between
levels
mid
low.
57
16 Tone
^ =
is
indicated
follows:
' =
Daly
couplets
(1978).
stress,
the
exclamation
which
downstepping
not
notation
a larger
list
be
I used
the
Diuxi
reanalysis
of
tone.
In
a process
word-
Coatzospan,
phoneme
causing
in
Peñoles
* modified high
Cahuatache
of
nonpredictable
t o n e . /Is an exception,
for
used
by
tone.
Schultze
(1938)
unequivocal.
list
sets
a feature
= low,
that
to
indicates
l o w t o n e and
totally
can
according
indicates
terrace
17 The following
time
is
mark
modified
The tone
given
The apostrophe
final
indicates
are
high, ~ = mid,
= r m h . Note
fhl, ^ = fml, " = rlh,
tonemic
is
as
contains
of
320
found
my M.A.
system
at
hand
and Josserand
(1982)
the
reduction
of
not
as
evident
reconstructed
in
a different
I had not
110 cognate
Of
(Dürr
are
in
255
but
for
there
at
done
(1983).
data,
from
these,
1982),
reconstruction,
and Josserand
they
selected
reconstructions
presented
as
forms.
thesis
of
the
sets
that
by
Bradley
/Is a result
several
details
consideration
of
may
of
seem
all
cognate
sets.
18 Josserand
final
(1983:
glottal
Zacatepec
19 This
stop
(both
form
area
cognate
is
A - high
20 The form
Diuxi
either
the
to
because
dialects
in
the
that
is
been
21 In
in
to
cited
as
Coatzospan
of
Jicaltepec
incompletive
58
enough
the
and
basic
data
aspect
(1 978)
and
basic
couplets
couplets
Ayutla
at
for
which
-
322)
in
corresponds
tone
reanalysis
of
hand
to
most
Diuxi
tone.
have
is
inconsistent.
cannot
occur
isolation
h i g h - h i g h ! (nouns)
high-low
(verbs),
couplet
form
to
tone
forms
couplets
special
or
Diuxi
standardize
notation
tonemic
have
and
word).
low-'high / 'low-high
data
source
of
from
a
feature.
Therefore
couplets
basic
of
reconstruct
and Oram 1976:
Daly's
(1978).
These
for
for
jectives)
22 In
not
not
a non-cognate
couplet
a number
isolation.
low-high !
high-rlh
in
lack
show
(Pike
tú ' t ú
Daly
of
did
a characteristic
tonemic
I bad
according
22)
continuative
tone
low-'low / 'low-low
Note,
set
variants:
and
!fhl-high,
high-high !
(ad-
high-high.
the
basic
continuative
form
either
a couplet
l o w - m i d has
high-high
or
as
low-low.
REFERENCES
A l e x a n d e r , R u t h Maria
1980
Gramática
Mixteca
Bradley, C. Henry
1970
A Linguistic
Sketch
- Mixteco
of Jicaltepec
de A tatlahuca.
Mixtee.
México.
Norman.
Bradley, C. H e n r y , and J. K a t h r y n Josserand
1982
" E l p r o t o m i x t e c o y sus d e s c e n d i e n t e s . " In Anales
( t o m o II): 2 7 9 - 3 4 3 , M é x i c o .
Daly, J o h n P.
1973
A Generative
Syntax
of Peñoles
Mixtee,
de Antropología,
19
Norman.
1973a
" T o n e Sandhi and Rule Ordering in Peñoles M i x t e e . " In Workpapers
of the
Summer
Institute
of Linguistics,
University
of North Dakota,
17: 8 0 - 105,
Grand Forks.
1977
" A P r o b l e m in T o n e A n a l y s i s . " In: Wilham R . Merrifield ( E d . ) : Studies
Otomanguean
Phonology,
pp. 3 - 20, Arlington.
in
1978
" N o t e s o n D i u x i M i x t e o T o n e . " In Workpapers
of the Summer
Institute
Linguistics,
University
of North Dakota,
22: 98 - 113, Grand Forks.
of
D a l y , J o h n P., and Margarita H o l l a n d de Daly
1977
Mixteco
de Santa Maria Peñoles.
M é x i c o . ( A r c h i v o de lenguas indígenas de
México 3).
Dürr, Michael
1982
Töne und
mixtekischer
1984
Tonveränderung
im Mixtekischen:
Vergleich
der
Dialekte.
M. A . thesis, Free University o f Berlin.
Tonsysteme
„ D i e E t y m o l o g i s i e r u n g v o n K o m p o s i t a im M i x t e k i s c h e n . B e m e r k u n g e n zur
P r o b l e m a t i k anhand einiger B e i s p i e l e . " In Indiana
9 (Gedenkschrift Gerdt
K u t s c h e r T e ü I ) : 1 8 9 - 2 0 6 , Berün.
D w y e r , David
1981
" L o m a , a Language with Inverted T o n e s . " \n Anthropological
23: 4 3 6 - 4 4 2 , Bloomington.
D y k , A n n e , y Betty S t o u d t
1965
Vocabulario
mixteco
de San Miguel
el Grande.
Linguistics,
México.
H o m b e r t , Jean-Marie
1978
" C o n s o n a n t T y p e s , V o w e l Q u a l i t y , and T o n e . " In: V i c t o r i a
( E d . ) : Tone - A Linguistic
Survey,
pp. 77 - 111, New York.
Fromkin
Hunter, G e o r g i a , and Eunice V . Pike
1969
" T h e P h o n o l o g y and T o n e Sandhi o f M o l i n o s M i x t e e . " In Linguistics,
2 4 - 4 0 , T h e Hague.
