Ebook 52 Van Den Berg Muna 12-11-12
Ebook 52 Van Den Berg Muna 12-11-12
Ebook 52 Van Den Berg Muna 12-11-12
A Grammar of the
Muna Language
SIL International
2013
SIL e-Books
52
ISBN: 978-1-55671-343-9
ISSN: 1934-2470
Fair-Use Policy:
Books published in the SIL e-Books (SILEB) series are intended for scholarly research and
educational use. You may make copies of these publications for research or instructional purposes
free of charge (within fair-use guidelines) and without further permission. Republication or
commercial use of SILEB or the documents contained therein is expressly prohibited without the
written consent of the copyright holder(s).
Editor-in-Chief
Mike Cahill
Managing Editor
Bonnie Brown
Compositor
Margaret Gonzlez
2
VERHANDELINGEN
VAN HET KONINKLIJK INSTITUUT
VOOR TAAL-, LAND, EN VOLKENKUNDE
139
A GRAMMAR
OF THE MUNA LANGUAGE
1989
FORIS PUBLICATIONS
Dordrecht-Holland / Providence-U.S.A.
Published by:
Foris Publicatons Holland
P.O. Box 5093300 AM Dordrecht
The Netherlands
Research for this book was made possible by a grant from the Netherlands
Foundation for the Advancement of tropical Research (WOTRO).
Preface xii
Conventions xiv
Maps xvi
1. General introduction 1
2. Phonology 15
2.5. Stress 25
2.6. Vowel sequences 25
2.7. Phonotactics 27
2.8. Morphophonemics 32
2.8.1. -um- allomorphy 32
2.8.2. Nasal accretion 35
2.8.3. Possessive suffix -ndo/-do 36
2.8.4. Allomorphy of -Ci and -Cao 36
2.9. Adaptation of loanwords 37
4. Verbal inflection 50
Bibliography 368
Index 371
Preface
This dissertation could not have been written without the help and assistance
of many people. I would like to take this opportunity to mention some of them.
First, I am grateful to the Indonesian government, who through the Lembaga
Ilmu Pengetahuan Indonesia (LIPI) allowed my wife and myself to do research on
Muna. Prof. Dr. Husen Abas, Dean of the Fakultas Sastra, Universitas Hasanuddin
in Ujung Pandang was our sponsor. Many thanks to him and his wife for their
support, friendship and hospitality. Local government offices in Ujung Pandang,
Kendari and Raha have always smoothly arranged the necessary paper work and in
this way facilitated our stay in Indonesia. At this point I would also like to
mention my former teacher, the late Prof. Dr. J.C. Anceaux, whose enthusiasm
for linguistics was contagious and who first suggested Muna as a field of
study. I have never been sorry for having taken up his suggestion. With
gratitude and grief I recall his efforts to read and evaluate the various
chapters of this dissertation. His lively and active spirit was in strong
contrast to his failing body. His death occurred a few days after we discussed
the last chapters of the first draft.
Secondly, numerous people on Muna have contributed in one way or another to
this project. Nilus Larangka taught us the first Muna words. His enthusiasm and
good humour we will always remember. La Ode Abdul Fattah spent many long
evenings working with us, often left amazed at the intricacies of his own
language. His patience, dependability and creative language use laid a firm
basis for much that we accomplished. Hanafi was always willing to discuss the
finer points of Muna grammar and vocabulary with us; I consider him a true
colleague. He and his family, together with Mainuru Hado became real friends.
We will never forget our outings to Pulau Munante.
When I first arrived in Raha in 1984, Siddo Thamrin and his wife gave me
accomodation for two weeks. Again they provided hospitality to us in 1985 for a
short period. La Ode Bolonene and his family let us have the upper part of
their beautifully located house near the sea. The family of Wa Kandiida in
Mabodo also deserve special mention. Wa Tini and La Rianta were brave enough to
lodge curious Westerners in their house a couple of times. They cannot
overestimate the importance of those times in Mabodo to us. By refusing to
speak Indonesian to us, La Hanamu contributed much to our fluency in Muna. His
company, friendship and young coconuts made a lasting impression.
Many other people taught us or provided help, such as La Kimi Batoa (espe-
cially in January 1988), La Mokui, Lengko Umar, La Ode Dini, Sadaria, La
Aso, Wa Sukia, Mustafa and La Kuti. In addition, the following people have
helped us by providing written stories: Wa Ode Asila, Muhamed Zahab Ihu, Yos
P., Wa Ode Hanafia, La Meri and La Fona and some fifteen SPG-pupils, whose
PREFACE xiii
Praise be to God.
Conventions
isa-ku
older.sibling-my
In the Muna material, square brackets [..] mark an infix or a nasal prefix, as
in na-k[um]ala 'he will go' or na-[m]ala 'he will take'. For the active
participle circumfix (-um-/-no, me-/-no, mo-/-no) the first part is not
glossed. Example:
me-late-no
-live-A.PART
no-ko-ana
3sR-HAVE-child
In other cases the affix is simply retained (in capitals) in the English
interlinear glosses, as in:
do-si-kala-ha
3pR-SI-go-HA
The reader is referred to Chapter 10 for the meaning and use of these affixes.
As for the English translations, it is helpful to bear in mind that Muna makes
no distinction between 'he' and 'she' and that verbs are not marked for tense.
CONVENTIONS xv
/ / phonemic transcription
[ ] allophonic transcription
< > orthographic representation
( ) optional
~ in variation with
-> results in
< derived from
> more than
- morpheme boundary (in Muna)
` primary stress
secondary stress
? questionable
* ungrammatical
2 two people
The following list explains abbreviations used in the word-for-word glosses and
some other abbreviations used in the English text:
General Introduction
1.1.1. Location
Muna is one of the bigger islands located off the southeast coast of the
crab-shaped island of Sulawesi, Indonesia (see map 1). To the east of Muna
lies the better-known island of Buton (sometimes spelled Butung following
the Bugis pronunciation), separated from Muna by the rather narrow Buton
Strait. To the west lies the mountainous island of Kabaena. Between the
northwest coast of Muna and the mainland of Sulawesi a large number of small
islands are found, the Tiworo archipelago. From north to south the largest
distance on Muna is approximately 110 km, while the biggest east-west distance
is roughly 60 km.
Concerning the name given to this island, older sources and maps (before the
twentieth century) generally use the name Pangasane, Pangesana or Pantjana.
The origin and meaning of this name are unknown to me, but Pancana is also
found as the name of a village in central Buton. In a note Mller (1839-44:89)
mentions that on French maps the name 'Mounan' is found. 'Muna' is certainly
an exonym, the endonym being 'Wuna'. In the language spoken on the island
this word is used for the island (witeno Wuna), for the people (miendo Wuna)
and for the language (wamba Wuna). The word wuna means 'flower', and according
to a local legend the name was given to this coral island soon after the
first inhabitants arrived, namely, the crew of one of Sawerigading's ships that
ran aground on the coral. Sawerigading was the legendary prince of Luwu'
2 A GRAMMAR OF THE MUNA LANGUAGE
1.1.2. Ecology
Geologically Muna is a low-lying coral island (reef limestone). The area just
south and west of the capital Raha (located on the northeast coast) is hilly
(highest hill over 240 m) but in the far south some hills are higher than 400
m. Especially in the south of the island the surface is very rocky, with only a
thin layer of top soil in holes and gaps between the rocks.
In certain hilly areas the island shows several layers of terrace-shaped
coral formations, suggesting a gradual rise of the island over the centuries
(Elbert 1911:145; Van Vuuren 1920:365-366). As a result of the porous soil,
there are a number of caves in these hills. In caves about 10 km from Raha
(Liangkobori and Metanduno) prehistoric cave drawings can be seen: hunting
scenes, animals, boats and other symbols (Kosasih 1983, 1984). The northern and
eastern parts of Muna are flat. The coral is here covered with a layer of
alluvial clay.
In the south and in the east the coast rises straight up from the sea, but in
the north and the west mangrove forests of varying width form the transition
between land and sea. In the coastal area off Loghia a large number of small
inlets and rocky islands (for example Pulau Lima) are to be found. North of
Loghia the shore is low and flat, consisting of sand and mud. During low tide
large areas fall dry (kaghotia), as wide as 1 km. Coral reefs line the south
and west coast of Muna, but the Buton Strait is relatively free from coral
(Verstappen 1957). Along the west coast a number of peat swamps fringe the
shore line.
As a result of the rocky coral underground there are relatively few rivers on
Muna. Only in those areas where there is sufficient earth and clay are a number
of streams found, most of which are unsuitable for navigation. Some rivers in
the north and west can be used by small vessels, such as the Lambiko and the
Wasolangka Rivers. Since the population was primarily located in the dry and
hilly western area, water shortage was and to some extent still is a major
problem on Muna. In these drier areas one is often dependent on rain water, and
in times of drought on 'water' tapped from banana trees and bamboo stalks. A
number of springs in these dry areas are of great worth, as they can provide
whole areas with water (for example Matakidi for Lawa and Jompi for Raha).
As for flora, parts of Muna are still covered with forest, but most of the
island has been cultivated. Parts are also overgrown with elephant grass. Of
the forests, special mention needs to be made of the teakwood (Tectona grandis)
forests. Shortly after 1910, when the Dutch started to administer this area
directly, the exploitation of these teakwood forests was begun, first by the
colonial government (Dienst van het Boschwezen), later also by a private
enterprise (Vejahoma). Currently Muna makes an important contribution to the
national export of teakwood and fortunately there is also attention paid to the
environmental aspects of this exploitation. Reforestation and the allocation of
protected areas are evidence of this. Brascamp (1918) offers an interesting
account of the discovery of these forests. Rattan and certain orchid species
are also found.
1 GENERAL INTRODUCTION 3
Fauna on Muna are limited in variety compared with mainland Sulawesi. The
only endemic bigger mammals are one monkey species (macaca), deer and wild
swine. Especially the latter occur in large numbers, causing much damage to
crops. Horses and buffaloes were probably introduced from elsewhere. The
animals unique to Sulawesi (anoa and babi rusa) do not live on Muna. The cuscus
is the only marsupial. Snakes are common, but rarely poisonous. A rare
freshwater snake was discovered not long ago near Raha (Iskander 1979).
Crocodiles are now rare on Muna. Birds are found in large numbers and in many
varieties: herons, storks, hawks, several dove species, cockatoos and
parakeets. The Buton Strait between Muna and Buton is full of fish and provides
good fishing water for the coastal population.
1.1.3. Demography
The two islands Muna and Buton constitute two kabupatens (regencies), but
confusingly the boundary between these two runs across the two islands such
that the southern one-third of Muna belongs to kabupaten Buton, while the
northern half of Buton belongs to kabupaten Muna (see Map 2). Until 1987
kabupaten Muna consisted of seven kecamatans (sub-regencies) of which five are
located on Muna: Katobu, Kabawo, Lawa, Tongkuno and Tikep (= Tiworo Kepulauan).
The other two, Wakorumba and Kulisusu, are found on northern Buton. In 1987
several of these kecamatans were split up; the total number is now twelve.
The total population of kabupaten Muna was 191,366 by the end of 1985. In
order to obtain the total number of speakers of the language, several factors
have to be taken into account:
- The population of kecamatan Kulisusu (22,668) does not speak Muna, but a
Bungku dialect.
- Raha, the capital of kabupaten Muna, has a number of ethnic minorities
such as Bugis, Bajau (also in other coastal areas of Muna), Chinese and civil
servants from all over Indonesia. Altogether these probably do not number more
than 5,000.
- The kecamatans Gu and Mawasangka in southern Muna (part of kabupaten
Buton) are part of the Muna-speaking area. Their population is 28,944 (Gu) and
23,749 (Mawasangka).
- Several villages on Buton (for instance in kecamatan Batauga) are Muna-
speaking; these people probably do not exceed 10,000.
A rough estimate for the total population that speaks Muna is therefore 225,000
people.
Houses were formerly scattered among swiddens, with the exception of one
large walled settlement, Kota Muna, the political and cultural centre of Muna,
which was abandoned in the nineteenth century after an internal power struggle.
Since the Dutch started colonizing Muna early in this century, people were
forced to build roads and live in villages along these roads. The centre of
population has for a long time been in Tongkuno, the district of Kota Muna, in
the dry and hilly eastern part of the island. The west and the north were
sparsely populated, probably because in these low-lying lands people were much
more susceptible to malaria than in the dry and higher east.
It is striking that this island population has never been a seafaring nation
and has generally avoided living along the coast. The socio-political situation
in the past probably accounts for this. Pirates were very common in this
part of the archipelago up to the nineteenth century and Muna people were
often taken away as slaves to such places as Makassar. It is perhaps not a
coincidence that the words for 'sea' and 'afraid' are homophonous in Muna
4 A GRAMMAR OF THE MUNA LANGUAGE
(tehi). An exception must be made for the village of Loghia, located on the
east coast and for centuries Muna's most important window on the outer world.
During this century there have been major changes in the settlement patterns.
Firstly, the choice of Raha as the capital of the local colonial government has
caused the population centre to shift from Tongkuno to Katobu. Raha is now a
small provincial town with a population of approximately 16,000. Secondly, in
the 1960s several villages from the driest areas in Tongkuno were transferred
to the northeast, where living conditions are more favourable. As a result
Tongkuno is now the least populated area on the island.
Living conditions and material culture are relatively simple. The staple food
is maize, grown in small fields sometimes a considerable distance from the
homes. Maize is supplemented by roots and vegetables and, in coastal areas,
fish. Dry-rice agriculture is found in some parts, but generally the area is
too dry for rice. Cash crops such as cashew and cacoa trees are increasingly
popular.
Traditional houses are built on piles, with woven bamboo walls and roofs of
dried grass. The building of wooden or even brick houses is encouraged by the
government, but such houses are still seen as a sign of wealth.
Art seems to be mainly functional. Weaving of sarongs is common and beautiful
patterns can often be seen. Certain villages specialize in producing earthen
jugs, but these are not decorated.
According to Couvreur (1935), social strata were still quite distinct on Muna
in the 1930s, but nowadays this division is increasingly felt to be
incompatible with modern life. Traditionally there were four classes: kaumu are
the high nobility, walaka the lower nobility, maradika the commoners (who in
turn are subdivided into three classes) and finally the slaves (ghata).
Each class had its own rights and prerogatives, relating to marriage, clothing,
ornaments and so on. Only the nobility were entitled to bear the title La Ode
(men) or Wa Ode (women) before their names.
These classes were also determinants in language use. Social deixis, for
example, was an important parameter in the choice of the different terms of
reference for 'wife'. A kaumu would use the term isamiu (lit. 'your (polite)
older sibling') to refer to his wife when speaking to somebody from a lower
class, whereas the lower class would use aimiu (lit. 'your (polite) younger
sibling') to the higher class. Presently such differences are tending to fall
into disuse.
1.1.6. Religion
The vast majority of Muna people are Muslim (97%), although pre-Islamic beliefs
and practices are still widespread. Islam entered Muna probably through Buton,
which was itself converted to Islam by the sultan of Ternate in the late
sixteenth century.
As a result of the work of Catholic missionaries starting in the 1930s, a few
villages in the south are predominantly Catholic (Wale-ale, Lakapera, Lawama,
Lolibu). The Catholic church has done much work in the area of community
development, schooling and health care. They were the first to relocate a poor
village from the dry south to a more fertile area (see Mingneau 1974).
Unfortunately no language material was published, although some of the mission-
1 GENERAL INTRODUCTION 5
aries had a good command of the language and parts of the New Testament have
been translated in draft form.
In addition to a Catholic church there are two small Protestant churches in
Raha, the members of which are almost exclusively non-Muna.
1.1.7. History
It is not yet clear where the Muna people came from and how long the island has
been inhabited. The greatest dialectal variation is found in the southeast
(Gu), suggesting a population movement from south to north. Also, no datings of
the cave drawings mentioned in 1.1.2 have yet been published
In 1.1.1 a legend was mentioned that the first people to arrive on Muna were
Sawerigading, the prince of Luwu, and his crew. The place where his ship is
said to have run aground on the coral is a hill near the village of Bahutara
in central Muna.
Another important legend concerning the origins of Muna is that of Bheteno ne
Tombula, the first king of Muna, who appeared from a bamboo stem and who
married Tandiabhe, the daughter of the king of Luwu who had been expelled from
her native country and reached Muna by sea. This legend is to be found in the
Appendix (Text 4).
It seems likely that the first inhabitants of the island were hunters and
gatherers. The old times are called dhamani momaano karaka 'the time of the
frog-eaters'. According to one story it was the Portuguese (miendo Porotugisi)
who introduced maize (kahitela < Castilla) to the island.
For a long time the people lived close together, but the growth of the
population led to the expansion of the people into four villages and later four
districts (ghoera). Gradually an intricate system of political organization
emerged, probably modelled after Buton, which developed into a politically
important area in the sixteenth and seventeenth centuries (Vonk 1937). At the
head of the Muna people stood the Sarano Wuna ('Muna council'), led by the
chosen omputo 'king, ruler' (lit. 'our lord'). Other members included
functionaries such as four district leaders (ghoerano), one bhonto bhalano, one
mintarano bitara and two kapitalao. Disputes and quarrels over succession were
not uncommon and in such times the sultan of Buton often played a decisive
role. The relationship between Buton and Muna after 1664 (when Ternate formally
gave up its claim to Muna) is usually characterized as that between an older
and a younger brother. Buton considered Muna to be one of its four autonomous
provinces (barata), which surrounded the sultanate. In Muna its independence of
Buton is usually stressed, but it is a fact that Muna has always stood in the
shadow of its 'big brother', whose political and cultural superiority were
obvious to all concerned. The influence of Buton in the areas of religion,
socio-political organization and language (borrowing) can probably not be
overestimated.
