LINGUIST List 16.454
Tue Feb 15 2005
Review: Morphology/Syntax/Lang Desc: Zeisler (2004)
Editor for this issue: Naomi Ogasawara <naomilinguistlist.org>
What follows is a review or discussion note contributed to our Book Discussion Forum. We expect discussions to be informal and interactive; and the author of the book discussed is cordially invited to join in. If you are interested in leading a book discussion, look for books announced on LINGUIST as "available for review." Then contact Sheila Collberg at collberglinguistlist.org.
Relative Tense and Aspectual Values in Tibetan Languages
Message 1: Relative Tense and Aspectual Values in Tibetan Languages
From: Jakob Dempsey <jakobsaturn.yzu.edu.tw>
Subject: Relative Tense and Aspectual Values in Tibetan Languages
AUTHOR: Zeisler, Bettina
TITLE: Relative Tense and Aspectual Values in Tibetan Languages
SUBTITLE: A Comparative Study
SERIES: Trends in Linguistics, Studies and Monographs 150
PUBLISHER: Mouton de Gruyter
Announced at http://linguistlist.org/issues/15/15-2883.html
Jakob Dempsey, Foreign Language Department, Yuan-ze University, Taiwan
The first one-quarter of the book is a theoretically-oriented introduction to
the issues which take Tibetan as their focus in the major portion of the
book. Among these issues are ways to conceptualise events, which include:
A) "Type of Actor" (presence or absence of Control, Agency, or Volition)
B) "Type of Event" - one breakdown would be: achievement ('to break'),
accomplishment ('to grow up'), activity ('to flow'), and state ('to sit'). These
categories can be morphemic alternations such as German or English
jagen 'hunt' (activity) vs. erjagen 'hunt down' (achievement).
C) Absolute Tense vs. Relative Tense. The former includes divisions into
past vs. non-past, past vs. present vs. future, and there may even be finer
distinctions such as remote-past, recent past. etc. The latter was originally
conceived to account for the Perfect in English and some related languages,
and includes: "taxis" (the temporal ordering of closely related events), vs.
ordering vis-à-vis a contextually given reference point which may or may
not be the same as the time of the speech-act, among other functions.
D) Phase: a way to classify verbal expressions by focusing on certain
intervals in the development of the event; expressions may be prospective,
inceptive, continuative, progressive (dynamism added), egressive, or
E) Quantification: classify events as distributive, iterative, habitual, etc.
F) Aspect: Zeisler lists five "possible prototypical perspectives on the
realisation of events":
1) event as such - the totality
2) course/internal stages
3) preparatory (external)
4) result (external)
5) indefinite quantification (habitual, iterative etc.), and further defining
Perfective as the marked (+totality) viewpoint vs. Imperfective, the
unmarked (0-totality) viewpoint. e.g. Russian "Dokazyval no ne
dokazal" "He tried to prove (Impf) but didn't (Pftv)". (Zeisler reminds us that
applying this Slavic system to other languages can result in confusion, for
example, in the Romance languages there is an opposition of, e.g. the
French imparfait vs. the Passé Composé which is "commonly accepted ... in
terms of Aspect [but] ... turns out to be a complex of aspectual, temporal,
modal, as well as pragmatic functions".)
G) Framing: marked (+internal perspective) vs. unmarked (holistic
perspective). This category is illustrated by the English "expanded form"
construction ( is/was ...ing), which is informationally marked and can serve
as a Frame for an event with open interval.
Zeisler discusses these concepts, and how they interact with TAM (Tense,
Aspect, Mood) in Ancient Greek, the Romance languages, Literary Arabic,
English, Russian, Bulgarian and other languages, but the limited nature of
this review suggests we should focus on these issues as discussed in the
major, Tibetan-oriented, part of the book.
Part II, "the Tibetan system of Relative Tense and aspectual values" starts off
with a classification of Tibetan within Tibeto-Burman, but based on a rather
old source; the inclusion of Lepcha as a close relative of Tibetan is probably
erroneous, cf. Bradley 1997, which also rearranges Zeisler's subdivisions of
Tibetan. Her suggestion of a close connection between Old Tibetan
(evidenced from early documents) and north-eastern or Amdo Tibetan is
intriguing, and would suggest something about the movements of ethnic
groups early on in Tibetan history.
