|Date: Tue, 24 Jan 2006 18:06:21 +1100
From: Malcolm Ross
AUTHOR: Holton, Gary
SERIES: Languages of the World/Materials 328
PUBLISHER: Lincom GmbH
Malcolm Ross, Department of Linguistics, Research School of Pacific
and Asian Studies, The Australian National University
This slim volume (iv + 99 pp) contains a sketch grammar (70 pages),
texts (27 pages), and a bibliography (2 pages) of Tobelo, a Papuan
language of the West Papuan family. Tobelo is spoken in parts of
northern Halmahera, an island in eastern Indonesia to the northwest
of New Guinea. Like other volumes in Lincom Europa's Languages of
the World/Materials series, it offers an outline of the grammar which is
aimed at readers with typological interests. It will also appeal to the
much smaller readership of Papuanist linguists.
The work has six parts. Part 0 (pp. 1-4) is an introduction which
places the language in its geographic, social and linguistic
environment and briefly surveys previous studies. Part 1 (pp. 4-12) is
a sketch of the phonology, with notes on differences between the
dialect described and other Tobelo dialects. It provides a brief account
of the phonemes, vowel sequences, syllable structure, stress and
Part 2 (pp. 12-30) describes Tobelo's lexical categories: nominals (inc.
pronouns and proper names), verbs, property concept words, and
minor lexical categories (adverbs, numerals and a preposition). Two
points of typological interest stand out in Part 2. Pages 17-19 and 27-
28 outline respectively the subsystems of demonstratives and of
directional locative adverbs, and these provide a first glimpse of the
complex directional system which occurs in Tobelo. The second point,
made albeit briefly on pp. 22-23, is that property concept words do not
form a lexical category of adjectives, but are a collection of roots
which, depending on context, behave morphosyntactically as both
nouns and verbs (the author has written about this in Holton 1999).
Part 3 (pp. 30-48) deals with morphology. Tobelo morphology is not
particularly complex, but it is pervasive insofar as every occurrence of
every word of a major lexical category is graced by a bound
morpheme which indicates the word's lexical category, as well as
serving other functions. Part 3 thus has subsections on the
morphology of nouns and verbs and on category-independent
morphology. Every noun is marked by one of the prefixes o- and ma-
(pp 30-32: I return to this below), every verb by a subject co-
referencing prefix. If the verb is transitive, the subject prefix is followed
by an object co-referencing prefix. Interestingly, the single argument
of an intransitive verb is coreferenced by a subject prefix if the verb is
semantically dynamic, but by an object prefix if it is stative (pp 37-39).
In the latter case, the object prefix follows a third-person non-human
subject prefix, functioning as a dummy. This sequence of dummy
subject prefix and argument-coreferencing object prefix is also used
as a construction which is functionally similar to an agentless passive
(p67; see also example 290). Hence a sequence of subject prefix and
object prefix is found with transitive, stative and passive-like verbs.
The subsection on category-independent morphology covers
aspectual suffixes, negation and directional suffixes.
Part 4 (pp. 48-71) describes Tobelo syntax. There are subsections on
the noun phrase, the directional system, word order, grammatical
relations, simple and complex clauses, and discourse phenomena.
Some of these subsections serve to pull together various threads
introduced in Parts 2 and 3. This is particularly true of the subsection
on the directional system (pp 50-53), which involves demonstratives,
locative adverbs and directional suffixes but encodes a single system
of semantic contrasts involving a seaward-landward axis and an
upward-downward axis. This system is significant in that it entails
speakers specifying direction in many circumstances where other
languages would not require it. Such systems are apparently quite
widespread in languages of eastern Indonesia, however, whether
Papuan or Austronesian (see van Staden 2001, Bowden 2002).
Part 5 (pp. 71-97) comprises two texts, one narrative, the other
My evaluation of this work falls into two distinct parts: content and
With regard to content, Holton provides us with a typologically
informed account of Tobelo, and squeezes a remarkable quantity of
information into 70 pages. He is able to do this because his analysis is
thorough and allows him to make categorical statements, and because
most of the time he has a fine command of the appropriate linguistic
terms. The small criticisms below should not be seen as outweighing
the generally excellent quality of the work.
As my description of the content above hopefully indicates, Tobelo
has some features of considerable typological interest. One is the
directional system, and Holton describes this well within the obvious
constraints of such a sketch. Another interesting feature is the
morphological behaviour of property concept words. Here the author
seems to assume that he need not provide a description because he
has published on it elsewhere. This is a pity, as it leaves the reader
ignorant of when a property concept word behaves like a verb and
when like a noun. Across languages it is common enough for property
concept words to be verbs or to be nouns, but it is less usual for the
same property concept roots to display both behaviours in a single
language, and a more detailed account of this phenomenon would
thus have been appropriate. A third matter of interest is how a
speaker chooses between the noun markers o- and ma-. As far as I
can see, this question is not answered, although we are given hints.