In the first decade of this century the Dutch began to administer the area
directly. Before that time, contacts with Europeans seem to have been minimal,
but thereafter a permanent military representation of the colonial government
was present in Raha, which, for ease of access, was chosen as the new capital.
After 1910 there followed a period of rapid changes, including the abolition of
the Sarano Wuna and the introduction of the system of forced labour (harendesi)
for the building of roads and the subsequent new village patterns. Schools were
introduced, a small hospital was built, run by the Catholic mission, and the
exploitation of the teak forests began. With the exception of the war years,
the Dutch ruled Muna until 1949, after which the island took its place in the
Indonesian Republic.
6 A GRAMMAR OF THE MUNA LANGUAGE
In Van den Berg (1988), I have given an overview of the language situation on
Muna and parts of Buton. Some of the information concerning the language
boundaries and the dialect situation of Muna is repeated below.
The dialect that is most widely spoken on Muna and enjoys the highest
prestige is found in the central and northern part of the island. This dialect
I call 'standard Muna'. Standard Muna is spoken with remarkably little dia-
lectal variation in a large area on Muna comprising the four kecamatans that
made up the old Muna kingdom: Tongkuno, Kabawo, Lawa and Katobu. It is also the
language of Tobea Besar, an island between Muna and the mainland of Southeast
Sulawesi. In Sneddon (1982) Tobea Besar is mistakenly assigned to the Tolaki-
speaking area. The Muna population on Tobea Besar consists of recent immigrants
from central Muna. There is also a Bajau settlement; Bajau are also found on
Tobea Kecil and Renda, two islands close to Tobea Besar.
The eastern boundary of Muna is on northern Buton. All along the western
coast of Buton (kecamatan Wakorumba) there are Muna settlements, some of which
are said to have been founded by people from Loghia, a fishing village on
Muna's east coast, a few miles south of the capital Raha. They all speak
standard Muna. It is unclear whether the southern boundary of these settlements
coincides with the kecamatan boundary between Wakorumba and Kapontori. There
is one village on the Wakorumba coast, Maligano, northeast of Raha, where the
1 GENERAL INTRODUCTION 7
original population speaks Taluki, a Bungku isolect which is about 75% cognate
with Kulisusu, the closest Bungku language. To my knowledge, the existence of
this isolect has never been reported.
In northwest Muna a different dialect is found in kecamatan Tikep (Tiworo
Kepulauan) on the Tiworo islands and around Kambara. The island population in
the Tiworo archipelago is a mixture of Bajau, Bugis, Muna and people from
Kadatua (an island southwest of Baubau). A number of islands are uninhabited.
On the mainland of Muna the dialect boundary between Tiworo and standard Muna
runs through Marobea (formerly Dandila), east of Kambara.
Moving south we encounter a very important bundle of isoglosses which roughly
coincides with an administrative boundary. The two southernmost kecamatans of
Muna, Gu and Mawasangka, belong to the southern group of Muna dialects.
Following Indonesian usage I will use the term Gumas to refer to these
dialects. Certain villages, however, are exceptional in that they are located
in the 'wrong' area. Tampunawou in northwest Mawasangka is standard-Muna-
speaking, whereas a few villages in southern Tongkuno, such as Wale-ale and
Lawama, speak a Gumas dialect. The present dialect boundary runs right through
Tongkuno's capital Wakuru, and causes dialectal differences between
generations. My impression is that the boundary is moving south, due to the
influence of standard Muna in Wakuru. Gu and Mawasangka are occupied by a
number of closely related subdialects. Gu shows the biggest number of dialectal
differences in the whole Muna-speaking area. The village of Tolandona, located
on the Buton Strait, is Wolio-speaking.
Muna dialects are also found on two small islands southwest of Baubau,
Kadatua and Siompu. Kadatua has a Gumas subdialect, but Siompu shows a
different southern dialect. On Siompu another language is also found,
Kaimbulawa. The middle island, Liwutongkidi ('small island'), is uninhabited.
Finally, a number of Muna dialects are spoken on the west coast of Buton,
just north and south of Baubau. North of Baubau there are the fishing villages
of Lowu-Lowu and Kalia-lia with slightly different Gumas dialects. South of
Baubau we find Katobengke, Topa, Sulaa and Lawela, all speaking Gumas dialects.
Moving further south, we first run into a separate language, Busoa, after which
follow Laompo (opposite Siompu) and Burukene. These two closely related
dialects are also part of the Gumas dialect complex.
Muna speakers have also moved to other areas in Sulawesi, such as the Telaga
islands off Kabaena, Kendari and Ujung Pandang. In Ambon there is also a group
of Muna speakers (from Batu Sori).
Our present knowledge of the Muna dialect situation allows the following
estimates of population figures.
On Map 2 the boundaries of the Muna language and the dialects are indicated.
The reasons for distinguishing these dialects (and not more) is partly
because of lexico-statistics: these speech varieties all share more than 80% of
their basic vocabulary. Secondly, there are important phonological differences
between these dialects, summarized in the following chart, in which <'> marks a
glottal stop:
8 A GRAMMAR OF THE MUNA LANGUAGE
Standard Muna | gh | h | r |
Tiworo | h | h | r |
Siompu | ' | ' | r |
Gumas | ' | ' | h |
Examples:
1.2.3. Subgrouping
Unlike Wolio, the official language of the sultanate on Buton, Muna has never
known a writing system. Written literature is therefore of very recent date and
almost exclusively of oral origin. As in other traditional cultures in
Indonesia, the oral literature comprises a wide range of forms and subject
matter. Prose stories include legends, folktales, fables and romances. Under
the heading of poetry can be reckoned such forms as kabhanti and kantola.
Pogala is a kind of traditional drama (Lanayo 1979). A good anthology of Muna
oral literature would be a very fruitful and worthwhile project.
It is generally felt that the influence of modern life and the mass media has
a negative effect on the creation and maintenance of these traditional forms.
Yet some attempts can be seen to counter this movement, such as the release of
a cassette a few years ago with traditional and new songs in Muna ('Wuna
liwuku').
1 GENERAL INTRODUCTION 9
As for language use, quite a sharp distinction needs to be drawn between the
capital Raha and the villages in the interior. In the villages Muna is
dominant, Indonesian being mainly restricted to the schools. Everybody is
fluent in Muna, but presumably a large percentage do not speak Indonesian. In
Raha this situation is reversed. Indonesian is dominant, even in families where
both parents are from Muna. The percentage of children born in Raha of Muna
parents but not fluent in Muna is probably quite high.
Foreigners living in Raha (Chinese, Bugis, Javanese) very rarely learn to
speak Muna, a fact which the Muna people themselves attribute to the difficulty
of their language. While Muna is certainly not an easy language to learn
(mainly because of its complex inflectional morphology), the socio-linguistic
situation in Raha is probably an equally important factor.
All over Muna education is in Indonesian, with Muna used as the language of
instruction in the lower forms. No educational material is available in Muna,
but at present a 'Muna language team' is trying to change this situation by
producing material in the local language. There are also plans for a dictionary
and an anthology of Muna prose and poetry.
1.3.1. Fieldwork
My first visit to Muna was in August of 1984 and was meant to be a fact-finding
trip. No real research was done, although I gathered some language data. The
main purpose was to make contacts at various levels, find suitable language
helpers and a place to live. Thanks to the kind cooperation of many people,
this visit proved a success.
In July 1985 my wife and I arrived in Muna and stayed there till November
1986 for an extensive period of fieldwork. The first four or five months were
spent acquiring a proficiency of the spoken language, although Indonesian
continued to be indispensable for discussing specific grammatical and lexical
points. We rented a house in Raha and lived there, but in addition we made
frequent tours to villages in the interior. During these first months the basic
pho-nology and the inflectional system were analysed. From January 1986 onwards
the emphasis shifted to the analysis of derivational morphology and syntax. In
order to achieve this it was crucial to have a good corpus of texts. Ultimately
we had some 75 texts of varying lengths at our disposal (approximately 150
typed pages). We acquired these texts in the following ways:
1. Record a spoken text and transcribe it or have someone else transcribe it;
2. Ask literate people to write a text;
3. Cull texts from published and non-published sources. The provincial office
for education and culture in Kendari has collected a number of Muna folk
stories. These unpublished stencilled collections contained much valuable
material.
Every text was carefully checked with language helpers for the meaning and the
use of particular structures and words. Then lists and charts were made for
various phenomena (for example, for negation, for each affix, for each
demonstrative and so on), on the basis of which preliminary conclusions were
drawn which were further checked against new material, corrected, supplemented
and so on. In addition a dictionary file was set up, eventually containing some
2,500 root entries.
On our return to the Netherlands the actual writing started, followed by a
short fieldtrip to Muna in January and February of 1988, in which the chapters
10 A GRAMMAR OF THE MUNA LANGUAGE
written thus far were checked and more information was collected on obscure or
difficult points.
This grammatical description of Muna is based on the speech variety in
kecamatan Katobu (to which Raha belongs), as this was the area where we lived.
Our main language helpers, Laode Abdul Fattah and Hanafi BA, were both natives
of Katobu (born in Loghia and Watuputi respectively). The speech in Katobu is a
subdialect of standard Muna, and where it deviates from subdialects spoken in
Tongkuno, Kabawo and Lawa, I have tried to indicate this.
1.3.2. Framework
Phonology
This chapter describes the sound system of the Muna language. The emphasis is
on the basic sound structures of the language, such as phonemes, syllable
structure, stress and phonotactics.
Two observations must be made concerning the scope of the description given
here. First, the phonetic description in 2.1 of the consonants and vowels is
necessarily impressionistic. The unavailability of Muna speakers in the
Netherlands and relatively simple recording techniques made high quality
acoustic descriptions impossible.
Second, this description is conservative in that no attempt is made to relate
the facts to current phonological theories. Hence one will not find formalized
rules or analyses of the facts in the words and notations of a particular
school of phonology (for instance, distinctive feature matrix, autosegmental
tiers). What is presented here is meant to be an adequate description of the
relevant facts.
In 2.1 the phoneme inventory of Muna is given, with subsequent phonetic
descriptions and examples. In addition to a list of words showing phonemic
oppositions (2.2), a table with phoneme frequencies is presented (2.3). In 2.4
syllable and root structure is discussed, followed by a treatment of stress
(2.5) and vowel sequences (2.6). In 2.7 the phonotactics of the simple root are
discussed, followed by a section on morphophonemics (2.8). Finally, 2.9
illustrates the phonological adaptations of loanwords.
a. Consonants
vd : voiced
vl : voiceless
prenas : prenasalized
16 A GRAMMAR OF THE MUNA LANGUAGE
Notes 1. The palatal consonants /c/, /j/ and /y/ are marginal loan
phonemes. The number of words containing these recent loan
phonemes is very low. Consequently, they will not be included
in the discussion of consonants.
2. /t/ and /nt/ are placed here with the alveolar /d/, although
they are phonetically apico-dentals. The difference between
/t/, /nt/ and the lamino-dental //, however, is much bigger,
hence their inclusion with the alveolars.
b. Vowels
In this section a phonetic description of all the consonants and vowels will be
given with illustrative examples. In these examples the phoneme under
discussion will be illustrated both in initial and in medial position within
the word. I will give the phonemes in phonemic transcription (marked by //),
2 PHONOLOGY 17
Consonants
paso 'nail'
dopi 'board'
buku 'bone'
bebe 'duck'
mpona 'long'
sampu 'go down'
mbololo 'gong'
lambu 'house'
These four bilabial plosives have flapped or trilled allophonic variants before
/u/ in stressed position. When the closure of the plosive is released, the lips
are vibrated once or twice fairly lightly while the lower jaw is moved forward
a little. This allophone only occurs in less careful speech and is considered
at best substandard. Since the International Phonetic Alphabet does not suggest
a symbol for this variation, I use the diacritic * to symbolize this 'rolling
release':
bhaguli 'marble'
tobho 'stab'
mata 'eye'
ama 'father'
/w/ <w> voiced labial approximant. The following are common allophones:
Before /u/ there is therefore free variation between [w] and []. This
whole system of allophones is quite unstable within the speech community.
With several speakers I recorded only the first or the second allophone.
The degree of lip protrusion in the second and third allophone also
varies, as does the amount of friction to be heard in the third
allophone. Whatever the phonetic norm, it is very clearly not a rounded
bilabial like the English /w/. The target seems to be a labio-dental or a
bilabial spread approximant, with lip protrusion and friction resulting
from the following rounded back vowels // and /u/.
foo 'mango'
mafu 'yam'
dhangku 'beard'
medha 'table'
tombi 'flag'
late 'live'
dada 'cook'
madaho 'later'
2 PHONOLOGY 19
/d/ may be realized as an implosive []. This is often the case before
/a/, regularly before /i/ and //, but rarely before // and never
before /u/. Before /a/ and // the plosive is sometimes made with a post-
alveolar or even a retroflex point of articulation:
ntabhala 'thumb'
lantoga 'bracelet'
ndau 'heron'
panda 'bottom'
nea 'name'
lani 'sky'
saho 'rafter'
lasa 'ten thousand'
nsara 'long'
nunsu 'beak, bill'
rase 'rust'
horo 'fly'
lembi 'naked'
ala 'take'
kori 'leprosy'
bhake 'fruit'
This sound has the usual allophones before front and back vowels: a
palato-velar [k] before /i/ and //, and a post-velar [k] before // and
/u/. Before /a/, /k/ is a regular velar plosive.
20 A GRAMMAR OF THE MUNA LANGUAGE
ganda 'drum'
laga 'tree ant'
ngkodau 'old'
rangko 'young'
nggela 'clean'
sanggara 'fried banana'
nginda 'shine'
pangaowa 'sail'
ghuse 'rain'
ragha 'branch'
hende 'rise'
maho 'near'
Vowels
ana 'child'
kala 'go'
elu 'saliva'
bhose 'shoulder'
2 PHONOLOGY 21
oe [oe] 'water'
wou [wou] k.o. tree
peo [peo] 'extinguished'
This section presents a list of minimal pairs showing some of the more
important contrasts establishing the phonemes in 2.1. The reasons for treating
the prenasalized segments as unit phonemes will be discussed in 2.4. I have
tried to limit the contrast to initial position in disyllabic roots, but in
some cases no good examples were available, either because they did not happen
to occur in the data gathered, or because of distributional limitations (ng and
ngg, for instance, are very rare in initial position). In such cases contrast
is shown in other environments. In one case (ngk - ngg) the environment is not
completely identical.
In a corpus of over 1,000 disyllabic roots (CVCV type, see 2.4) the frequency
(in percent) of the consonant phonemes in word-initial and word-medial position
is as follows (because of rounding off, the total is not exactly 100):
penultimate final
i 11 17
e 14 14
a 35 32
o 22 18
u 18 19
The structure of the syllable depends crucially on the analysis of the pre-
nasalized segments mp, mb, nt, nd, ngk, ngg and ns. Are these phonetically
complex sound units or are they sequences? And what is the syllable division of
words like lambu 'house'?
The analysis adopted here treats these segments as unit phonemes in all
positions in view of the following considerations:
lambu 'house'
ka-lambu-lambu 'small house'
pulangku 'staircase'
ka-pula-pulangku 'small staircase'
la.mbu 'house'
pu.la.ngku 'staircase'
From the second example it is clear that the sequence au in ngkodau is split up
since it belongs to two different syllables, the final u constituting a
syllable in its own right.
Some examples of syllable divisions:
Roots display various combinations of syllable types. The following chart shows
the root structures that are most common in the language. Although most of
these forms are plain roots, some roots of more than two syllables may have
frozen morphology. Based on a sample of approximately 2,500 roots, the relative
frequency of root types is given, but structures that occur less than five
times in the corpus have been ignored (0.1%).
The following conclusions can be drawn concerning the structure of the root:
1. The majority of roots are disyllabic, although trisyllabic roots are also
common. Roots consisting of one, four or five syllables are marginal
(altogether less than 10%).
2. CV syllables are much more common than V syllables.
3. V syllables tend to occur in root-final position.
4. Trisyllabic roots contain at least one CV syllable (there are no V.V.V
roots).
5. Sequences of three vowels are very uncommon in roots (0.7%).
2.5. Stress
`ani 'bee'
`bhosu 'water jug'
ka`bhongka 'main road'
ghunde`letu 'worm'
netula`tula 'he tells a story'
netulatu`lamo 'he has told a story'
In sequences of two vowels, all five vowels are found as either the first or
the second element. As noted in 2.4.1 the second vowel constitutes a separate
syllable. Examples:
26 A GRAMMAR OF THE MUNA LANGUAGE
bhei 'rotten'
bhee 'crazy'
bhea 'betel'
feo 'squeeze'
deu 'needle'
ghai 'coconut'
ghae 'cry'
gaa 'marry'
dhao 'learn to stand'
ghau 'lung'
doi 'money'
oe 'water'
bhoa 'hawk'
foo 'mango'
wou k.o.tree
pui 'coccyx'
ghue 'rattan'
dua 'two'
suo 'bamboo trap'
tuu 'knee'
When the two adjacent vowels are different, stress is regular and falls on the
penult. When the two vowels are identical, however, the resulting sequence is
phonetically one long vocoid. Examples:
In other words, the final unstressed vowel is absorbed into the preceding
identical vowel. This accounts for the fact that when Muna people do write
their language, they tend to write word-final sequences of like vowels as one
vowel (for example fuma for fumaa 'eat').
The disyllabic character of the sequence reappears when it occurs in non-
final position because of affixation (for example fumaa-mo, nee-ku), when the
phonetically long vocoid is stressed on its second part. In such cases the
sequence is always written as a double vowel.