This reviewer should remind the reader that the comments expressed here
are from the viewpoint of a general-background scholar in Tibetan
linguistics with a particular interest in Lhasa Tibetan, who also has a desire
to learn something about the specialist studies in Tibetan verbal systems
which Zeisler has undertaken; the deep understanding of theoretical issues
in the first part of the book coupled with a familiarity and personal
investigation of many varieties of Tibetan, modern and ancient, is an
achievement which perhaps belongs uniquely to Zeisler among the global
community of language scholars.
Zeisler devotes quite a few pages to analyzing and explaining the
phonologies and transcription-systems she will be using. In matters of a
preferred transliteration for Tibetan, it is probably useless to assert that
one's own is less confusing or more rational, one can never satisfy
everybody. For Lhasa Tibetan, she takes the slightly centralised high front
and back vowels, transcribed in Chang & Shefts as e and o (both with raised
dot above) as non-phonemic, but that is not really the case in this type of
Lhasa speech which she is taking as her main source. I would agree with her
about the shortcomings in phonetic accuracy of Goldstein's extensive
publications on modern Tibetan.
This section also introduces us to the Tibetan verb, which lacks components
related to mood, voice, person, or number. Instead, there is a fundamental
division into verbs representing controlled or intended action and those
representing accidental or nonvolitional action. We are introduced to the
different possible stems of the Classical Tibetan verb -- four at the most,
where A is the unmarked stem, expressing simultaneity vs. stem B which
expresses anteriority; stems C and D express modal concepts. In presenting
such material, I think Zeisler's approach is a little too abstract to be a proper
guide for many readers: for example, at the very point of first opening the
discussion about the four verb stems, it would be helpful to give a short list
or little table of some common examples, so the reader could see some
concrete depiction of what is under discussion.
Part II contains a large section dealing with verbal forms in Old and Classical
Tibetan. Compound constructions, with some sort of auxiliary after the
verb-stem, are not as nearly common here as they become in later Tibetan,
but already can be seen. Zeisler bids us look more carefully at the semantic
content inherent in the suffixes, for example "bsad-par-'ong" would not
just be another kind of future expression, but specifically "is going to [have]
you killed." Her analysis of the Simple Future in these texts reminds one of a
common use in Lhasa Tibetan: a deliberate decision of the speaker to act,
also often carrying the connotation that one is morally or socially obliged to
act, thus laws, precepts and prohibitions are often expressed with this
Simple Future. Under #275 in this section, it seems the phrase "ensnaring
deer" should be changed to "ensnared deer". Zeisler concludes this section
stating that "the stems of Old and Classical Tibetan can be best described in
terms of Tense-R [relative tense] and Mood." Thus not in terms of Aspect or
Framing. In terms of Tense-R, "the present stem is positively marked for
simultaneity", whereas the simple past stem is "aspectually...unmarked", and
not expressive of Absolute Tense (Tense-A).
By the time that we get to modern dialects, such as Lhasa Tibetan, the
widespread conflation of the old separate stems leaves the differentiation of
Tense-A, Phase/Quant (Quantification) and Mood as the job of
compounded verbal constructions. The introduction of Aspect, to be
connected with ergative agent marking in past tense constructions, leads,
as Zeisler states, to more confusion and muddle than it helps; an article by
this reviewer pointed out years ago that the role of ergative marking is
much broader than this, whether in Lhasa speech or in earlier semi-
Zeisler's "phonemic transcription" of Lhasa Tibetan contains some
inaccuracies or misleading representations: e.g. p.472 "sha-'di zo!" (Eat this
meat!). I would agree with Zeisler that the falling contour on the
imperative "zo" is better attributed to sentence intonation, and should not
be marked the way it is in the Chang and Shefts publications, but the vowel
is short; this example also includes a peculiar marker of Zeisler's, a sort of
dislocated integral sign which is supposed to indicate "non-phonemic
suprasegmental features due to sentence intonation". Frankly, I don't see
the point of this device: for example, p. 495 has this symbol appearing in
the midst of "paa-lags" ( father ); Zeisler's transcription has a
falling tone on the second syllable, which is not appropriate since it is
completely unstressed, the vowel in "paa" is long simply because that is the
form of the word, there is no "prosodic trick" which can relate it to the short-
vowel written form.