47:
59
H y m a n , Larry M.
1978
" H i s t o r i c a l T o n o l o g y . " In: V i c t o r i a F r o m k i n ( E d . ) : Tone
Survey,
-
A
Linguistic
pp. 257 - 270, New York.
Josserand, J. K a t h r y n
1983
Mixtee
Dialect
History.
Ph.
D. dissertation, Tulane University,
New
Orleans.
Kuan, Y u - c h i e n
1977
Die Grundregeln
des modernen
Hamburg.
Hochchinesisch.
L o n g a c r e , R o b e r t E.
1957
Proto-Mixtecan.
International
Journal
of
American
23.4,
Linguistics,
part 3, C h i c a g o ( P u b l i c a t i o n 5 o f the Indiana University Research Center
in A n t h r o p o l o g y , F o l k l o r e , and Linguistics).
Mak, Cornelia
1953
" A C o m p a r i s o n o f T w o M i x t e e T o n e m i c S y s t e m s . " In International
nal of American
1958
Linguistics,
" T h e T o n a l System o f a Third M i x t e e D i a l e c t . " In International
American
Linguistics,
jour-
19: 85 - 100, Chicago.
Journal
of
24: 61 - 70, Chicago.
Mak, Cornelia, and R o b e r t E. L o n g a c r e
1960
"Proto-Mixtec P h o n o l o g y . " In International
tics, 26: 23 - 40, Chicago.
Journal
of American
Linguis-
Merrifield, WiUiam R . ( E d . )
1981
Proto-Otomanguean
Kinship.
Dallas.
N o r t h , J o a n n e , and Jana Shields
1977
" S i l a c a y o a p a n M i x t e e P h o n o l o g y . " In: William R . Merrifield ( E d . ) :
in Otomanguean
1978
Phonology,
Studies
pp. 21 - 3 3 , Arlington.
Gramática Popular del Mix teco de Silacayoapan.
México.
Overholt, Edward
1961
" T h e T o n e m i c System o f G u e r r e r o M i x t e e . " In: B e n j a m i n F. Elson ( E d . ) :
A William Cameron Townsend en el Vigesimoquinto
tuto Lingüístico de Verano, pp. 597 - 626, México.
Pankratz, L e o , and Eunice V . Pike
1967
" P h o n o l o g y and M o r p h o t o n e m i c s
Journal
of American
Linguistics,
Aniversario
o f A y u t l a M i x t e e . " In
del Insti-
International
33: 287 - 299, Chicago.
Pensinger, Bren da J.
1974
Diccionario
tepec, pueblo
60
mixteco-español,
de Chayuco).
español-mixteco.
México.
(Mixteco
del este de
Jamil-
Pike, Eunice V . , and J o h n H. C o w a n
1967
" H u a j u a p a n M i x t e e P h o n o l o g y and M o r p h o p h o n e m i c s . " \n
cal Linguistics,
9.5: 1 - 1 5 , Bloomington.
Anthropologi-
Pike, Eunice V . , and T h o m a s Ibach
1978
" T h e P h o n o l o g y o f the M i x t e p e c Dialect o f M i x t e e . " In: M o h a m m a d Ali
Y a z a y e r y , Edgar C. P o l o m é and Werner Winter ( E d s . ) : Linguistic
and
Literary
Studies
in Honor of Archibald
A. Hill, v o l . 2 (Descriptive
Linguistics), p p . 271 - 2 8 5 , T h e Hague.
Pike, Eunice V . , and J o y Oram
1976
"Stress and T o n e in the P h o n o l o g y o f D i u x i M i x t e e . " In Phonetica,
3 2 1 - 3 3 3 , Basel.
33:
Pike, Eunice V . , and Priscilla Small
1974
" D o w n s t e p p i n g Terrace T o n e in C o a t z o s p a n M i x t e e . " In: R u t h M. Brend
( E d . ) : Advances
in Tagmemics,
p p . 105 - 134, A m s t e r d a m .
Pike, Eunice V . , and K e n t Wistrand
1974
" S t e p - U p Terrace T o n e in Aeatlán M i x t e e . " In:
Advances
in Tagmemics,
p p . 81 - 104, A m s t e r d a m .
R u t h M. Brend ( E d . ) :
Pike, K e n n e t h L.
1947
" G r a m m a t i c a l Prerequisites t o P h o n e m i c A n a l y s i s . " In Word,
Worehester, Mass.
Pike, K e n n e t h L.
1948
Tone Languages.
Ann
3 : 155 - 172,
Arbor.
R e n s c h , Calvin R.
1976
Comparative
Otomanguean
Phonology.
Language S c i e n c e M o n o g r a p h s 14).
B l o o m i n g t o n . (Indiana University,
Schuh, Russell G .
1978
" T o n e R u l e s . " In: V i c t o r i a F r o m k i n ( E d . ) : Tone
pp. 221 - 2 5 6 , N e w Y o r k .
Schultze Jena, L e o n h a r d
1938
Indiana HI — Bei den Azteken,
del Sur von Mexiko.
Jena.
Mixteken
und
- A Linguistic
Tlapaneken
der Sierra
Survey,
Madre
Zylstra, Carol F.
1980
" P h o n o l o g y and M o r p h o p h o n e m i c s o f the M i x t e e o f Alaeatlazala, G u e r r e r o . " In S. /. L.-Mexico
Workpapers,
4 : 15 - 4 2 , M é x i c o .
61