As noted in 2.4.2, sequences of three vowels are quite uncommon. The follow-
ing were recorded in roots (in derivations there are many more possibilities):
kaue 'swing'
siua 'nine'
soua 'pimple'
waea 'bat'
`bhuia `siua
`daoa `soua
`kaeo `waea
`kaue
bhe`au
bhoko`eo
bu`ea
The rules governing these phenomena are still unclear. Notice that a word like
siua 'nine' cannot be reinterpreted or analysed as *siwa, because of the clear
phonetic differences between these sounds: w does not show any lip rounding
before a, whereas u is a fully rounded vowel.
Finally, there are two roots with sequences of three like vowels, although
the first syllable is in both cases probably a frozen morpheme. In these
examples a glottal stop occurs between the first and the second vowel:
This glottal stop, which is non-phonemic, also occurs optionally between the
first and the second vowel in a sequence of three vowels in complex words:
2.7. Phonotactics
C1/C2 p b mp mb bh f t d dh nt nd s
p v - v - - - v v v v v v
b - v - v - - v v v v v v
mp - - - - - - - v - - - v
mb - - - - - - v - (v) - - -
bh - - - - v - v v (v) v (v) v
f v - - - - v v (v) - - v -
t v - v v v v v v v v v v
d v - v v - - v v - - v (v)
dh - (v) v v - - - - v - (v) v
nt (v) - - - - - - (v) - - - -
nd - - - - - - - - - - - -
s v v v v v (v) - v v v v v
ns - - - - - - - - - - - -
k v v v v v - v v (v) v v v
g - (v) v v - - v v v v v v
ngk - (v) - - - - - - - - - -
ngg - - - - - - - - - - - (v)
gh - v v v v v v v - v v v
h (v) - - v (v) - - v (v) (v) v -
l v v v v v - v v - v v v
r v v v v - v v v - v v v
w - - - v (v) - v - (v) v (v) v
m - - - (v) - (v) v v v v v v
n v - - - (v) (v) - - - - - (v)
ng - - - - - - - - (v) - (v) -
2 PHONOLOGY 29
p..b.. p..bh..
bh..p.. b..p..
k..g.. g..k..
This constraint does not apply to t and d, which is evidence of the fact
that these sounds are not homorganic (apico-dental versus apico-
alveolar).
Closer scrutiny reveals that the fricative f and the approximant w share
this non-occurrence with contra-voiced bilabial plosives. There are no
examples of the following patterns:
f..b.. f..bh..
b..f.. bh..f..
w..p.. w..f..
p..w.. f..w..
For the labials the rule can therefore be extended to exclude all co-
occurrences of contra-voiced labial obstruents (plosives, fricative and
approximant).
The behaviour of the uvular fricative gh shows that combinations with the
velar plosives do not occur:
gh..k.. k..gh..
gh..g.. g..gh..
This may mean that phonologically these sounds are homorganic and hence
the velar area falls under the same generalization as the labial area. In
that case the non-occurrence of g and gh could be accidental.
p..mb.. mb..p..
b..mp.. mp..b..
bh..mp.. mp..bh..
f..mb.. mb..f..
w..mp.. mp..w..
k..ngg.. ngg..k..
g..ngk.. ngk..g..
might mean that this rule does not hold for velar obstruents, or,
alternatively, that gh is voiceless at some level of its representation
or was voiceless in the past. The latter possibility is most plausible in
view of the fact that Muna gh reflects PAN *q, possibly a voiceless
uvular plosive (see Van den Berg 1988).
mp..mb.. mb..mp..
ns..ngk.. ngk..ns.. etc.
l..r.. r..l..
p..ng.. gh..p..
s..t.. l..f..
w..d.. m..gh..
Consonant-vowel constraints
in combination with vowels. Among all the root types there is only one
exception to this rule, the word bhuia 'card game', possibly a loan from Wolio.
What the reason is for this constraint is not clear, but possibly there is a
correlation between the 'backness' of u and the frontness preferences of
implosives. It is typologically well known that bilabial implosives are much
more frequent than alveolar implosives, while velar implosives are very rare.
The backness of u may function as a kind of block to the implosion of the
previous consonant. Alternatively, the lip rounding of u may be of some
influence in the process.
Other non-existent combinations involve low-frequency consonants (nse, ngu)
and are best considered accidental.
2.8. Morphophonemics
The infix -um- serves to form the irrealis of certain verbs, and, in com-
bination with the suffix -no, the active participle of those verbs (see 4.5 for
meaning and usage). This infix has a number of allomorphs which are given in
order of importance, first in combination with roots, then with prefixes and
finally with reduplication.
baru 'happy'
bhala 'big'
manda 'repent'
nale 'soft, weak'
mbolaku 'steal'
ndiwawa 'yawn'
realis irrealis
The last example is irregular in that the regular form (*a-[m]omaa) does not
exist (or no longer exists). For all three examples the subject prefixes have
been reinterpreted as belonging to the irrealis ao-class (see 4.5).
When roots are affixed with verbal prefixes (that is, when the resulting
derivation is a verb) the behaviour of -um- is not predictable, but has to
be specified for each prefix. Three cases can be distinguished (for meaning
and usage of these prefixes, see chapter 10), of which the first two are
illustrated by two examples each.
a. The prefix is amenable to -um-: the normal allophonic rules for roots
apply. This is the case with the following prefixes:
34 A GRAMMAR OF THE MUNA LANGUAGE
b. The prefix is resistant to -um-: the rules do not apply. This is true for
the majority of prefixes:
c. The prefix is blocked for -um-: the rules cannot apply because of the
shape of the prefix (initial nasal or prenasalized consonant). Examples
are the following prefixes:
mansi- mpo-
manso- ngko-
mba- nsa-
These facts can be reanalysed in terms of weak and strong morpheme boun-
daries before a given prefix, such that a certain prefix is specified for
its morphological boundary, a weak one allowing for -um- allomorphy, while a
strong boundary is resistant to such rules.
With fully reduplicated roots (that is, repetition of the first two syl-
lables of the root) only the reduplicated part is affixed with -um- while
the base remains unchanged. In these examples I ignore inflection and
meaning (for reduplication see 10.3).
This is true for almost all cases, except when the base has an initial p or
f. In these cases, both the reduplicated part and the base undergo nasal
substitution:
With partial reduplication (that is, only the first syllable of the base is
repeated) both regular um-infixation and blocking occur as variants,
although infixation seems to be more common:
p mp
f mp (mf)
b, bh, w mb
t nt
d nd
s ns
k, gh ngk
g ngg
Of all the plosives, only dh (a loan phoneme) does not participate in this
process. The implosive bh loses its implosive character, while the continuants
f and w are occlusivized to mp and mb respectively. The prenasalized counter-
part of gh is ngk, which is remarkable in that there is a difference in place
of articulation (uvular versus velar) as well as in voicing. This supports the
earlier observation that gh was possibly a voiceless plosive at an earlier
stage.
As for f, in a very limited number of words f can have the prenasalized
variant mf in addition to mp, mostly after the prefix para-. Since mf is not an
independently established phoneme and clusters do not occur in the language,
this poses a problem. The best solution seems to be to regard mf as a juncture
allophone of mp, that is, an allophone which is only found at a special
morpheme boundary.
Nasal accretion occurs in the following three cases:
The regular third person plural possessive suffix -ndo 'their' (see 5.4.1) has
a variant -do which is found with bases containing a prenasalized consonant. In
some high-frequency words the variant with -do is the only possible one, but in
many cases doublet forms are attested:
Although strictly speaking the following problems are not related to morpho-
phonemics, this seems the best place to discuss them.
The suffixes -Ci (transitivizing, often a locative or petative meaning, see
10.2.16) and -Cao (intensifier, see 10.2.15) each have a number of allomorphs.
In the case of -Ci the following consonants can take the C position:
p, t, k, f, s, gh, h, m, n, ng, l, r, w
2 PHONOLOGY 37
(that is, all the voiceless plosives, the fricatives, nasals, liquids and the
approximant). These consonants thus seem to form a natural class over against
the remaining consonants (voiced plosives, the implosive and the prenasalized
consonants).
To illustrate this suffix I give two examples (for further information see
10.2.16):
In the case of the suffix -Cao the C position can be occupied by one of the
following consonants:
p, t, f, s, h, l, r
This is a smaller set, also excluding the nasals, the velar/uvular k and gh and
w. For examples and usage of -Cao see 10.2.15.
As for the status of these consonants in the derivation, a possible solution
is to treat them as underlying final consonants of the root, which are deleted
in every environment except before the suffixes -i and -ao. Positing underlying
final consonants however, violates an exceptionless rule in the (surface)
phonology of the language, namely that there are no syllable-final consonants.
Also, one runs into problems with verbs that take more than one suffix. I
therefore simply choose to describe the different allomorphs as bound
allomorphs for which the verb has to be specified in the lexicon.
For a discussion of a similar case of allomorphy in Maori, see Hale (1973).
Like all other languages, Muna has borrowed and still borrows a large part of
its vocabulary. The source languages are Malay/Indonesian, Bugis, Wolio and
(formerly) Dutch. In this section the main phonological adaptations loanwords
have undergone will be discussed. In many cases there is a known corresponding
lexeme in Malay and hence I take the source language to be Malay (possibly
through Bugis or Wolio), although in many cases the words are ultimately from
Sanskrit, Arabic, Dutch and other languages.
Muna Malay/Indonesian
The existence of the word dosa 'guilt, debt' next to dhosa 'sin' suggests
an interesting but complicated history of borrowing and semantic change
of these words.
But more usual is the addition of a vowel. The conditions specifying when
the final consonant is deleted and when a vowel is added and which vowel
that must be, cannot be given at present.
3. Breaking of clusters
This chapter will form a prelude to the syntax and morphology chapters of this
Muna grammar. I will first discuss the difference between word, affix and
clitic, while the rest of the chapter is concerned with the division of words
into word classes and a succinct discussion of these classes. A number of
issues relating to the area of word classes will conclude this chapter.
3.1.1. Words
In this section I will define some concepts which will be used throughout this
grammar. Based on Reichling (1935) the word can be defined as a unit of
phonological and morphological constancy and syntagmatic mobility. In other
words, in a word such as fotu 'head', the four phonemes are in a given fixed
order and cannot be changed without altering the meaning or resulting in a non-
existent word. Morphological constancy is shown by the fact that in a word such
as no-feka-nggela-hi-e-mo 'she has already made it clean', the order of the
morphemes is fixed. Given these six morphemes, this is the only possible
ordering. Syntagmatic mobility can be shown by replacement, movement, separa-
bility and isolation. In combination these tests will show the division between
words, groups of words and affixes. Take for example the following two
sentences, which are almost identical in meaning (although there is a dif-
ference in emphasis):
The fact that the two constituents of these two clauses can be moved around
shows that they are separate words.
42 A GRAMMAR OF THE MUNA LANGUAGE
3.1.2. Affixes
ndawu 'fall'
fo-ndawu 'drop'
tomba 'basket'
ka-tomba-tomba 'small basket'
Regularity, productivity and change of word class are not the distinguishing
parameters in this definition. Rather, inflection covers the traditional
category of verbal conjugation, which in this case is extended to 'nominal
conjugation'.
Clitics take a position midway between words and affixes. They constitute
meaningful elements that differ from affixes in, for example, their position
and their behaviour with respect to stress rules and other phenomena. In Muna
there exists only one clitic element, the pausal clitic -a (for a full dis-
cussion of meaning and usage of this clitic, see 9.22). The reasons for calling
3 WORDS AND WORD CLASSES 43
`lima 'arm'
li`maku 'my arm'
lima`kumo 'my arm' (emphatic)
lima`kumoa 'my arm' (emphatic pausal)
Grammatically it differs from affixes in that it is not added to the word, but
rather to the phrase (for instance after negators). Also, its meaning as a
'pausal marker' is markedly different from the kind of meaning derivational
affixes usually have. In the illustrative sentences in this grammar, -a will
simply be glossed as 'CL' (clitic).
Cliticization is a phonological process whereby an independent word is
phonologically attached to the preceding or following word. Two such processes
occur in Muna, whereby a monosyllabic word is pre-cliticized to the following
verb and loses its vowel:
The forms so and pa in the first column are full words. Their distribution is
parallel with other prepositions (Chapter 6) and negators (8.5) that do not
cliticize. In the second column, however, these elements have become clitics,
as they are part of the following word.
This phenomenon of vowel merging is also found with the prefixes sa- and ta-
(see 10.2.44 and 10.2.48).
Any systematic grammar of a language will have to define the parts of speech or
word classes that are relevant for that language. We cannot work on the
assumption that the traditional division into parts of speech is universal; a
system has to be set up which is based on the language under investigation. It
turns out, for example, that the category 'adjective' cannot be maintained for
Muna, as there is no independent criterion by which supposed adjectives can be
satisfactorily defined apart from verbs.
Word classes are distinguished on two bases: inflectional/derivational
possibilities and syntactic properties. I distinguish between declinable and
nondeclinable words, and then arrive at morphological classes dividing the
declinable words into noun, verb, numeral and pronoun. Syntactic criteria such
as substitution, expansion and function divide the nondeclinable words into
several classes.
44 A GRAMMAR OF THE MUNA LANGUAGE
For Muna I propose the following ten word classes, of which several will have
subdivisions:
1. Noun
2. Verb
3. Pronoun
4. Numeral
5. Quantifier
6. Adverb
7. Preposition
8. Conjunction
9. Particle
10. Interjection
There is a primary division between noun and verb on the one hand and all the
other word classes on the other hand. Noun and verb are open classes comprising
several thousands of members and continually expanding because of borrowing.
The other word classes are all closed classes, the members of which can be
listed. Also, nouns and verbs have a wide array of derivational possibilities,
whereas derivation is impossible for many of the others (quantifier, adverb,
preposition, conjunction, particle and interjection) or very limited (pronoun
and numeral). The differences can be shown in the following chart:
In this section I will give operational definitions for noun and verb and
illustrate the other word classes.
3.3.1. Noun
3.3.2. Verb
d. *no-mba-rabu-rabu
e. *no-feka-rabu-e
b. *no-ghosa-e
c. *ni-ghosa-no
b. *no-kala-e
c. *ni-kala-no
d. *no-mba-kala-kala
e. *no-feka-kala-e
On the relationship between the verb classes a-, ae- and ao- and the transitive
and stative/dynamic intransitive classes, see 4.4.
Stative intransitive verbs are usually translated as adjectives in English.
To use the term 'adjective' for these words obscures the important fact that in
Muna these words behave in every respect as verbs. Both predicatively and
attributively these putative adjectives are marked in exactly the same way as
verbs, that is, with subject inflection and as active participles, as illus-
trated by the following two pairs (both verbs are ao-verbs):
Because of their structural similarity, ghae and kesa are grouped together as
verbs: they differ in their derivational possibilities, as illustrated above
3 WORDS AND WORD CLASSES 47
by ghosa and kala, hence the categorization into stative and dynamic in-
transitive verbs.
3.3.3. Pronoun
There are two, structurally different, sets of pronouns: personal pronouns and
demonstrative pronouns.
The set of free personal pronouns (to be distinguished from pronominal af-
fixes such as subject markers) comprises six basic members, for example inodi
'I'; ihintu 'you' (see 5.3).
The set of demonstrative pronouns comprises six members, for example aini
'this'; awatu 'that (far)' (see 5.5).
Other traditional pronominal categories are not needed, since 'possessive
pronouns' are in fact possessive suffixes (5.4), while positing a category of
interrogative pronouns would obscure the fact that question words are drawn
from more than one word class (see 8.6.2).
3.3.4. Numeral
3.3.5. Quantifier
3.3.6. Adverb
Adverbs are words that modify the verb or the whole clause. They form a closed
class with several subclasses:
1. Temporal adverbs, for example indewi 'yesterday'
2. Intensifying adverbs, for example sepaliha 'very'
3. Focusing adverbs, for example dua 'also, too'
4. Negators, for example miina 'no, not'
5. Disjuncts, for example hadae 'maybe'.
Adverbs are further discussed and illustrated in the chapter on clausal
syntax (7.11).
3.3.7. Preposition
3.3.8. Conjunction
3.3.9. Particle
There are two monosyllabic words in Muna that do not fit into any of the above
categories, the article o and the plural marker ndo. They share the feature of
pre-nominal position, but otherwise differ markedly from each other. These
particles are treated in 5.6 as part of the nominal phrase.
3.3.10. Interjection
Interjections form a closed class of words that are aberrant in their syntactic
behaviour from other word classes in that they constitute single-word clauses.
The following types of interjections can be distinguished:
1. emotive interjections, for example adede 'ouch' (expresses pain)
2. interjections to call or chase away animals, for example sio (call to chase
away chickens)
3. the word uumbe 'yes'.
For a fuller list of interjections see 7.12.2.
In this section I want to mention several issues that are connected with word
classes, without treating these issues in any detail.
1. There are regular processes by which words can change from one class to
another, usually through the addition of morphological material. Rules
describing morphological processes must therefore specify the input and
output in terms of word classes, as is done in Chapter 10 on derivational
morphology. Examples of word-class changes:
2. Some words are members of more than one word class or seem to have a derived
usage in another word class. Such words are best considered to have a basic
word-class membership and a derived one through a process which is irregular
and lexically determined. Examples:
3 WORDS AND WORD CLASSES 49
3.5. Conclusion
A comparison of Muna and, for example, English shows that the category 'verb'
is often used in Muna where English has other word classes. This is not only
true for the English category 'adjectives' but also for such words as no-langka
'rarely', ne-nturu 'often', no-nea 'usual(ly)'. Morphologically they are verbs
because of the subject markers ne- and no-, but semantically they correspond to
adverbs in English. To this category also belong words such as nae-tolu 'in
three days', nae-fato 'in four days', verbs which are based on numerals.