Further down on the same page, the high-toned "ni" , which frequently
occurs at the end of phrases in spoken Lhasa Tibetan, is also marked with
this split-integral; I suspect that Zeisler is assuming the word is basically the
written "ni", which should be low-tone, so "something prosodic" must be
making it high-tone. Aside from the fact that this is simply the expected
tone for this morpheme in Lhasa Tibetan, there being thus no need
to "explain" its relationship to any written form, the historical antecedent is
more likely to have been "a.ni / e.ni" which still pops up all over the place in
Lhasa sentences and has the high tone already in place in the second
syllable. - In her transcription Zeisler has certain syllables spaced
completely apart, others written together, and others connected by a dash,
but the rules for such spellings are not at all clear and can in some cases be
quite misleading unless the reader is already quite familiar with Lhasa
For example, why does the unstressed genitive marker -khi get placed off
with a dash, but the unstressed dative marker -la gets stuck right onto the
noun it follows? (p. 473) Also, it seems that Zeisler's system marks the tone
on the first syllable of a disyllabic unit, and then the second syllable's tone is
predictable. Example #341b starts out with "ngarang", where we can predict
the low tone indicated on the first syllable to be followed by a raised (fairly
level) tone on the second, but later on we have "zaga:" (I have modified her
symbols for the sake of electronic transmission) where, with the same tone-
markings, we would expect the same tonal pattern on the second syllable,
but actually it is completely different, it is unstressed and low-tone. Why?
Because the connection between the syllables in "ngarang" is a close
juncture found within syntactic words, whereas the connection between "za"
and '"ga:" is a loose juncture found between syntactic words within a
phrase. This is obvious from #341c on the next page where, in "phebs-ga:"
(my retranscription) the tone on the first syllable is falling (as it would be in
pause-final) instead of leveled out as it would be in close juncture,
e.g. "phebs.ki-red". If one connects syllables with two kinds of juncture-
markers (I use a mid-raised dot vs. a dash) then the sentence structure and
the phonology are both clearer; so for example p. 482 #358 should
be "gjap-nä", the second syllable is quite unstressed, not falling tone as she
has it. The same goes for the ergative "-gis" suffix in #359 (p.483). On the
other hand, for example p.491 #375, in "byas-byas.par" the first two
syllables should, in Zeisler's system, be written together with no dash since
they are in close juncture: the first syllable's falling tone is leveled out.
Zeisler's linkage of the three syllables implies the opposite of the facts.
After this critical "excursus" on Lhasa phonology, let us, as far as this
reviewer is able, say a few more words about Zeisler's treatment of verbs in
various modern dialects. The Lhasa Tibetan section shows us at least 18
different types of verbal constructions, expressing prospective, experiential
past, "coming onto the scene", past habitual etc. It is not very easy to find
one's way around among all these types, a problem we will consider more
below. - The Lhasa section cites a number of sentences with verbal
expressions ending in "-'dug", but it is doubtful whether any of these can
be taken as real Lhasa speech: they do not occur in the Chang and Shefts
corpus, nor in Wang 1994. The Lhasa forms would be: stem + -'dug > -
zhag, stem + -ki + 'dug > .kiq (falling tone), stem + -pa + 'dug > -pa: .
Forms with 'dug only appear in negative or interrogative constructions.
The next section in Part II, on the Eastern dialects, is longer than the Lhasa
section, but we will only look at it briefly here. Verbs in these dialects have
preserved the original Present stem, "but it has taken over all functions of
the future stem." A distinctive feature of Eastern dialects is that the
nomonaliser "-pa" is replaced by "-le" (Kham) and "-n?" (Amdo), e.g. "brjod-
pa-red" (said) > "dze-leq". Zeisler mainly compares one northern Kham
dialect (Nangchenpa) and two of the Amdo group (Rebkong and
Themchen). Verbal behavior shows similarities to Lhasa: present tense
constructions are zero-marked for Tense-A, and in terms of Tense-R the
past stem is marked for anteriority, but the simple past is not commonly
seen, more limited to a perfective use.
