Banksira, Degif Petros (2000) Sound Mutations: The Morphophonology
of Chaha, John Benjamins, hardback, xxxi 332 pp., $75.00, ISBN:
Reviewed by Peter Unseth, SIL International, Graduate Institute of
Linguistics, and Univ. of Texas at Arlington.
Publisher's announcement posted at:
This book is a cause for celebration: a native speaker of Chaha (the
most phonologically complex Semitic language of Ethiopia), has
described its morphophonology in a rigorous, theoretically informed,
and comprehensive way. With this book, the study of Chaha takes an
undisputed place in the front rank of Ethiopian linguistic studies.
For the wider audience of phonologists, this book provides an
examination of fascinating problems and an application of the tools of
Feature Geometry and of underspecification to difficult problems that
have been resistant to analysis or have been previously analyzed in
different ways. The author, Degif Petros Banksira (DPB), has
previously published several items related to Chaha (under the name
"Degif Petros", Ethiopian naming customs conflicting with Western
ones). This monograph, derived from his dissertation, covers many
points of Chaha's complex phonology.
I must first state that I am reviewing this book more from the
perspective of an Ethiopian language scholar than a pure phonologist.
A more specialized phonologist, especially one without a background in
Ethiopia, would value and evaluate the book in different ways. DPB's
theoretical approach involves extensive use of feature geometry and
underspecification, with only two mentions of Optimality. He also
uses the approach of Distributed Morphology to organize his discussion
of subject and object morphology.
I am disappointed that the book's title follows the trend of putting
only the topic in the title, relegating the name of the language to a
subtitle, at most. This practice often results in books being cited
in lists (catalogues, bibliographies, "publications received") without
their subtitles, leaving readers unaware of the language(s) a book
describes. Admittedly, this opinion is colored by my descriptivist,
rather than theoretical, outlook.
I suggest that readers begin by reading the Introduction and the four
page Conclusion, a good summary of the book's findings.
Some of his many fascinating and notable points include:
* /t/ is a default consonant, unspecified for laryngeal features, an
obstruent rather than a sonorant (p. 79).
* Roots of the consonant pattern 1212 have different rules than roots
of the shape 1234 (p. 180)
* He eliminates the need for a separate class of Type B verbs in
Amharic, interpreting them as containing a second radical that is a
vocoid (p. 83-88).
* DPB gives a principled explanation for why some words have
labialization on velar and labial consonants, arguing for a biphonemic
analysis of these, a topic which has long been debated in Ethiopian
Semitic circles (p. 136,137).
* He proposes that there is no phoneme /k/, but that it is an
allophone of /x/. This is counter to pattern symmetry in the phonemic
chart and to what has been analyzed in all other Ethiopian Semitic
languages, but he uses this analysis to write consistent rules.
* He presents a principled explanation for why some verbs appear to
violate the Obligatory Contour Principle by having roots of the
consonantal shape 112. He derives them from 1212, with a deletion of
one consonant (p. 71-73). His proposal seems to also be applicable to
similar exceptional verb roots in other Semitic languages, Tigrinya
having a notable number of such verb roots.
* Despite the claims of some, he argues Chaha is templatic, not a
surprise for a Semitic language (p. 114).
* He argues that labialization and palatalization of different
segments in a word are often caused by the segment /U/, the [high] and
[back] features linking to different segments, e.g. t'IBBIs 'roast'
and t'IbWIS 'well roasted' ("I" here is his epenthetic "barred i", and
"S" is palatalized "s") (p. 231), also (p. 187,201).
DPB uses the term "doubling" to refer to different things, a practice
that can lead to confusion. He uses it to refer to roots with the
consonantal sequences 122 & 1233, also for roots of the shape 1212,
and for the results of reduplicating the penultimate consonant in the
intensive form 1223 < 123. He uses the term "geminate" to refer to a
phonetic lengthening of a consonant, different from the way many
non-Ethiopian Semitic scholars use "geminate" referring to roots of
the 122 pattern.
In healthy contrast to a broad tradition in Ethiopian Semitic
linguistics, DPB's approach is very strongly synchronic, (though he
admits that some cases of /A/ act differently because they are from
different historical origins p. 107). He makes good use of
comparative data on a number of points (e.g. p. 108, 194, 199,
227-229), but derives forms by applying rules to contemporary (if
abstract) underlying representations, even to the point of proposing
an abstract vowel represented by the /ae/ digraph (p. 107). (Some
will be reminded of the controversies over history and levels of
abstraction in The Sound Pattern of English.)
His handling of the status and behavior of the epenthetic vowel (his
barred "i") is not as clear as I had hoped. The vowel is inserted by
rules at certain points (p. 25ff), but I did not always understand how
this vowel appeared in various forms, a problem intensified by the
fact that it was not always clear if cited forms were phonetic or some
level of abstraction (p. 110, 157, 172, 181, 203). I presume the
problem is not a flaw in the author's analysis, but simply a matter of
his presuming that readers will retain and correctly apply a large
number of rules and lexical information without much additional
guidance. But the word 'fire' is problematic, with an initial
epenthetic vowel (p.258).
DPB is gentle in disagreeing with others, sometimes citing other
publications for the reader to compare a different analysis (e.g. p.
128, fn. 5). He corrects some forms that others have cited, pp.
237-9, but he does so gently. He is also very forthright about
listing forms that are exceptions to various rules. (The two
exceptional forms on p. 137 might conform to his rules if he tweaked
his rules to delete the vowel slot when the following consonant links
to the preceding C slot.)
The last two chapters cover subject affixes and object clitics. At
first, this may seem unrelated to the study of phonology, but DPB
shows how his analysis of the phonology provides a more unified
explanation of certain problems in the allomorphy. In his discussion
of object clitics, he runs into problems on p. 267 by not noting the
differences between datives, benefactives, and accusatives that
originate as datives (e.g. "She gave me a plunger", where "me",
originates as a dative).
Two errors in references may confuse some. First, the article by
Prunet (1990) was published in the International Journal of AMERICAN
Linguistics. Secondly, the article cited in the text as Banksira
(1999) is cited in the References as "2000", having been printed in
Lingua Posnaniensis 2000, pp. 7-18.
Though I have pointed out some shortcomings in the book, these faults
are miniscule in a book that proposes innovative and comprehensive
analyses of a language that has been studied by many great scholars.
His work will not be the ultimate word in Chaha studies, but certainly
he has moved work further forward than any other author.
Since many of the issues that DPB handles are similar in other
Ethio-Semitic languages, the book will obviously be required reading
for those studying any Ethio-Semitic languages, as well as
phonologists who work with feature geometry. Since the book is rich
in data it will be useful for years to come, so libraries should be
encouraged to buy it now so that future phonological theories can be
applied to the Chaha data.
I worked in Ethiopia 12 years, sent by SIL, most of my time there at
Addis Ababa University. I am now finishing my dissertation at the
University of Texas at Arlington, writing on a reduplication pattern