Another example is provided by manner adverbs marked by feka-, which are
optionally inflected for subject (see 7.10.3 and 10.2.5). Other words that can
optionally receive subject inflection include conjunctions such as kansuru 'at
once' and the preposition peda 'like' (6.2.4).
Finally, it should be mentioned that the division into word classes is both
necessary and meaningful for basic - that is, underived - words. With certain
derived words problems arise, however. Participles, for example, retain some
verbal features, acquire a number of nominal characteristics and have other
unique properties.
The following chart shows the spreading of some of these features over the
basic categories of noun and verb and the active and passive participles:
1. function as subject + + + -
2. basis of co-derivation + + - -
3. negator suano pata pata miina
4. person marking P P - S
5. modified by demonstrative + + + -
or descriptive attribute
Notes: 1. Verbal subjects (parameter one) do occur, but are very rare.
2. For ko-derivations, see 10.2.21.
3. For negators, see 8.5.
4. In the fourth parameter (person marking) P stands for possessive
inflection (4.9), while S signifies subject inflection (4.1).
Active participles are not marked for person.
5. For demonstratives see 5.5, for descriptive attributes 5.9.2.
The conclusion must be that passive participles are very noun-like in their
syntactic behaviour, more so than active participles. Both, however, have
unique properties because of their special negator pata. Other derivations take
up other intervening positions. The terms noun and verb should therefore be
used with caution with respect to derived categories. In discussions of the
participles the terms noun and verb will therefore be avoided.
Chapter 4
Verbal inflection
This chapter deals with verbal morphology, one of the most complex aspects of
Muna grammar. More specifically, it is restricted to verbal inflection. First,
subject inflection is discussed with its complexities, such as the three
different paradigms for subject markers (4.3, 4.4), the influence of the
irrealis mood on subject inflection (4.5) and the so-called definiteness shift
(4.6). A first analysis of these topics is given in Van den Berg (1987). Direct
and indirect object inflection are treated in 4.8. The chapter ends with a
discussion of inflection in morphologically complex words (4.9).
In (1a) the prefix no- marks the subject of the verb leni as third person
singular (it also indicates the realis (R) mood, see 4.5), while the prefix a-
in (1b) marks the subject as first person singular (realis).
These clauses can be expanded by a full nominal or pronominal subject; a free
pronoun usually indicates emphasis (see 5.3.1):
In these cases there is agreement between the noun phrase functioning as the
subject of the clause and the prefix indicating the subject on the verb.
Henceforth these prefixes will be called subject markers. The following clauses
are therefore ungrammatical:
4 VERBAL INFLECTION 51
b. *inodi no-leni
The following chart presents the subject markers for one class of verbs. For
ease of reference the full paradigm of subject inflection illustrated on the
verb kala 'go' is given together with the free pronouns, which are further
treated in 5.3.
1. -Vmu, the plural marker on first person inclusive and second person, is a
derivational suffix (see 10.2.30). The first vowel of the suffix is a copy
of the base to which it is attached (vowel harmony); compare the following
examples:
3. The do-forms can also be used for the second person singular, in which case
they have a rather formal connotation. Their use does not imply politeness
but rather aloofness; the form is especially suitable for addressing
foreigners. The following possibilities exist for the second person
singular:
4.2. Agreement
As explained in 4.1, there is agreement between a full nominal subject and the
subject markers on the verb with respect to number. There are, however, certain
regular exceptions to this rule:
3. A third person singular nominal subject that has plural reference takes a
plural subject marker. This usage is common with names and titles:
Muna has three verb classes with partly different subject markers. First
the formal side of these classes (the different subject markers) will be
4 VERBAL INFLECTION 53
Further evidence for the existence of these three verb classes is offered by
the morphological shape of the locative noun derived from these verbs:
(19) *ae-lente
*kae-lente-ha
*a-late
*kao-late-ha
The full inflection of the three verb classes is as follows. For the sake of
comparison the a-class is repeated.
A few examples of ae- and ao-verbs will illustrate these subject markers:
ae-verbs:
ao-verbs:
The existence of three verb classes automatically raises the following ques-
tion: are there factors which determine or predict to which class a given verb
will belong? Phonological conditioning is ruled out. Compare the following
three verbs which all have the same initial two phonemes, yet belong to three
different classes:
3. transitive verbs:
1. transitive verbs:
3. transitive verb:
bhalo 'answer'
In order to obtain some idea of the correlation between the verb classes and
their member verbs I have taken the first 200 verbs of the Muna dictionary file
and classified them as belonging to one of the subgroups as specified above.
The result is as follows:
A few conclusions can be drawn from this chart. Transitive verbs have a strong
preference for the ae-class, whereas dynamic intransitive verbs are usually
(but not always) members of the a-class. Stative intransitive verbs prefer the
ao-class, although a fair number of them are also found in the a-class. In
other words, there are some strong tendencies that point to the following
correlations:
The many exceptions force us to handle this rule rather loosely. The main
pattern is clear, but there are many unexpected counterexamples.
So far only basic underived verbs have been discussed. Derived verbs also
belong to one of the three verb classes. Notice the following points about
verb-class membership of derived verbs:
a. The causative prefix fo- moves a verb to class ae- (see 10.2.6):
2. Reduplicated verbs (without other affixes) always belong to class ae- (see
10.3):
bheta 'sarong'
ae-bheta 'I wear a sarong'
bhose 'paddle'
no-bhose 'he rows a boat'
4 VERBAL INFLECTION 57
All the verb forms discussed so far can refer to either the past or the
present. They can therefore be called the realis mood, to set them apart from
the irrealis, a differently inflected verb form. The irrealis has two primary
uses:
1. it refers to the future, or it expresses a wish, a desire or an intention;
2. it is obligatorily used in negative clauses.
Examples of the usage of the irrealis will follow below; first the formal
side of the irrealis will be dealt with. The most important difference between
the realis and the irrealis is the different set of subject markers that are
prefixed to the verb.
The following chart shows the different sets of realis and irrealis subject
markers for each of the three verb classes.
Notice that a number of subject markers are identical in the realis and in the
irrealis (sg 1, sg 2, pl 2, pl 1 ex).
For the ae-class and ao-class these irrealis subject markers can be directly
attached to the verb form, but this is not the case with the a-class. In the a-
class the irrealis subject markers are prefixed to the so-called um-form. The
um-form is a bound form that is the result of the operation of a morphological
process of which the most common form is the infix -um-. For a detailed
discussion of the allomorphy of -um-, see 2.8.1. For ease of reference the main
points are summarized here:
58 A GRAMMAR OF THE MUNA LANGUAGE
Realis Irrealis
A. Class a-:
B. Class ae-:
C. Class ao-:
The partial similarity between the a-class and the ao-class has disappeared in
the irrealis. Take for example the following two verb forms in the third person
singular:
In the realis these verb forms seem to belong to the same class. In the
irrealis, however, it appears that they belong to different verb classes:
In order to illustrate usage and meaning of the irrealis, clauses will be given
in which irrealis forms occur. As mentioned above, there are two primary
usages:
2. It is the only verb form that can be used in negative sentences, that is, in
the presence of negators such as miina 'not', miina-ho 'not yet' and pa
'will not'. The presence of a negator before the verb often triggers the
clitic -a on the verb (see 8.5 and 9.22):
When the negator is placed after the verb for reasons of emphasis, the irrealis
is also used:
c. *ne-rabu-e
Rabu is an ae-verb, hence the subject marker ne- in (36a). When the direct
object is 'definite' (for example, a pronominal suffix), the verb shifts to the
a-class; hence the form no-rabu-e in (36b) instead of the expected but non-
existent *ne-rabu-e.
This shift only applies to transitive ae-verbs. Transitive a- and ao-verbs,
of which there are only a few, do not change:
Examples of these definite objects will be given below. The numbers refer to
the specification of the direct objects given above; where appropriate in the
(a) examples the 'normal' usage of the verb is illustrated, in the (b) examples
the definiteness shift is given.
2. Free pronoun:
3. A personal name:
a. a possessive suffix:
b. a demonstrative pronoun:
d. a temporal adverb:
6. An 'implicit' object:
The verbs ghompa in (49) and tei in (50) are both ae-verbs. The expected form
in (49) is therefore ne-ghompa or no-ghompa-e. The actual form no-ghompa can be
accounted for by 'direct object suppression'; since the direct object is
retrievable from the same sentence, the pronominal suffix -e can be 'sup-
pressed', but the effect on the verb (the definiteness shift) is retained. The
suffix -e can be introduced without any apparent change in meaning. The
circumstances under which this suppression occurs are not clear.
When the object noun is modified by another plain noun or by the question
word hae 'what', the definiteness shift does not seem to be obligatory: both a-
and ae-inflections occur (but see the partitive usage in 4.6.3):
When an ae-verb has become an a-verb because of the definiteness shift, the
regular processes of um-formation for an a-verb now apply too. In other words,
this verb has become an a-verb both in the realis and in the irrealis. Consider
the following examples:
4 VERBAL INFLECTION 63
The definiteness shift is triggered because of the possessive suffix -no. When
these sentences are negated, the difference between the ae-inflection and the
a-inflection becomes even more transparent:
The following chart shows the distribution of the definiteness shift over the
three verb classes in simple and derived transitive verbs.
a- a-
ae- a-
ao- ao-
a- ae- a-
ae- ae- a-
ao- ae- a-
a- ae- ae-
ae- ae- ae-
ao- ae- ae-
a- ae- a-
ae- ae- ae-
ao- ao- ao-
There are certain exceptions to the definiteness shift. These can be grouped
under the following categories:
This opposition only seems to take place with direct objects that have been
made definite by nominal modification such as a possessive suffix or another
noun. With pronominal suffixes this option is not open:
b. *ae-fo-ndawu-e
3. The verbs wura/wora 'see' and rabu 'make'. Both are ae-verbs, but with
definite objects such as modified nouns they regularly do not shift to a-
verbs. The reasons for this are still unclear.
In spite of the regularity of the definiteness shift and its exceptions, there
remain a number of cases in the text material where the shift does take place
and the object is not definite, or alternatively, where the shift does not take
place while the object is definite. (See, for example, Appendix text 3,
sentences 1 and 23). These residual problems need further study.
This is clearly not a definiteness shift. Rather, the opposite is true. The
noun is generic and the activity is seen as a whole; the object is incorporated
in the verb. Other possible translations of (67b) and (68b) are: 'I am sarong-
weaving' and 'They are deer-hunting'. Because of this object incorporation the
verb is treated as an intransitive verb and therefore changes to the a-class,
which is typically intransitive.
Furthermore, the object of such verbs can be left out while the verb remains
in the a-class:
Certain other transitive verbs also allow this 'object suppression', but then
the verb remains in the ae-class:
The circumstances under which this detransitivization shift takes place are not
fully understood. Possibly this shift indicates a completive or habitual
aspect. Compare the following two short dialogues, which were provided to
illustrate the differences:
The following chart shows the pronominal suffixes marking direct and indirect
object, in combination with the free pronouns and the possessive suffixes. The
last two categories will be further discussed in Chapter 5.
Direct and indirect object suffixes are part of the verb. This is most clearly
shown by the position of the suffix -mo in the verb. This -mo follows all other
suffixes. It also follows the pronominal suffixes.
1. when the object is the goal, target, patient or causee of the verb:
3. with the question word hamai 'which, where' (8.6.2), the existential verb
naando 'be' (7.2.1), the negator miina 'not' (8.5.1) and demonstrative
derivations with ha- (5.5.8 and 10.2.11):
2. to express the semantic function 'goal' with Cao-verbs (10.2.15). When the
suffix -Cao is followed by one of the indirect object suffixes or by the
suffix -ghoo, the final o of -Cao is deleted. Only with the first person
suffix -kanau is this rule optional.
3. with certain intransitive verbs to express goal, such as kanu 'make ready to
go', asi 'like, love', limpu 'forget'. These verbs are treated as
intransitives on formal grounds since they cannot be suffixed with a direct
object pronominal suffix. Indirect object suffixation (either pronominal or
with -ghoo (see 7.9)) is optional with kanu and limpu, but seems to be
obligatory with asi.
For a full discussion of -ane (and its relation to -ghoo) see 7.9.
Indirect object suffixes, unlike direct object suffixes, do not trigger the
definiteness shift:
b. *a-gholi-angko pae
4 VERBAL INFLECTION 71
It is possible to have both a direct and an indirect object suffix on the verb,
but in that case the direct object is limited to -e; the indirect object
precedes the direct object:
The definiteness shift is triggered in all these cases, because of the direct
object suffix -e. Combinations of other direct object pronominals with indirect
object pronominals are ungrammatical:
(99) *no-owa-kanau-da
3sR-bring-me-them
This meaning can be expressed in the following way, where the recipient is no
longer a suffix but a free pronoun:
When -ane and -e are both suffixed to a verb, the two e's are frequently fused
into one single -e; the verb remains in the a-class:
The chart presented at the beginning of 4.8 raises the question how the first
person inclusive dual and plural are expressed when they are direct or indirect
pronominal objects. As can be seen, there are no suffixes for these categories.
The language uses two other mechanisms:
a. The full pronoun is used (an option which is also open for all the other
persons for emphatic purposes, see 5.3.1):
b. The prefix fo- is used (the detransitivizer, not the causative fo-, see 8.3
and 10.2.7):
For the first person inclusive indirect object this fo- can also be used,
together with the indirect object marker -ghoo. In this case the class affixes
me- and mo- also surface (see 10.2.28):
When there is no direct benefit for the hearer, -kanau will be used instead.
Notice the difference between the following two sentences:
In (109) there is a clear implication that getting the chair is also for the
benefit of the hearer, because, for example, a guest has arrived and the host
commands his son to get a chair. The son as part of the family is also
responsible for the well-being of the guests. (110), on the other hand, would
be uttered by a father to his son when he needs a chair for himself.
Finally, -kaeta is used in polite adhortative sentences, where the speaker
also includes himself:
Simple verbs all have the common characteristic of subject inflection. This is
not the case with morphologically complex words derived from verbs. As far as
inflection goes, the following formal categories can be distinguished for these
derivations, which will be discussed in turn:
1. Regular subject inflection. This comprises most verbal prefixes and suffixes
(for instance fo-, po-, -Ci, -Cao), as well as reduplication. (See Chapter
10 for a detailed treatment of derivational morphology.)
2. Nominal inflection. This covers not only nominalizations such as the affixes
-ha, ka- and ka-/-ha, where the possessive suffix marks real possession, but
also certain derived words which retain many verbal properties but where the
agent is expressed by a possessive suffix. Nominal inflection is illustrated
by the following derivations:
1. the passive participle, marked by the prefix ni-/ne- (see 9.1.2 and
10.2.33):
2. the temporal subordinative prefix sa- 'when' (see 9.7.1 and 10.2.44):
3. the temporal prefix paka- 'when just, when first' (see 9.7.1 and
10.2.37):
a. active participle
The form of the active participle is dependent on the verb class to which
the verb belongs. Its formation is as follows:
b. imperative
The form of the imperative is also dependent on the verb class. The im-
perative for each verb class is as follows (see 8.7):
c. deverbal derivations
1. simple deverbal base following miina bhe 'there is no...' (see 7.2.2 and
8.5.1):
4 VERBAL INFLECTION 75
This chapter deals with the structure of the nominal phrase (NP). Starting with
the formula for its internal structure (5.1), the various possible heads of
nominal phrases and their modifiers are discussed. The noun as head of an NP
will be discussed in 5.2, the personal pronoun in 5.3, the possessive suffix
and modifying nouns in 5.4, the demonstrative pronoun in 5.5, nominal particles
in 5.6, the measure phrase (including numerals, classifiers and quantifiers)
in 5.7. Relative and simple verbal clauses functioning as heads are treated in
5.8. Question words can also head an NP, but they will be discussed in 8.6.2.
This chapter is not only concerned with the syntax of noun phrases but also
with the morphology of its constituent parts. Word classes whose primary
function is to head or modify an NP will be analysed here in terms of their
derivational possibilities (noun, personal pronoun, demonstrative and numeral).
The head of a nominal phrase can be a noun (N), a personal pronoun (Pro), a
demonstrative pronoun (Dem), a relative clause (Rel) or a simple verbal clause
(Cl). With these heads a number of modifiers are possible, such as a measure
phrase (MeasPhr), the article (Art), a quantifier (Quan) or an attribute
(Attr). The greatest number of attributes is found when the head of the NP is a
noun. In other cases the possibilities of modification are more restricted.
The unmarked internal structure of these types of NPs is as follows (marked
orderings are discussed in 5.10; heads of NPs are capitalized and not in
brackets):
a. (Art)
(MeasPhr) N (NP) (-Poss) (Attr) (MeasPhr) (Attr) (Dem)
(Quan)
c. DEM
e. CL (Dem)
In addition to common nouns, the following subtypes of simple nouns (that is,
morphologically not complex) can be distinguished:
The morphological and syntactic features of these nouns are different from
those of the common noun. Names, for example, do not normally receive affixes
(see 5.2.3). Classifiers and measure nouns can be prefixed with a prefixed
numeral. They are treated in 5.7.4 and 5.7.5. The noun wuto is discussed in
5.4.