Example 165 shows then when a generic referent is first introduced in
discourse, it is marked with o-, and on its second mention it is marked
with ma-. This is intriguing, as the example shows that the distinction
is not one of specificity or definiteness, as Holton notes on page 67.
He hints on page 66 that o- is used for first mentions and is therefore
referential indefinite, but this is evidently not the whole story, as it tells
us little about the status of ma-. We are also shown that ma- is used
instead of a possessor prefix when the possessor is non-human (pp.
Linguistic infelicities are very few. One concerns the use of an
applicative prefix to 'reference' an instrumental argument. We are told
that because the instrumental argument is not coreferenced by a
prefix on the verb, 'the object remains an oblique' (p. 42). This is an
odd statement. If the coreness of an argument is defined only by the
presence of a coreferencing prefix, then what is the function of the
applicative prefix? It is surely to render the instrument a core
argument, its coreness being marked in this instance by the
applicative prefix rather than by a coreferencing prefix.
A possible infelicity of a different kind is the assertion that certain
Tobelo features are 'contact-induced Austronesian features ' (p3).
Such statements are common in the literature, but it is impossible to
substantiate them. Holton says that among the North Halmahera
languages Tobelo is the most conservative in its retention of non-
Austronesian (i.e. Papuan) typological features. This might or might
not be true, but the assertion presupposes that the history of the West
Papuan family is well known, and it isn't. Furthermore, its relationship
(if any) with other Papuan families is unknown, so we can have little
idea which morphosyntactic features are inherited and which copied
through contact. Holton takes the presence of subject prefixes on
verbs to be 'Papuan'. He may well be right, but other specialists in the
region would disagree, as such prefixes are the norm in local
Austronesian languages. He assumes that the inclusive/exclusive
distinction in first-person plural pronouns and the directional system
are of Austronesian origin, but the inclusive/exclusive distinction is not
uncommon among Papuan languages, and directional systems like the
one in Tobelo crop up in Papuan and Austronesian languages from
eastern Indonesia eastward into western Oceania. As Austronesian-
speakers settled the region from west to east, this allows an inference
that such directional systems were present in the Papuan languages
of eastern Indonesia and were copied by Austronesian-speakers as
they moved through the region.
The presentation of the book detracts sadly from its content. The
reduction of evident A4 typesetting by 50% to an A5 page-size makes
reading a real physical effort, as the Times font is effectively only 8-
point size (the usual publisher's size is 11-point). This is compounded
by a strange typsetting glitch which frequently introduces a space
after the first character of a word. When several such gaps occur in a
line, readability is severely affected. For example, on page 39: 'As n
oted i n s ection 4.8.1 o nly referential p articipants m ay b e i ndexed
u sing t he o bjective paradigm.'
There are signs that the final copy was not proofread. How else
could 'know' appear for 'known' in the first line of the text on page 1?! I
will not list further English typos (they are many). On p. 11 -aoikoka
appears in apparent error for (I assume) -aikoka. I am left wondering
what typos occur elsewhere in the Tobelo examples.
There are just a few errors that were evidently already in the
submitted electronic manuscript. On p. 18 we are told that 'there is no
downward punctual demonstrative', but it is present in the table
immediately above this statement. It is the downward areal
demonstrative that is missing. On p. 32 we find 'proceeded'
where 'preceded' was intended.
It is a little distressing that a piece of work of pleasing quality should
be rendered so troublesome to read by poor typesetting and miniscule
print size. I have no doubt, however, that it will finish up on many
linguists' bookshelves, as it is a well-organised sketch which contains
Bowden, John (2002) Taba: Description of a South Halmahera
Austronesian language. Canberra: Pacific Linguistics.
Holton, Gary (1999) Categoriality of property words in a switch-
adjective language. Linguistic Typology 3: 341-360.
Staden, Miriam van (2000) Tidore: a linguistic description of a
language of the north Moluccas. PhD dissertation, University of
| ABOUT THE REVIEWER:
ABOUT THE REVIEWER
Malcolm Ross is a professor in the Department of Linguistics in the
Research School of Pacific and Asian Studies at the Australian
National University in Canberra. His interests include the Austronesian
and Papuan languages of the New Guinea region and the
Austronesian languages of Taiwan, with a particular focus on the
linguistic histories of these regions.