Part III of the book, dealing with Western Tibetan, could be called the Main
Course, both from its size and from the perspective of Zeisler's personal
research background. She earlier carried out a major research project on an
oral epic transcribed in this area (Lower Ladakh), and has done on-the-
scene field work recording several Tibetan dialects in this area; she presents
us with a very thorough survey of the literature on West Tibetan dialects,
and displays an equal mastery in discussing the phonological variations
across the dialects. The West Tibetan verb shows a primary opposition
between present and past stem, but in a sense they are hardly
different "stems", since the past form has become generalised as merely
adding an "-s" to the present stem, at least for controlled action verbs. This
group of dialects is also distinguished by having widespread use of
an "Aesthetive subject": dative/locative marker with transitive accidental
event verbs, only rarely found in Lhasa Tibetan. Zeisler has good reason to
suspect this may be due to Indo-Aryan/Dardic influence. As she points
out, "Western Tibetan varieties show the most extensive differentiation of
constructions": this part of the book thus contains quite a long list of
different verbal forms, with many examples from Balti, Ladakhi and other
Finally, Part IV gives us a comparative recapitulation, for example, "in terms
of Framing....none of the four stems would be marked for the +internal
perspective." There is an extensive discussion about possible "potentalis"
implications in the imperative form. Zeisler attempts to solve the problem
about the original function of the past stem: if it were neither aspectual, nor
modal, nor temporal....then what? She revives an older theory that the "b-"
prefix associated with the past stem may derive from an early verb *ba (to
do), which leads to the possibility that the d-/g- prefix of stem form C is
not merely directional (cf. Wolfenden) but could have been derived from a
separate verb. There is a lot of additional material on different types of
complementary verbs, expressing movement, transfer, deposition etc. but
there is no space here to discuss details.
An unfortunate shortcoming at the end of the volume is the lack of any
index to the various topics discussed in the book. There is an index to the
various researchers and where they are cited discussing various broad
topics in the text, but with such a large book (nearly a thousand pages), and
so many technical terms and different grammatical topics covered, there is a
critical need to be able to locate just where such terms were first and then
later used, just where one could find information, spread across the various
dialects, on this or that topic. I would hope that Zeisler could spend some
time and generate such an index of perhaps 3 or 4 pages which could be
kept next to the back cover by the reader.
Although English does not seem to be her native language, the language
throughout the volume is clearly written and showing good mastery of
technical terms, with only the occasional peculiarity.
Nowadays a great amount of research on Tibetan is written in Chinese; I
would recommend that where possible, Zeisler consult such sources more.
For example, the large grammar of spoken Lhasa Tibetan (Wang 1994)
would have been profitably consulted for many of the discussions in
Bradley, David 1997. "Tibeto-Burman Languages and Classification" , Pacific
Linguistics Series A-86. Dept. of Linguistics, Research School of Pacific and
Asian Studies, Australian National University
Chang, Betty Shefts and Chang, Kun 1978-1981. Spoken Tibetan Texts.
Taipei : Institute of History and Philology, Academia Sinica.
Wang, Zhi-jing 1994. A Grammar of Colloquial Lhasa Tibetan [in Chinese].
Beijing: Central Minorities University.
Yu Dao-quan et al. 1980. Tibetan-Chinese (Lhasa Colloquial) Dictionary [in
Chinese and Tibetan]. Beijing: Min-zu chu-ban-she.
ABOUT THE REVIEWER
Jakob Dempsey: M.A. in Tibetan linguistics and folk-literature, Ph.D. in Asian
Linguistics (Sino-Tibetan historical phonology) - both from University of
Washington, Seattle. Since 1997 on faculty of Department of Foreign
Languages, Yuan-ze University, Taiwan. Research papers on Tibeto-Burman
and Old Chinese phonology as well as the Tibetan language. Currently
doing a funded project translating Wang 1994 into English. This nearly 600-
page grammar of Lhasa Tibetan has no equivalent in European-language
Respond to list|Read more issues|LINGUIST home page|Top of issue