1. Nominal derivations:
lambu-hi 'houses'
b. prefix si-; meaning 'be one in respect to; have the same' (10.2.46):
3. Unproductive derivations:
The derivations discussed so far are all regular and productive for common
nouns. A few common nouns, however, can also occur in derivations which are
only partly productive, that is, these processes only apply to certain
categories of common nouns. Notice the following 'minor' derivations:
a. foko-/-u (10.2.9):
e. po- (10.2.41):
In the preceding section the simple common noun was described in terms of its
derivational possibilities. In this section the starting point will be at the
opposite end, that is, the morphologically complex noun is described in terms
of its bases and affixes. Necessarily there will be some overlap with the
preceding section, since certain nouns have nominal roots. These affixes are
discussed in detail in Chapter 10.
The following types of derived nouns exist:
2. Circumfix ka-/-ha (and its allomorphs kae-/-ha and kao-/-ha) on verbal roots
(10.2.18):
Unproductive derivations:
Compounds are relatively rare in Muna. One type which is formally marked
through prenasalization is discussed in 10.4. Examples:
Proper nouns (names) are used for identifying persons and locations. Proper
nouns have very limited inflectional and derivational possibilities; only
80 A GRAMMAR OF THE MUNA LANGUAGE
locative names can be affixed with ko-/-ha-e (5.7.3), while possessive inflec-
tion with personal names is not impossible but rather unusual.
Notice the following points about proper nouns referring to persons:
1. Names are usually preceded by the articles la (for men) or wa (for women),
both as term of address and as term of reference. When names are written for
official purposes, these elements are capitalized. In normal speech la is
often reduced to a.
La Aso
Wa Sukia
2. When somebody belongs to the nobility, he/she is allowed to place ode before
his/her name. In writing this is usually capitalized:
La Ode Malefu
Wa Ode Hanafia
3. Names are often abbreviated. This is especially the case when people are
directly addressed (see 10.6):
4. Names can be preceded by the particle ndo, which denotes plurality (5.6.5).
It means: 'X cum suis; X and his friend(s)/relative(s)'. It is also found
with animal names when they figure as characters in a story:
5. When animals are the main characters in a story, the common name is made
into a proper name by reduplication, preceded by la (or the reduced form a).
The reduplication may be full (two syllables) or supernumerary (three
syllables) (see 10.3):
The last form is the most usual one. Some other examples that occur in
texts:
5 THE NOMINAL PHRASE 81
This last example is surprising, since the article a is found only on the
reduplicated part, not on the root. It seems that when the root contains
more than two syllables, reduplication precedes prefixation of the article
a. Disyllabic roots are first prefixed with a or la and then partially or
fully reduplicated.
Reduplication is also used to form names out of phrases:
The free pronouns were presented in 4.1 in combination with the subject
markers. They are repeated here, along with the possessive suffixes.
Inodi, ihintu and ihintu-umu are full forms, idi, hintu and hintu-umu are
reduced forms (5.3.3).
82 A GRAMMAR OF THE MUNA LANGUAGE
5.3.1. Usage
a. after prepositions:
(9) omo-lolu sepaliha ihintu itu 'YOU are very stupid indeed'
2sR-stupid very you that
c. *fumaa-kanau inodi
5.3.2. Derivation
The possibilities of derivation with the free pronouns are very limited:
The reduced personal pronouns idi, hintu, hintu-umu can occur in exactly the
same positions and seem to have the same usage as the full free pronouns. In
(8) and (12) examples have already been provided of these reduced pronouns.
Other examples:
The only difference between the full and the reduced free pronouns is that the
derivational process with sa- as described in 5.3.2 cannot apply to the latter.
Forms such as ta-hintu 'just you' and idi-mo 'it is I' are less usual but not
impossible, but forms such as *sa-idi-idi do not occur.
Also, pronouns can be modified by a full appositive NP, which is then actually
outside the head NP (see 5.9.1):
5.4. Possession
The paradigm of possessive suffixes, which was presented in 5.3, is used for
nominal inflection. This inflection is found with nouns, passive participles
and certain other derivations (see 4.9). Examples:
As mentioned in 2.8.3, the third person plural suffix -ndo alternates with -do
when suffixed to bases that contain a prenasalized consonant.
To emphasize the possessive pronouns, the following two possibilities exist:
1. free pronouns are added, usually after the head noun (see 5.3.1):
2. the emphatic/reflexive noun wuto 'self' is added. This noun wuto itself is
also obligatorily suffixed with a possessive:
(43) suano ka-ghosa-no wuto-no 'it was not his own strength'
not NOM-strong-his self-his
Both possessive suffixes have to be present; the following forms are therefore
ungrammatical:
b. *lambu wuto-ku
house self-my
The third person singular possessive suffix -no is also found in a number of
adverbs and conjunctions. This is probably a special lexicalized use of -no:
When a head noun is modified by another noun or a noun phrase, the possessive
linker -no is suffixed to the first noun. This linker -no is a special use of
the third person possessive suffix -no. As in most languages, the relation
between the two nouns is typically one of possession, but there is a wide array
of meaning relations such as partitive, provenance, characteristic of, located
in and so on. Some examples:
b. *pughu-no-ku kalei
c. *pughu-ku-no kalei
Notice that constructions like (51) and (52) are structurally ambiguous. The
final possessive suffix can theoretically modify the whole phrase or just the
last noun. In (51) -ku only modifies ama 'father', whereas in (52) -mu modifies
the whole phrase pughu-no kalei.
When the noun phrase has plural reference, the plural possessive linker -ndo
can be used. The plurality may either be determined by the head noun or by the
modifying noun. The use of -ndo is not obligatory; -ndo stresses the plurality
of the head noun and therefore of the whole phrase:
The use of either -ndo or the plural suffix -hi marks the whole phrase as
plural. The occurrence of -hi in both the head noun and in the modifying noun
is avoided.
When the modifying noun is animate, the possessive linker can be omitted
without any change in meaning:
a. when the second noun is the name of the material from which the first noun
is made:
d. when the second noun is a name, even when this name is derived from a common
noun:
Compare:
c. kambea-no mpatani 'mpatani flower'
e. when the second noun is a derivation with ka- or ka-/-ha that functions as a
descriptive phrase, a relative clause or an apposition to the head noun:
5.5.1. Forms
There are six basic demonstrative pronouns in Muna, which occur in two sets,
one with and one without initial a-. These two sets are as follows:
1 aini ini
2 aitu itu
3 near amaitu maitu
3 far:neutral awatu watu
3 far:high atatu tatu
3 audible anagha nagha
On the difference between the two sets, see 5.5.5. In addition to these basic
sets various combinations are possible, such as amaitu-ini. These will be
discussed in 5.5.7. Minor forms are the dialectal asotu and the combination
awagha-itu.
All demonstrative pronouns can be used to refer to the location in space of a
given entity. In addition, several demonstrative pronouns can also refer to
location in time and location in discourse. I will call this the locative, the
temporal and the anaphoric usage of demonstratives.
The first person demonstrative aini is used for whatever is near the speaker,
that is, for what is within his reach. Aitu, the second person demonstrative,
refers to an entity that is closer to the hearer than it is to the speaker,
but not necessarily as close as aini is to the speaker. Aitu can also be
used for something near the speaker when aini is already in use ('not this
one, but that one', when both objects are at the same distance). When the
entity referred to is about the same distance away from both speaker and
90 A GRAMMAR OF THE MUNA LANGUAGE
hearer, several options are open. The third person amaitu typically refers to
an object that is not far away, especially when it is contrasted with awatu,
which is further away. Obviously, 'far away' is a relative term and much
depends on the perspective of the speaker. In any case, awatu seems to be the
most neutral form in the third person series. It contrasts with atatu in that
atatu has an extra semantic component of 'height' (from the point of view of
the speaker) which is lacking in awatu. Informants would usually say that atatu
refers to a high location and awatu to a low location, but from conversations
and texts it is evident that awatu is in fact the neutral form, which can even
be used for relatively high points. Only in opposition to atatu does it mean
'low' or 'level'. In certain contexts atatu seems to mean 'far away' or
'further away' in opposition to awatu, rather than 'higher than'. Anagha,
finally, is used for an object that cannot be seen by either speaker or hearer,
but is audible. A crying child or a barking dog that is not visible can be
referred to by means of the demonstrative anagha. The primary usage of anagha,
however, is anaphoric (see 5.5.4).
In the glosses and in the translation no attempt is made to reflect these
distinctions: 'this' is the equivalent of aini, 'that' of all the other
demonstratives.
In addition to these demonstratives in standard Muna, there is also a
dialectal demonstrative asotu. Its meaning is probably equivalent to awatu. Its
use, however, is limited to certain villages in the kecamatan Katobu and it is
considered substandard by Muna speakers from other areas and even from Katobu
itself.
All demonstrative pronouns can function as the head of an NP or as a
modifying attribute. Examples of this locative usage:
(77) ...na-t[um]aburi bhe sau atatu 'it will press down that tree'
3sI-press with tree that
(78) dahu anagha dahu-ku 'that dog (the one you hear) is
dog that dog-my mine'
5 THE NOMINAL PHRASE 91
One basic demonstrative can have a temporal meaning, namely aitu. Other
temporal meanings occur with the complex demonstratives as discussed in 5.5.7.
Aitu has two meanings:
These meanings are obviously related, the difference being their 'deictic
anchoring'. Aitu refers to a point which has already been established in the
previous discourse, and only when no such point is present, does it refer to
the present moment, that is, the actual moment when the speaker is talking.
Examples:
(80) ane nao-maa-kanau, aitu nao-repu 'when he eats me, he will croak'
if 3sI-eat-me now 3sI-croak
In (80) a moment in the future has already been established, hence aitu means
'then, at that time'; in (79) no such point is present, hence the meaning
'now'. The 'now' of aitu is only a very short moment; for longer stretches of
time ampa-aitu 'now, nowadays' is used.
In one text aini occurs as a temporal demonstrative with the same meaning as
aitu 'then, at that time'. This usage of aini is very rare:
Certain demonstrative pronouns are often used to refer back to an entity that
has already been introduced in the preceding context. In such cases aini, aitu,
amaitu and anagha can be translated either as demonstratives or as definite
articles, depending on the context, although they are still glossed as
demonstratives. Awatu and atatu do not have this anaphoric function.
Notice also the difference between peda aini 'like this (future)' and peda aitu
'like this/that (past)' in the following sentences:
(88a) identifies one among many of my friends, while (88b) refers to 'my
friend' who has already been introduced and is now the topic in the discourse,
or he is close to the speaker at the moment of speaking.
5 THE NOMINAL PHRASE 93
(89a) implies that there are other places called Raha, while (89b) implies that
the phrase is uttered by someone who is in Raha.
The referential set is therefore used when there is no need to identify or
re-identify the noun or noun phrase under consideration. This is especially the
case when:
1. when the noun is modified by the article o (see 5.6). Since the 'meaning' of
o has no relation with definiteness or givenness, this restriction on the
co-occurrence of o and the identifying demonstratives cannot easily be
accounted for.
2. when the NP is found in a prepositional phrase. In those cases the
referential demonstratives are usually found, even when there is a new,
unmodified head noun. When the demonstrative follows the preposition
directly, the referential demonstrative must be used:
ne ini 'here'
ne watu 'over there'
a. ini
'they did not forget to take provisions with them, but their pro-
visions were only one kasopa (= traditional bowl-like container)'
b. itu
(99) no-wanu kaawu anahi itu 'when the child got up'
3sR-get.up after child that
The difference between inodi ini ('I this') and inodi itu ('I that') is that
the former is used when the speaker gives objective information about himself,
whereas he uses the latter when he gives his opinion about something else.
There is a subjective element in inodi itu which is lacking in inodi ini.
5 THE NOMINAL PHRASE 95
(101) inodi ini, no-mpona-mo ao-saki 'I have been ill for a long time'
I this 3sR-long-PF 1sR-ill
(102) inodi itu, a-asi-ane anagha 'as for me, I like that'
I that 1sR-like-it that
c. maitu
d. nagha
e. watu
f. tatu
There is some variation between the identifying and the referential set.
Especially with unmodified given nouns there is freedom as to which of the
two sets is used. When a narrator uses the identifying demonstratives in
such a case, he wishes to re-identify his main character; but when he uses
the referential demonstratives, he assumes the entity is still known. The
96 A GRAMMAR OF THE MUNA LANGUAGE
following example, taken from a story, is about a man whose wife goes to spy on
him and then returns home. Both demonstratives can be used in this example,
since the woman constitutes 'given' information, which is not modified by
another element:
Sometimes the referential demonstratives are used when they function neither as
the head of an NP nor as a modifying element of an NP. Often some kind of
deictic meaning component can be detected, but it is not possible to do more
than list the usages that have not been dealt with in the previous section.
a. ini
naewine 'tomorrow'
naewine ini idem
indefie 'when?'
indefie ini 'in the past, formerly'
3. ini follows the irrealis in main clauses where it indicates that the
planned activity is being performed at the moment of speech:
b. itu
(123) ingka ama-mu bhore itu 'it's your father, you silly
ENIM father-your silly that thing'
alo 'night'
alo itu 'tonight'
gholeo 'day'
ghole itu 'today'
naewine 'tomorrow'
naewine itu idem
c. maitu
maitu, like ini, can also modify a clause of which the main verb has already
been mentioned or suggested, thus acting as a clause linker:
d. nagha
(134) sau awatu-ini no-bhari bhake-no 'THAT tree has got much fruit'
tree that 3sR-many fruit-his
(136) anagha-ini naando se-mie moghane 'once upon a time there was a
that be one-CLAS man man'
But it can also refer to an object that is not in view, but which still forms
the topic of conversation:
2. Suffix -e (10.2.2):
Awatu and atatu can be emphasized by adding the suffix -e. This suffix is
pronounced on a very high pitch and is drawn out considerably. Compare also -e
on vocatives in 7.12.1.
____
_____ ____/
(152) lambu awatu-e:: 'THAT house over there'
house that-E
5.6. Particles
In Muna there is one article, the particle o, which always precedes the noun it
modifies. Its usage and meaning are completely different from that of the
article in Indo-European languages. There is no relation with definiteness or
specificness; it does not serve emphasis nor is it a topic marker. In fact,
there is no clear meaning in the sense of referential meaning. The term
'article' is used, following Anceaux (1952) in his description of Wolio. The
function of o is to signal the relative syntactic freedom of a noun or a noun
phrase. In other words, a noun modified by o has a lesser degree of syntactic
binding to the rest of the clause than a noun which is not so marked.
Therefore, we typically find NPs modified by o in the following constructions:
The article is not usually found before an NP following the main verb, either
as subject or as object, nor in possessive constructions after the linker -no.
In those cases the syntactic binding is too strong; but see 5.6.4 for variation
in usage.
5.6.2. Usage of o
Notice also the following two examples with the connective tawa 'or';
especially the second example is interesting, as o is here used before a noun
which stands in a possessive relationship to the governing noun. As mentioned
above, such a noun does not normally take o. In this case it can take o because
of the coordination with tawa:
(164) suano o mbadha so ne-ghondo 'it is not the body that will
not ART body FUT P.PART-look be looked at'
4. Preverbal subject:
b. *kapoluka no-bisara-mo
5. Appositions:
The only affix that can be added to the article is the prefix ta- 'only, just'
(10.2.48):
There are restrictions on the modifiers that can combine with the article o in
an NP:
(175) a. *o lambu-ku
ART house-my
b. *o ka-kesa-no
ART NOM-beautiful-its
The only exception is o hae-no 'what?' used as a question to ask for the
identity of a specific object (see 8.6.2), although it is not completely clear
whether -no is in fact a possessive suffix here.
But when the measure phrase follows the noun, o can be used:
5 THE NOMINAL PHRASE 105
b. *welo o lambu
b. *so o anahi
the examples given above amply show that o can occur in combination with other
modifiers. Notice also the following example, where the first two nouns (which
are unmodified) take o, whereas the last two nouns, which are modified by a
'clipped participle' (see 5.9.2 and 10.2.28), do not:
lalo mo-asi
heart A.PART-pity
What was said in 5.6.1 and 5.6.2 is true for the speech of most people and is
consistent with a fair number of texts. However, in a few texts, o is found
much more frequently than can be explained by the criteria set up above. The
following sentences from such texts illustrate these cases:
These examples show the use of o modifying a postverbal object (186), a post-
verbal subject (187) and a dependent noun after the possessive linker (188).
When these sentences were checked with other informants, some of them reacted
against the use of o in these contexts. All these informants came from basi-
cally one dialect area, although there were age differences.
In order to find out to what extent the usage of o was subject to individual
preference, a random test was organized. Clauses from texts containing the
article in an 'unexpected' place were transformed into an exercise for filling
in o. Every NP was given a blank space before it in which informants could
indicate whether o was obligatory (+), impossible (-), or optional ().
The following syntactic positions were distinguished (each position was
represented by at least three sentences; overall order was random):
1. preverbal Subject (S);
2. preverbal Object (O);
3. postverbal S (intransitive verbs);
4. postverbal O;
5. postverbal S (transitive verbs);
6. dependent nouns after the possessive linker;
7. nouns after a preposition.
Some of the results of this test (a few clauses for four informants) are as
follows:
5 THE NOMINAL PHRASE 107
A B C D
1. preverbal S:
a. o karambau no-mente
'the buffalo was amazed' + + + +
b. o oe no-sampu dua
'the water also went down' + + + +
2. preverbal O:
3. postverbal S:
4. postverbal O:
f. no-salihi o karambau
'he admonished the buffalo' - -
g. ne-gholi-mo o pae
'she bought rice' -
h. de-fo-naando o rompu-ha
'they organized a meeting' - -
6. dependent noun:
j. amba-no o pikore
'the bird said' - - - +
7. after preposition:
To sum up our conclusions concerning the article, the following rules governing
the application of o can be stated:
These observations point towards the fact that o signals relative syntactic
freedom within the clause of the NP which it modifies. Single-word clauses and
enumerations are typical examples of 'free' positions, and a preverbal subject
is (in intransitive clauses) the marked position (see 7.3). Postverbal nouns
are more 'bound' (in whatever sense) to the governing verb or to the clause as
a whole than preverbal nouns, at least in Muna, where the preverbal position is
reserved for special pragmatic marking. O in itself, however, does not have a
pragmatic meaning, as is clear from the examples. But this 'binding' theory
does not explain the co-occurrence restrictions with, for example, possessive
suffixes.
Another problem is the variation that exists. Possibly the use of o is linked
up with other features such as animacy or the intervention of adverbs between
the verb and the postverbal subject. Possibly, too, this variation is connected
with speech tempo. One informant suggested that in slow, careful speech the use
of o is more appropriate with postverbal nouns than in normal speech. This
points in the direction of 'intonational binding' as an explaining mechanism. I
was not able to confirm this hypothesis by independent recordings. Clearly,
this aspect of Muna grammar still needs much research.
The particle ndo signals explicit plurality of the noun phrase which it
modifies. Usually the head noun is animate. The form a-ndo is also found, but
how it differs from ndo is not clear. Ndo is found modifying the following
nominal types:
1. Single names. The resulting construction means 'X cum suis, X and her
friend(s), companion(s)'.
(189) do-kala-mo tora ndo Wa Mbaruisa 'Wa Mbaruisa and her sister
3pR-go-PF again PLUR ART Mbaruisa went again'
2. Coordinate names. Here ndo redundantly marks the phrase as plural; it does
not mean: 'A + B cum suis'.
5 THE NOMINAL PHRASE 109
se-ghonu liwu
one-CLAS land
In addition to ndo anagha 'those (plural)', the form ndo danagha is also found.
In isolation, danagha does not occur.
The particle ndo can only be suffixed with the plural suffix -hi:
{ (Numeral + Classifier) } N
{ (Quantifier) }
110 A GRAMMAR OF THE MUNA LANGUAGE
The measure phrase usually precedes the noun it modifies, but it may also
follow. There is a subtle difference between a pre-nominal and a post-nominal
measure phrase. In the case of a pre-nominal measure phrase the numeral is
stressed, whereas in a post-nominal construction it is the nominal that is
stressed:
Measure phrases cannot be expanded, except by the adverb labhi 'more'. This
labhi can be added to a measure phrase, but it can also form the basis of a ko-
derivation which may or may not be verbally inflected:
The lower numerals (one to nine) have three variants, which can be called the
free form, the prefixed form and the reduplicated form, as shown in the
following chart:
5 THE NOMINAL PHRASE 111
The free forms are only used for counting, that is, they can only occur in
isolation, when they do not follow or precede a noun.
The only exception to these rules are ompulu 'ten' instead of the expected but
non-existent *se-fulu, and moghono, which already means 'one hundred' and
cannot be prefixed by se-: *se-moghono, but raa-moghono 'two hundred'.
1,000 se-riwu
2,000 raa-riwu
10,000 se-lasa
1,000,000 se-juta
se-lila 'innumerable, without number'
but not: *raa-lila
As shown above, the prefixed forms occur before units of ten, a hundred, a
thousand, ten thousand, and a million. They are also found before classifiers
and measure nouns:
Examples of the reduplicated forms in counting after units of ten have been
given above. The main use of the reduplicated numerals is as the base of
inflected verbs (a-class, no um-form), when the referents are human. When the
number is higher than ten, the complex numerals illustrated above are employed:
When these numeral verbs are used attributively, the active participle form of
the verb is usually employed, but the inflected form is also found:
The participle form is also found in the idiomatic expression bhasitie ompulu
rua-fulu-no 'the extended family' (lit. 'the ten twenty relatives').
Reduplicated numerals are also found in temporal phrases which are translated
literally from Indonesian. This usage is frowned upon by many speakers of Muna:
Ordinals do not exist in Muna. For 'first' either bhaa-bhaano is used (which is
not a numeral but probably a participle) or the adverb paka-paka. It is not
clear how the ordinal concept of 'second' 'third' and so on is expressed in
Muna.
Nor do fractions exist, apart from se-tanga 'a half' and se-wunta 'a half';
'a year and a half' is se-taghu se-tanga.
Finally, mention must be made of the increasing use of the Indonesian num-
erals in Muna everyday speech. Some of these forms are partially adapted to the
phonological shape of Muna; thus ampa 'four' (< empat) and ana 'six' (< enam).
The following derivations of the numerals exist. The first three are closely
connected:
These forms can only occur in combination with units of ten and the word -
gha 'day' (a bound form of gholeo 'day'). There is one irregularity: the
bound form of two is not raa- in this case but -fua.
2. The prefixed numerals can form the base of a verb which functions as a
temporal adverb. These verbs belong to the ae-class and have a defective
paradigm in that they are only inflected for the third person singular. For
past reference ('x days ago') the suffix -mo is added to the realis form,
for future reference ('in x days') the irrealis form is employed. Again -fua
is the bound form of 'two' in this series. With siua 'nine' as base, the
word -gha is needed.
When periods of ten or more days are constructed in this way, the forms
under (1) are employed, with past reference suffixed by -mo and with future
reference prefixed by na-.
4. The suffix -ghoo is added to free or reduplicated forms meaning 'in the x-th
place':
These forms are abbreviations of longer forms: ntopisa 'first cousin', topendua
'second cousin' and topentolu 'third cousin'. In everyday speech these longer
forms are rarely used.
1. Prefix na-, indicating futurity (10.2.32). This prefix must be used with the
measure phrase when the verb is in the irrealis indicating futurity:
(208) nae-late ne ini na-raa-taghu 'he will live here for two years'
3sI-live loc this FUT-two-year
(209) nae-late ne ini ta-na-se-ahadhi 'she will live here another week'
3sI-live loc this TA-FUT-one-week
4. Prefix ti- (variant te-), meaning 'as much/many as', used in concessive
clauses (9.14 and 10.2.50):
When this derivation is used attributively with a noun, the ko-part need not be
present:
7. Prefix fe-, meaning 'divide into x; do the action denoted by the verb x
times' (10.2.4):
5.7.4. Classifiers
Wua and ghonu are the most commonly used classifiers; they are also found with
new objects that do not fit neatly into one of the other categories, for
example se-wua sentere 'a flashlight', but also: se-ghonu sentere. The dividing
line between wua and ghonu is not always clear; size is probably the
distinguishing factor: ghonu is used for either small round objects or really
big objects, wua takes the objects in between:
The reduplicated numeral se-ise 'one' can also function in a measure phrase, as
in se-ise ka-nea 'a/one custom'. In such cases the reduplicated numeral is
probably re-analysed as the prefixed numeral se- plus a classifier ise. It is
found with certain non-concrete objects: kanea 'custom', panaki 'illness',
ilimiu '(magical) knowledge', ponamisi 'feeling', and seems to be limited to
the numerals 'one' and 'two' (raa-ise). The derivation ko-ra-ise-ha-e 'both of
them' is also found.
Measure nouns constitute an open class. They differ from other nouns by virtue
of the fact that they can be immediately preceded by the prefixed numerals.
This is not the case with common nouns, which have to be pre-modified by a
classifier in such cases. Measure nouns also differ from classifiers;
classifiers come in between the numeral and the head noun, whereas measure
nouns are the head of the measure phrase themselves; they do not further modify
another noun.
So far the following measure nouns have been found; they can be grouped in a
number of semantic categories:
1. parts, sections
3. extents of time
gholeo 'day'
alo 'night'
wula 'month'
kamentae 'morning'
taghu 'year'
ahadhi 'week'
bhara 'rainy season'
tanda gholeo 'dry season'
4. lengths
rofa 'fathom'
dhari 'arm's length'
tuda 'span'
siku 'cubit'
inere 'body length'
5. areas
6. weights
7. volumes
kasopa 'bowl'
piri 'plate'
kopo 'handful'
kadu 'bag'
ghobho 'bundle'
tonde 'glass'
rupia 'rupiah'
ringgi '2 1/2 rupiah'
see '1/10 rupiah, 10 cents'
doi '1/100 rupiah, 1 cent'
kepe '1/1000 rupiah, 1/10 cent'
gobha '2.5 cents'
tibha-tali '25 cents'
tali '30 cents'
bhoka '240 cents' (now generally 24,000 rp)
120 A GRAMMAR OF THE MUNA LANGUAGE
9. metric terms
10. others
5.7.6. Quantifiers
In Muna there are only six quantifiers. They will be summed up and illustrated
in this section. In 5.7.7 quantification will be discussed that is not ex-
pressed by overt quantifiers.
The six quantifiers are:
eano 'every'
sabhara 'all kinds of, each and every'
malingu 'every, each'
bhari-bhari-e 'all'
keseno 'all; only'
sanea 'all; only'
Quantifiers precede the nouns they modify, with the exception of keseno and
sanea, which may also 'float' in the clause. I will discuss the quantifiers one
by one.
When eano modifies a noun and a locative relative clause, this relative
clause may also precede the noun. Normally relative clauses follow the noun:
This fronting is not possible when the relative clause consists of more than
one word:
5 THE NOMINAL PHRASE 121
When eano and a relative clause are part of a prepositional phrase, eano has to
follow the head noun:
Eano can also be used as a conjunction meaning 'every time' (see 9.7.2).
There does not seem to be a difference between eano and malingu. The difference
between these two words and sabhara is that sabhara stresses the diversity of
the objects, not just their individuality.
4. bhari-bhari-e 'all':
When the head noun is human, the final -e can be replaced by -ndo. The use of -
ndo makes the human reference explicit; this is left unspecified when the form
with -e is used:
1. a. bhari-bhari-e
b. bhari-bhari-ndo
2. a. ko-si-bhari-e
b. ko-si-bhari-ndo
3. a. ko-si-bhari-ha-e
b. ko-si-bhari-ha-ndo
4. a. ko-si-bhari-bhari-e
b. ko-si-bhari-bhari-ndo
5. a. ko-si-bhari-bhari-ha-e
b. ko-si-bhari-bhari-ha-ndo
The difference between the (a) and the (b) forms has been explained above. The
forms without -ha can only be used attributively; forms with -ha can either
modify a head noun or function as the head of an NP. The non-reduplicated forms
are not commonly used. The most frequently found forms are (5a) and (5b): ko-
si-bhari-bhari-ha-e and ko-si-bhari-bhari-ha-ndo.
The vowels of ko-si- are sometimes metathesized to form ki-so-bhari-bhari-ha-
ndo.
Bhari-bhari-e can also be followed by an object pronominal suffix, in which
case the final -e is lost: bhari-bhari-kaeta-amu 'all of you (polite)'.
Keseno is probably related to the nouns kese and kese-keseno 'dry, without
side dish':
5.7.7. Quantification
This section discusses how quantificational concepts such as 'few, some, many,
none' are expressed. This issue is worth investigating, as there are only a few
overt quantifiers (most of which mean 'all') and the question arises how other
quantificational concepts are realized in the language. The following concepts
will be dealt with in this section: 'none, nothing, nobody, a little, a few,
some, many, much, all, every'.
a. sendai 'little'. This word can only be used attributively with mass
nouns, not with count nouns. It is often the head of an NP. Finally, it
may be used predicatively in an equative clause:
(247) kurusi-mani sendai kaawu 'we have only got a few chairs;
chair-our little only our chairs are only a few'
b. sehae 'how much, how many' when used rhetorically means 'not much, a
little':
3. 'some'
a. the noun sigaa 'part' is used as a measure phrase, either before or after
the head noun:
4. 'many, much'
This notion is expressed by the verb bhari 'to be much/many', which belongs
to the a-class:
When the referent of 'many' is human, the verb may receive plural
inflection:
(255) do-bhari pata l[um]ulusu-no 'there were many who did not
3pR-many NEG pass-A.PART pass (the exam)'
5. 'all, everything'
6. 'each, every'
7. 'whole'
Normally a relative clause modifies a head noun (5.9 and 9.1), but headless
relative clauses (sometimes referred to as 'free relatives') also occur. In
such a case the relative clause becomes the head of the NP. There are several
types of relative clauses, but in this section only the subject relative clause
(formed with the active participle) and the object relative clause (formed with
the passive participle) will be illustrated. The terms 'subject' and 'object'
in subject and object relative clauses refer to the function of the relativized
element (participle or noun) in the relative clause, not to the function of the
relative clause in the main clause. For a full discussion of relative clause
formation, see 9.1.
(267) [m]ande-no wamba Wolio no-bhari 'there are many who know Wolio'
know-A.PART language Wolio 3sR-many
5.9. Attributes
1. relative clause (see 9.1 for the various types of relative clauses):
These adverbs can sometimes even float within an NP, where they end up between
other constituents:
8. appositive noun phrase, which is co-referential with the main noun phrase.
An appositive noun phrase constitutes a new nominal phrase which is outside
the main noun phrase:
In Chapter 3 it was made clear that the category 'adjective' cannot be main-
tained for Muna, since supposed adjectives behave like verbs. In this section
the two strategies for the formation of descriptive attributes will be
illustrated and contrasted. The term 'descriptive attributes' is meant to cover
words referring to such concepts as size, shape, age, quality and so on.
Predicatively such concepts are realized as verbs, but attributively there are
two possibilities:
1. the active participle form of the verb is used, either full or clipped (see
below):
There is a subtle difference between the (a) and the (b) forms in (304) and
(305), as is also suggested by the English translations. In the (a) examples
the noun phrase is presented as a unity that can take any position in the
sentence. The (b) examples, on the other hand, present salient information
5 THE NOMINAL PHRASE 131
about the head noun and the expectation is that what is said in the remainder
of the sentence is related to the emphasized quality of that noun. In (308) and
(309) the two noun phrases of (304) have been made part of a sentence which
illustrates the difference:
Reversing the two attributes here would result in very stilted sentences: mo-
rubu-no is the natural choice in (308), since the rest of the sentence is
clearly related to the size of the stone, which is not the case in (309).
The following is a non-exhaustive list of stative intransitive verbs that can
be nominalized for attributive purposes:
1. colour terms:
2. others:
When two descriptive attributes modify the head noun, one of them is a ka-
nominalization, the other a participle. The order is fixed: the participle has
to follow the ka-nominalization:
Alternatively, both verbs are nominalized (when this is lexically possible, see
above), in which colour has to precede size:
Such double descriptive attributes are, however, rather unnatural. The most
natural way of expressing such concepts is by making the second attribute a
separate appositional noun phrase, marked by the article o:
An apposition is found outside the NP proper and therefore always follows the
other constituents.
Examples of combinations of attributes and other modifying elements will be
given below. Nouns are frequently modified by one or two modifiers, but a
combination of three modifiers is already much rarer, whereas an example of
four modifiers was only found once. N signals the head noun.
a. N - NP - RelCL - Dem
b. N - ka-NOM - RelCL
c. N - Temp - AppCl
d. N - RelCl - AppCl
e. Quan - N - NP - RelCl
f. N - NP - RelCl - Dem
g. N - NP - PrepPhr - Temp
h. Quan - N - NP - RelCl
The following example was made up for the purpose of expanding the number of
modifiers; it is rather artificial but considered grammatical. In this example
there are two relative clauses, making a total of five modifiers:
A number of modifying categories can be filled twice. This has already been
illustrated in (334) with two relative clauses. Another example of two relative
clauses is (335). Also, a case of two quantifiers within one NP was found (see
338).
a. When two relative clauses are found in a single NP the object relative
clause (marked on the verb by ka- or ne-/ni-) precedes the subject relative
clause (marked by the active participle):
5 THE NOMINAL PHRASE 135
But if the object relative clause consists of more than one word it may also
follow the subject relative:
1. Poss - ka-Nom
2. Temp - RelCl
3. RelCl- AppCl
4. Relative clauses may precede the head noun. This is a marked order (see
9.1):
The three local prepositions are we, te and ne. All three can indicate both
location and direction. Their distinguishing components are:
1. Relative height. Te is used for locations and directions that are higher
than the point of orientation; we and ne are unmarked for height.
2. Specificness. Ne is used for specific locations and directions and with
certain verbs denoting specific actions. We and te are unmarked for this
meaning.
The meaning of te is more specific than we. The following usages can be
distinguished:
b. *na-k[um]ala we molo
Relative height plays an important role in Muna, not only in the use of
demonstratives such as watu and tatu and the prepositions we and te, but
also in the choice of verbs of motion. When a certain location is clearly
higher than the point of orientation, the verb kala 'go' is not normally
used; rather foni 'go up' is the correct verb, as in (11). Alternatively,
sampu 'go down' is the usual expression when the goal of the movement is
lower than the point of orientation, as in (12):
b. It is used for places that are located to the east of Muna. Sometimes
places in the north also take te:
6 THE PREPOSITIONAL PHRASE 139
b. ne watu 'there'
loc that
c. ne hamai? 'where?'
loc where
b. It is the local preposition used before names and certain nouns modified
by a referential demonstrative:
c. It is used for animate recipients, source, path and also for agents in
nominalized relative clauses marked by ka-/-ghoo (see 9.1.3):
It is not clear what these verbs or clauses share semantically. Possibly the
action as a whole is a specific action directed towards one object, which,
although it is not marked as definite, is still taken to be specific by the
hearer in the sense that only one referent is meant. If this is true, it can
be linked with the 'specific' usuage of ne illustrated under (a) - (d).
There are four non-local prepositions: bhe 'with', so 'for', ampa 'until' and
peda 'like'. These prepositions can all be suffixed by -mo.
The basic meaning of bhe is 'with', but its use is extended to other areas
beyond the traditional realm of prepositions. The following uses can be
distinguished:
(37) de-late-mo bhe ina wee-ndo 'they lived with their step-
3pR-live-PF with mother step-their mother'
(41) na-t[um]aburi bhe sau awatu 'it will also press down that
3sI-press with tree that tree'
This 'inclusive' usage following such verbs as owa 'bring' and ampe
'bring up' can best be translated as 'along':
When the noun after bhe is reduplicated, this indicates that the
inclusion of this noun is unexpected:
b. Coordination of NPs and verbs. This use of bhe is closely connected with
bhe as a conjunction marking simultaneity (9.6):
(50) ne-gholi pae bhe kambulu 'she bought rice and vegetables'
3sR-buy rice with vegetables
c. Comparison:
(53) ka-kodoho-no bhe tehi se-kilo 'it is one kilometre from the
NOM-far-POS with sea one-km sea'
(55) do-po-ghawa-mo bhe anahi amaitu 'they met (with) the child'
3pR-REC-get-PF with child that
6 THE PREPOSITIONAL PHRASE 143
(61) miina bhe doi ka-rubu-a 'I do not have any change'
not be money NOM-small-CL
There remain a few cases where the function of bhe is not perfectly clear,
for example in (64), where it is optional:
a. Beneficiary:
b. Purpose:
The meaning of ampa is 'until, as far as, as long as'. Although it is a non-
local preposition, it may have a locative meaning in addition to a temporal
function.
a-[m]eka-mate-ko
1sI-CAUS-die-you
'If you have not seen (found) those seven buffaloes, I will kill
you'
Peda differs from the other prepositions in that it can receive subject
inflection (a-class), but only for the third person singular. An active
participle can also be formed:
In addition to these three basic non-local prepositions, there are also two
deverbal prepositions in Muna:
These words are either inflected as verbs, or they occur uninflected, in which
case they function as deverbal prepositions:
(84) lapasi i-tolu do-po-ghawa tora 'after three days they met again'
after I-three 3pR-REC-get again
(85) a-pake sau aini mbali katuko 'I use this stick as a walking
1sR-use wood this as cane stick'
146 A GRAMMAR OF THE MUNA LANGUAGE
(86) a-pake sau aini so katuko 'I use this stick for a walking
1sR-use wood this for cane stick'
In (85) the implication is that the piece of wood is already used as a walking
stick, whereas (86) states the intention to do so. Sometimes the two co-occur:
(87) de-tando sau mbali so lambu-do 'they cut wood for their houses
3pR-cut wood for for house-their (= to build their houses)'
Illustrated in clauses:
(89) tae-late ne wiwi-no tehi 'we (ex) live beside the sea'
1eR-live loc side-POS sea
(90) ama-ku ne-ngkora ne soriri-ku 'my father sat down beside me'
father-my 3sR-sit loc side-my
probably depending on whether the speaker takes the noun phrase as given or
specific:
(91) do-fewule-mo we/ne ghowa-no foo 'they rested under a mango tree'
3pR-rest-PF loc bottom-POS mango
The following examples are taken from one text; all three prepositions are
used:
The form welo 'in, inside' is a reduced form of the complex preposition we
lalo. It is found in locative and temporal functions, as well as preceding
subordinate clauses, and glossed as 'in':
(94) no-rato welo timbu 'he came during the east monsoon'
3sR-arrive in east
(98) no-foni telo lambu 'he went (up) into the house'
3sR-go.up in house
Chapter 7
The clause
When a subject NP is present, it may either precede or follow the verb. I take
the order Verb-Subject (VS) to be the unmarked order for the following reasons:
VS 107 69%
SV 48 31%
It seems likely, however, that these prepositional phrases have been moved
out of their position and are therefore 'right-dislocated'. These states of
affairs are presented as afterthoughts and in spoken language this may (but
need not) be reflected in an intonation break between the rest of the clause
and the PP.
(48) a. o-k[um]ala ne hamai itu bhela? 'where are you going, friend?'
2sI-go loc where that friend
b. Topic establishment
karumbu
woods
'once upon a time there was a little monkey who lived near the
woods'
t[um]anggo-da
(A.PART-)support-them
c. Topic shift
This is the first time the cock is mentioned (hence the indefinite
article in the translation), but as the story is located around a house,
the presence of domestic animals such as cocks is inferentially given.
'when he saw other people with their axes going to clear the
forest, he went along as well'
7 THE CLAUSE 157
d. Result or conclusion
e. Climax
At climactic points in the discourse, there is often a tendency for the main
participant to be fully expressed and to be put in preverbal position. In
Muna the conjunction kansuru 'at once' is often added to give the climactic
action extra force. This kansuru can also be inflected as an ae-verb (see
9.21, example (301)). The following examples are all taken from the
climactic point in the narrative:
Thus, syntactic and pragmatic considerations are shown to have an effect on the
place of the subject in relation to the verb. It is important to realize that
especially pragmatic factors relate to the speaker's intent to mark something
as, for instance, topic of contrast. In other words, the speaker is free to
choose a particular strategy, but he is not compelled to do so. This explains
the VS order in those cases where there is a marked contrast with another
entity, so that preposing the subject to become 'topic of contrast' is
certainly a possibility. Compare the following example, where there is a
contrast but the second subject is still in postverbal position:
This whole area needs much more research. The syntactic factors have to be
carefully checked, operational definitions have to be found for the pragmatic
factors and quantitative studies must show the extent of variation.
The closest equivalent to the existential verb 'to be' in English is the
irregular verb naando. It is irregular in that it is not usually inflected in
the realis. The verb is typically used in clauses introducing the main
participant of a story, in which case the order is always VS, followed by
another clause which is a description of that participant. The following
examples are all story-openers:
The verb naando is uninflected in all these examples, but the subject marker
no- can be added freely. In the irrealis, however, the verb has to be
inflected:
c. *miina n[um]aando
At this place I will also treat some other uses of naando. The first of these
is that naando may be object-inflected: the direct object pronominal suffixes
are added to the base naando, in which case the addition of subject markers is
ungrammatical (see 4.8.1) The meaning of an object-inflected naando is the same
as with subject inflection, but the object-inflected forms are used either when
the subject of the clause precedes the verb or when the subject has first or
second person reference.
In the second place naando may precede another verb, in which case the meaning
is not existential but signals a progressive aspect. In some instances it may
even be translated by 'still' or 'while':
In addition to the verb naando, the word bhe can also be used in existential
clauses. The preposition bhe has already been discussed in 6.2.1, where its
usage was shown to be broader than the other prepositions.
In existential clauses bhe seems to function as a verb, but since it is not
and cannot be inflected, we can not call it a verb. I will use the term
'existential preposition' to refer to this bhe, which is glossed as 'be'.
Examples of bhe in positive clauses:
More usual, however, is the occurrence of bhe in negative clauses (see 8.5.1):
(93) miina-ho bhe bhirita ampa aitu 'so far there is not any news
not.yet be news until now yet'
(97) suru gholeo miina bhe sampu 'all day long there was no
long day not be go.down coming down; he did not come
down'
Notice that there is no overt subject expressed in these clauses; the context
normally makes this sufficiently clear. The semantic difference between such a
root form and a ka-/-ha-nominalization (10.2.18) is not entirely clear.
Finally, naando and bhe may also co-occur in a clause:
After illustrating the normal transitive word order, with full and pronominal
NPs, I will discuss and illustrate the following points:
- zero objects;
- object agreement;
- object focus: OV, VS, OVS;
- subject postposing: VOS;
- subject + object postposing: VSO.
3. VO transitive clauses:
(112) o katogha ne-mbolaku kenta topa 'the crow stole dried fish'
ART crow 3sR-steal fish dry
ama-mani
father-our(ex)
'my younger siblings still did not know our (ex) father's face'
that the direct object is present at some point. In any case the suffix is
always retrievable from the context and can be inserted without any change in
meaning; it only makes the object explicit.
'she took a stone and threw it into the middle of the heap of dead
leaves'
'after he had caught him, he pulled out his feathers and put him
under a log'
In the following example the verb fumaa is used in exactly the same way; again
the suffix -e may replace the implicit object. There is no definiteness shift
involved here, as fumaa is already an a-verb.
The suffix -e 'agrees' with the full direct object and is therefore more or
less redundant.
Object agreement is frequently found with -Cao-verbs (although strictly
speaking these verbs require morphologically indirect objects, see 4.8.2):
In most of the cases of object agreement the direct object is a known entity
that is supplied for the sake of clarification, almost as an afterthought.
Since there is no break in the intonation pattern, however, it is not likely to
be a case of right-dislocation.
In the following examples, the direct object has been mentioned before:
7 THE CLAUSE 165
(125) no-wora-e kaawu foo amaitu-ini 'after he had seen the mango'
3sR-see-it after mango that
However, in the following examples the object is not a known entity, and hence
a functional explanation cannot be given. It is possible that a system of
object agreement is gradually coming into existence, on the analogy of the
subject agreement system, in which a full NP (either nominal or pronominal)
agrees with the subject markers. But this must remain a hypothesis.
In the unmarked SVO order the information is presented from the perspective of
the subject. It is also possible, however, to take the object as point of
departure and to background the subject. This is done by using OVS word order
or, when either S or O is not a full NP, OV or V-eS. When the object precedes
the verb, it is customary to find a pronominal suffix on the verb that agrees
with the fronted object. The actual structure of most of the examples is
therefore OV-eS. I will discuss OV, V-eS and OVS in turn.
1. OV clauses
2. V-eS clauses
isa-hi-hi-no
older.sibling-PLUR-PLUR-his
'his sick child was lying down surrounded by its older brothers
and sisters'
The fact that it is possible for the subject of a transitive clause to follow
the verb may result in seemingly ambiguous clauses, in which it is not clear
whether the nominal constituent following the verb is the object (with
pronominal agreement on the verb) or the subject. Compare the following
example:
7 THE CLAUSE 167
This ambiguity does not exist in spoken language, since the intonation patterns
of the two clauses differ:
----------------------------------
_________ --_
- _
----------------------------------
(140) a. no.fu.ma.a.e.mo ke.nta 'she was eaten by the fish'
----------------------------------
_____-- --_
- _ -
----------------------------------
b. no.fu.ma.a.e.mo ke.nta 'she ate the fish'
3. OVS clauses
we tehi Sampuru
loc sea Sampuru
'Wa Ndai Pitu has been thrown into the Sampuru sea by her older
sisters'
In a number of clauses the subject follows the verb and the object. This seems
to be the case for the following two reasons:
168 A GRAMMAR OF THE MUNA LANGUAGE
(145) o-[m]ala hae-no-no itu hintu? 'which part of it will YOU take?'
2sI-take what-its-its that you
(146) a-[m]ala ghole-no ini idi-a 'I will take the top'
1sI-take top-its this I-CL
These and the following verbs functioning in experiential clauses can also
occur as intransitive verbs, in which case the subject marks the experiencer.
There is no difference in meaning:
There is one clause type of which the base is a verb that can take two
arguments: a subject and a complement. The second argument cannot be called
170 A GRAMMAR OF THE MUNA LANGUAGE
Examples:
The order is not necessarily S-V-Complement; the subject may also follow the
rest of the clause:
Equative clauses are verbless clauses that consist of a nominal subject and a
complement, which can be an NP or a PP. Derivations involving the prefix sa-
(10.2.44) and sa-/-ha (10.2.45) also function as complements in equative
clauses. Semantically, equative clauses usually make statements about the
identity or the location of the subject. Here, 'subject' should be understood
as the known or given entity in the discourse (for example pronouns,
demonstratives, nouns modified by possessive suffixes). The complement contains
new information about this subject.
The usual order in an equative clause is Subject-Complement, although the
reverse order is also found. The following four types can therefore be
distinguished (sa- and sa-/-ha-derivations are grouped with NPs):
7 THE CLAUSE 171
a. NP-subject, NP-complement;
b. NP-complement, NP-subject;
c. NP-subject, PP-complement;
d. PP-complement, NP-subject.
a. NP-subject, NP-complement
b. NP-complement, NP-subject
When the complement precedes the subject, this is because it is a question word
or because it is emphasized:
c. NP-subject, PP-complement
d. PP-complement, NP-subject
(189) soo-mo bara-ndo andoa watu 'those goods were for THEM'
for-PF goods-their they that (lit. 'that was for THEIR
goods')
(193) ka-mbaka-no kenta topa! 'how nice that dried fish is!'
NOM-nice-POS fish dry
(200) aitu maka anahi ini ka-tehi-no 'the child was very afraid'
that EMPH child this NOM-afraid-his
With such fronted NPs it is even possible to leave out the possessive linker
-no:
174 A GRAMMAR OF THE MUNA LANGUAGE
(208) ka-mbaka-no kalei bhe ka-wehi! 'how nice that banana was and
NOM-nice-POS banana with NOM-full how full (I am)'
7.8. Fronting
(209a) is a neutral statement which does not presuppose any previous discourse
knowledge. (209b), on the other hand, presents pae as the established topic of
discourse, about which something is said, namely that its stalk is weak. The
corresponding clause type in Indonesian is termed 'topic-comment sentence' by
Macdonald (1976). Other examples:
This fronting of a dependent NP can even occur when the NP is 'carried' over
two verbs:
no-fumaa-e bhake-no
3sR-eat-it fruit-its
'in forty days he had not yet finished eating the fruit of that maize'
A basic clause can be further extended with non-nuclear arguments that give
extra information about the state of affairs described in the basic predicate.
Some of these non-nuclear arguments refer to place and time of the action, to
be discussed in 7.10. In this section I will deal with those arguments that can
be expressed through indirect object pronominal suffixes as discussed in 4.8.
These arguments will therefore be called indirect objects (IO), but note that
this is a syntactic notion. Semantically an indirect object may express such
functions as beneficiary and recipient, but also instrument and reason.
In the following discussion, subdivisions are made according to the full
nominal or pronominal suffix character of both direct object (DO) and IO.
When the IO is a full NP, the verb is suffixed with -ghoo, a suffix that
signals the presence of an IO (see 10.2.10). Examples of indirect objects on
176 A GRAMMAR OF THE MUNA LANGUAGE
a. Beneficiary:
(216) do-dhoa-ghoo mate-no ini 'they pray for the dead man'
3pR-pray-IO die-A.PART this
b. Recipient:
c. Instrument:
d. Reason:
e. Referent:
f. Comitative:
(231) miina dae-rato-ghoo fonoti 'they did not come home with
not 3pI-arrive-IO shellfish shellfish'
Comitatives are also found after the preposition bhe 'with' (see 6.2.1).
There are a few other constructions in which an extra argument on the verb
is signalled by the suffix -ghoo. The difference between these and the examples
in the preceding section is that these arguments cannot be replaced by the
IO pronominal suffixes. For these indirect objects I use the term 'oblique
178 A GRAMMAR OF THE MUNA LANGUAGE
a. When the extra argument is the reflexive/emphatic noun wuto 'self', which is
obligatorily possessed:
In this last example the oblique indirect object is a clause which stands in
an appositional relationship to the noun bhirita (see 9.2.4).
c. After numeral verbs; the oblique object signals one of the included members
of the set, not an extra member:
robhine amaitu
woman that
'with that woman included, the village chief has ten wives'
With transitive verbs it has to be noted that the definiteness shift is not
triggered by a definite IO (verb remains in the ae-class), but only by a
definite DO (verb shifts to the a-class):
The IO pronominal suffixes are also employed to signal the patient ('semantic
direct object') followed by a full indirect object. In other words, to express
the combination of the DO suffix -e + -ghoo, the IO pronominal suffix -ane is
used. It is crucial to distinguish this -ane from -ane in the preceding
section, since in the latter case the definiteness shift is triggered. The DO
is definite, even if it is expressed in the form of an IO suffix. In the
interlinear translation I will use the gloss 'it/IO', 'you/IO' for this use of
the IO suffixes to indicate its double function. Examples:
In this last case it is the definiteness shift that shows the (indirect object)
function of the noun rea; compare (257) with (258), where rea is direct object:
When the DO and the IO are both pronominal suffixes, they can only co-occur
when the DO suffix is -e:
This last example is surprising, since the IO suffix -anda clearly refers to
the patient in the context. Apparently the combination of IO pronominal
suffixes and the DO suffix -e allows for a wider interpretation, possibly
because this is the only ordering that is allowed. Thus -anda-e (and not *-da-
ane or *-ane-da) can mean 'it for/with them' and 'they for/with it'. For other
examples and the shortening of -ane-e to -ane, see 4.8.
In (267a) -ane only refers to the indirect object, whereas in (267b) -ane
refers to the preverbal direct object, while at the same time it marks the
presence of a full IO NP.
As is the case with DO (see 7.3.3), indirect objects may also show object
agreement. In other words, an IO pronominal suffix can be followed by a full
IO, usually for reasons of clarification:
(269) no-lili-ane kenta anagha 'he went round with that fish'
3sR-go.round-it fish that
In certain cases, the full indirect object is put in a PP, in which case the
pronominal suffix is still an indirect object, but the full NP is no longer so:
Finally, similar to DO zero marking, there may be indirect object zero marking
(suppression) when the IO referent is clearly established in the discourse.
However, the verb is still suffixed with -ghoo in these cases:
In this section we will be concerned with the clause periphery, that is, those
elements that provide extra information on the state of affairs designated by
the clause core and that are not syntactically indirect objects. A number of
peripheries are introduced through prepositions and have already been
illustrated in Chapter 6. Here I will concentrate on three periphery types:
time, location and manner.
182 A GRAMMAR OF THE MUNA LANGUAGE
The temporal periphery answers the question 'when?' or 'how long?' in relation
to the clause core. The temporal periphery may be filled by a number of
different words and phrases, all of which I will call 'temporal' in this
section. The following temporals may be distinguished (the list is not
exhaustive):
2. noun phrase:
As far as the place of the temporal periphery is concerned, the usual position
is clause-initial:
(283) miina dao-maa dua se-gholeo 'one day we will not eat'
not 1pI-eat also one-day
Certain temporals may also occupy the position between the preverbal NP (either
subject or object) and the verb:
Not all temporals can occur in every possible position, however. The following
five temporals were tried out in different positions in a clause:
These five temporals were inserted in the numbered positions in the following
clause:
1 2 3 4 5
isa-ku ne-gholi roti we daoa
brother-my 3sR-buy bread loc market
'my brother buys/bought bread in the market'
1 2 3 4 5
aniini x x ? x x
indewi x x * x x
tolu wula kaomponano x x * () x
sadhia x x * () x
tantigho * x * * *
x = possible
* = impossible
() = possible, but change of meaning (the temporal is seen to be part of
the locative; for example, three months residence in the market)
? = questionable
This suggests that temporals may belong to different subclasses defined on the
basis of their possible position in the clause. Such a subclassification,
however, still remains to be worked out.
In Chapter 6 the prepositional phrase was analysed and the various prepo-
sitions, including the locative prepositions, were discussed and illustrated.
Here I will only discuss the locative periphery in terms of its position
7 THE CLAUSE 185
Occasionally it may be found in initial position. This may be the case when a
location is introduced in an existential clause or for reasons of emphasis or
surprise:
As can be seen, the manner adverbial normally follows the verb, but in the
imperative it may also precede the verb.
Frequently, such a manner adverbial is inflected like a verb, that is, it is
prefixed by the subject markers from class a-. The resulting construction is
again a juxtaposition of a manner verb to a main verb:
Although combinations of peripheries are not very frequent within one clause,
especially at the beginning of a text or episode one may encounter both
temporal and locative peripheries. The following examples show a few such
clauses and also combinations of time and place with other peripheries such as
beneficiary (with so) and comitative (with bhe).
wamba Inggirisi
language English
'at that time there were very few teachers of English here in Raha'
'this man lived with his wife and their five children
on the edge of the forest near the sea'
7.11. Adverbs
In Chapter 3 adverbs were defined as a closed class of words that do not form
the head of a phrase, that cannot take prefixes or suffixes and that can serve
a variety of purposes. On the basis of their meaning, several subtypes of
adverbs can be distinguished. The primary division, following Quirk and
Greenbaum (1973), is that between adjuncts and disjuncts. Adjuncts are to some
extent integrated into the structure of the clause, whereas disjuncts are
semantically outside the clause; they convey the speaker's comment on or
evaluation of the content of the communication.
7.11.1. Adjuncts
Types of adjuncts:
indewi 'yesterday'
kansuru 'continually'
tantigho 'all the time'
sebantara 'in a short time'
ndai-ndaino 'in a while'
sepaliha 'very'
daano 'indeed'
mpuu 'indeed, really'
kotughu 'true, real'
5. Focusing adverbs:
These also have to follow the nominal or verbal head to which they belong.
Within the clause, focusing adverbs follow the verb immediately; they must
precede a postverbal subject or object:
(323) dae-ngkora deki ne ini 'we will sit here for a while'
1pI-sit first loc this
7.11.2. Disjuncts
Disjuncts are only loosely connected with the meaning of the clause; they
indicate the speaker's or narrator's point of view on the content of the
clause:
(334) miina da-ko-diu kaasi-ndo 'they did not move, the poor
not 3pI-HAVE-move pity-their things'
In this section I want to have a short look at a number of other elements that
function at the clause periphery and that cannot be subsumed under other
headings. These elements are vocatives, interjections and ideophones.
7.12.1. Vocatives
1. Proper names
2. Kinship terms:
ama 'father'
ina 'mother'
awa 'grandparent'
7.12.2. Interjections
(such as pain, fear, amazement). Second, there are interjections used for
animals, either to call them or to chase them away.
The following list is a sample of the most common emotive interjections in
Muna, the intonation on which they are normally uttered and the emotions they
convey.
(344) ee, noafa o-angka-ghoo ne ini? 'hey, why are you going this
hey, why 2sR-go-PURP loc this way?
(351) me-ngkora ne ini konae 'why don't you sit down here'
IMP-sit loc this -
Certain words that belong to other word classes can be used as interjections.
Examples:
7.12.3. Ideophones
In narrative texts the Muna language makes little use of ideophones. The only
examples found are words expressing the sounds made by animals or instruments:
There are, however, many words that form the basis of a verbal derivation with
ko- which have the meaning 'to make the sound of'. There are probably scores of
these onomatopoeic verbs and the following list is a small sample of these:
Clausal modifications
8.1. Transitivizing
d. *ae-ghumu-ti-e
196 A GRAMMAR OF THE MUNA LANGUAGE
d. *a-late-ghi-e
When the underived verb is an ao-verb, the derived Ci-verb is still an ao-verb,
and hence no definiteness shift is involved:
a-leni 'swim'
ae-leni-fi 'swim to get something'
a-kala 'go'
ae-kala-hi 'go to (somebody)'
a-mai 'come'
ae-mai-hi 'come to (somebody), visit'
Examples in clauses:
8.2. Causatives
8.2.1. Factitives
The causee is expressed as a direct object, as shown by the use of the direct
object pronominal suffixes. Definite objects trigger the definiteness shift:
locutional causative). Only in a few cases have I tried to show these ranges of
meaning in the translation.
In this case the causee is also expressed as a direct object, which may trigger
the definiteness shift. Examples in clauses:
(24) miina da-[m]o-sampu-e ne hale 'they did not put her down on
not 3pI-CAUS-go.down-her loc floor the floor'
When a causative verb is formed on the basis of a transitive verb, the prefix
fo- is used again, and the derived verb belongs to the ae-class:
It is more usual to find these constructions with object agreement, that is,
the causee is expressed in a pronominal suffix and the full causee NP is put in
clause-final position:
b. *a-fo-fumaa-ko-e
The fourth causative type in Muna is different from the others. It is formed
only on the basis of transitive verbs, marked by the prefix fe- (10.2.4), which
draws all these derivations into the ae-class. Its meaning is more specific
than a fo-causative. A fe-causative implies that the causer has interacted
verbally with the causee with the purpose of having him or her perform an
action for the benefit of the causer. Hence the name 'locutional causative'.
The normal translation of a locutional causative is therefore 'ask, request,
command to do X for the benefit of the subject'. The degree of politeness and
compulsion (request versus order) is left unspecified.
b. *a-fe-buri-e
In the examples above it is not always easy to translate the component 'for
the benefit of the causer' properly, but this meaning aspect is clearly
present. In (37), for example, the implied meaning is that it is the causer's
202 A GRAMMAR OF THE MUNA LANGUAGE
In (47) -angko signals the beneficiary, whereas in (49) -angko refers to the
causee. In (47) and in (48) the meaning component 'for the benefit of the
causer', however hard to translate, is still present.
The following examples of causative stacking were found in texts, and others
were elicited from speakers of the language. The latter case is marked as such.
2. fe- + feka-:
4. fe- + fo-:
To end this section I will give two causative paradigms on transitive verbs:
a-fumaa 'eat'
ae-fo-fumaa 'feed'
ae-fe-fumaa 1.'order to eat'
2.'ask to be eaten'
ae-ada 'borrow'
ae-fo-ada 'lend'
ae-fe-ada 1.'order to borrow'
2.'ask to be borrowed'
8.3. Detransitivizing
It is very striking that Muna should have another prefix fo- (10.2.7) which is
valency-reducing. To distinguish it from the causative fo- I will call it fo2-.
It is attached to transitive bases and has the effect of replacing the object.
204 A GRAMMAR OF THE MUNA LANGUAGE
Apart from its syntactic behaviour, fo2- differs from the causative fo- (fo1)
morphologically:
2. The irrealis allomorph of fo1- is [m]o-, but fo2- does not change:
(59) lapasi aitu do-fo-walu-mo 'after that came the putting into
after that 3pR-DETR-shroud-PF a shroud'
The resulting predicate with fo2- cannot be extended with a direct object,
neither as a full noun nor as a pronominal suffix:
b. *no-fo-sia mie
3sR-DETR-bite people
c. *no-fo-sia-e
3sR-DETR-bite-him
8 CLAUSAL MODIFICATIONS 205
Notice that in (61) the verb sampu is made transitive by means of the suffix
-niki and detransitivized by the fo2- prefix.
It is not unusual to find combinations of the detransitivizing fo- and the
causative prefixes feka- or fo-:
In combination with the causative prefix, the predicate may be extended with an
indirect object (marked by -ghoo) indicating the instrument or the means of the
action:
Finally, the verb fo-guru 'teach' must be mentioned here, since it appears to
be exceptional. Fo-guru is related to po-guru 'learn' and to guru 'teacher'; it
is syntactically a causative verb, in which the causee (the pupil) is expressed
in an indirect object and the causand (the subject taught) in a direct object:
When the causee is a pronominal suffix and the causand a full noun, the causee
may also be realized as a direct object pronominal suffix for reasons that are
not clear (compare also examples (34) and (35) in 8.2.3):
When the causee is not mentioned, the detransitivizing fo- is used. Notice,
however, that fo- replaces an indirect object in this case, and not the usual
direct object:
When the suffix -ghoo is added, it may signal the subject taught or the means
of instruction:
206 A GRAMMAR OF THE MUNA LANGUAGE
This indirect object may be pronominalized, and the causee (the pupils) may re-
emerge in a prepositional phrase:
8.4. Reciprocals
(79) do-suli we lambu do-po-feena-mo 'when they (2) came home they
3pR-return loc house 3pR-REC-ask-PF asked each other'
clause and the unmarked non-singular is that in this case the action is viewed
from one actant only, although it is clear that the action is reciprocal:
(85) no-po-semba bhe isa-no 'he and his brother kicked each
3sR-REC-kick with older.sibling-his other'
8.5. Negation
These negators will be illustrated one by one in this section, with the
exception of ko/koe/koise, which is discussed in 8.7.4. Other words involving a
negative meaning component are treated in 8.5.5.
When a verbal clause referring to the past or present is negated, the word
miina is put before the verb. The effects on the verb are twofold:
1. The verb form changes from realis to irrealis (see 4.5). It is impossible
for miina to be followed by a realis verb form. Note, however, that the
irrealis in this case still refers to the past or present. Only the first
verb after the negator is affected by this irrealis shift.
2. The verb is often (but not always) suffixed with the clitic -a.
(88) miina na-[m]ooli-a no-foni-si-e-a 'he was not able to climb it'
not 3sI-able-CL 3sR-go.up-TR-it-CL
the negation. This last usage is relatively rare, possibly because of its
ambiguity (does the irrealis point towards negation or the future?).
no-ko-katoo-ghoo ndoke
3sR-HAVE-partner-IO monkey
'so the girl does not have you (pl) as her marriage partner, her
marriage partner is a monkey'
A number of intensifying and focusing adverbs may modify the negator miina, in
which case they occupy a position between the negator and the verb:
(93) miina tora bhe mo-bhalo-e 'again there was nobody who
not again be A.PART-answer-him answered him'
(95) na-[m]oni te lani, miina dua 'he has not gone to the sky
3sI-go.up loc sky not also either'
(96) miina bhe ka-lele-ha dua miina 'there was no place to cross
not be NOM-cross-LOC also not either'
In existential clauses with bhe, miina is also the regular negator. The noun
following bhe can be a regular noun (97) or a nominal derivation as in (96) and
(98), a deverbal noun (root form) as in (99), a reduplicated noun as in (100)
or a ka-/-ha-derivation marking emphasis as in (101):
(97) miina bhe doi ka-rubu-ku-a 'I do not have any change'
not be money NOM-small-my-CL
This bhe may also be found at the beginning of a verbal clause, while miina
occurs in clause-final position. Such a construction denotes contra-expectation
and can be adequately translated with 'even':
The suffixes -mo and -ho can be added to miina, resulting in miina-mo 'no more'
and miina-ho 'not yet':
Finally, like the existential verb naando, the negator miina can also receive
object-inflection, that is, it functions as a kind of verb of which the subject
is expressed in direct object pronominal suffixes (see 4.8.1).
When a verbal clause referring to the future is negated, the negator pa or pae
is used, which seem to have no difference in meaning. Occasionally pe is found,
possibly a dialectal form. Verbs following these negators are also in the
irrealis form. With vowel-initial verb forms pa may be cliticized to the verb:
pa-o- > p-o- (see 3.1.4).
210 A GRAMMAR OF THE MUNA LANGUAGE
The suffixes -mo and -ho can only be added to the form pae, not to pa.
(117) ane pae ao-limpu, lima taghu-mo 'if I have not forgotten (it),
if FUT.not 1sI-forget five year-PF it was five years ago'
(118) ane paise, dae-karadhaa hae? 'if not, why should one work?'
if FUT.not 3pI-work what
(124) do-bhari pata l[um]ulusu-no 'there were many who did not pass
3pR-many NEG pass.exam-A.PART the exam'
Certain focusing adverbs may intervene between the negator and the verb form:
The negator suano (dialectally also soano) is used for constituent negation,
that is, before an NP. Again, -mo and -ho may be added, though this is less
common than with miina. The clitic -a is frequently attached to the last
element of the NP which is being negated:
(134) aini suano medha-a, tamaka kurusi 'this is not a table, but a
this not table-CL but chair chair'
(137) suano kaawu inodi, do-bhari 'not just me, there were many
not just I 1pR-many of us'
In addition to these five negators with their variants and derivations, there
are a few other words that have a negative meaning component: the emphatic
negative adverb mau-na and the negative verbs ghuri and kiido.
(138) no-tola, mau-na bhe mie-mie 'he called, but there was
3sR-call not be RED-people absolutely nobody'
2. The negative verb ghuri means 'absolutely not' and may either precede or
follow the negated clause, of which the verb must be in the irrealis. This
verb ghuri only occurs in the third person singular realis:
3. The negative verb kiido 'not want, refuse' is also followed by the irrealis:
Raising of negative words from the subclause to the main clause ('I don't think
it is raining') is ungrammatical in the great majority of examples. Only one
example was elicited where the scope of the negator is possibly the subclause:
----------------------------
_ _ _ _ _ -
- - - -
----------------------------
(146) de-aso ghunteli ghole itu? 'do they sell eggs today?'
3pR-sell egg day that
--------------------
_ _ _ _ -
- -
--------------------
(147) o robhine isa-mu? 'is your older sibling female?'
ART woman older.sibling-your
Other examples:
(149) miina ome-salo tulumi we sigaahano? 'do you not ask others for
not 2sI-ask help loc other help?'
The second type of yes-no question is a leading question, in which only the
nucleus is spoken on a high tone followed by a mid tone; the speaker expects a
positive answer from the hearer, he wants to make sure he has heard it
correctly or drawn the right conclusion:
-------------------
- _
- - - - -
-------------------
(151) no-kala we sikola? 'has he gone to school?'
3sR-go loc school 'he has gone to school, hasn't
he?'
----------------------
- _
- - - - -
---------------------
(152) isa-mu o robhine? 'your older sibling is female,
o.s.-your ART woman isn't she?'
Other examples:
Answers to yes-no questions (of both types) can be single-word clauses such as
uumbe 'yes', miina 'no', paise 'no (will not)', bhahi 'possibly' and beane or
bhaiane 'who knows'. It is also possible to repeat the verb in a positive
answer:
The question tag ela (variant elae) may be added to a yes-no question when the
speaker is fairly certain that the answer will be positive. It is a feature of
informal spoken Muna:
(159) o-k[um]ala we sikola maka paise? 'are you going to school or not?'
2sI-go loc school then FUT.not
(160) o-ghawa-e-mo itu maka miina-ho? 'have you got it or not yet?'
2sR-get-it-PF that then not.yet
Content questions differ from yes-no questions in that specific question words
are used that ask for information. In this section I will treat the Muna
question words one by one, giving illustrations of regular and aberrant usages.
As for the intonation, the stressed syllable of the question word is normally
spoken on a high tone. Notice that in the majority of cases the question word
occupies the same position as the corresponding non-questioned constituent in a
statement.
The following question words can be distinguished:
Hae may fill the position of any noun phrase: subject, direct and indirect
object, an NP within a prepositional phrase and a dependent NP within another
NP. In addition it may be used attributively, in which case it means 'which'.
When hae functions as the subject of the clause, it is preceded by the article
o:
(161) o-fumaa hae itu ghane? 'what are you eating, boy?'
2sR-eat what that boy
(164) omo-ghae-fi hae itu bhela? 'what are you crying for,
2sR-cry-TR what that friend friend?
(167) o-[m]esua ne sikola hae itu? 'which school did you go to?'
2sI-enter loc school what that
When hae follows the suffix -ghoo it often refers to purpose, hence its
translation as a why-question. In these examples -ghoo can be analysed both as
a purpose suffix and as an indirect object suffix (see 7.9, 9.17, and 10.2.10):
The possessive suffix -no can be added to hae with the resulting meaning of
'what part?' It is usually suffixed by another possessive suffix indicating the
object of which the questioned constituent is part:
(172) o-[m]ala hae-no-no itu hintu? 'which part will you take?'
2sI-take what-its-its that you
8 CLAUSAL MODIFICATIONS 217
But o hae-no is also found in simple questions asking for the identity of a
specific entity:
Hae-no followed by a possessive suffix may also occur with the preposition so
'for', in which case it means 'to what purpose?', 'what use is that for X?' It
often functions as a rhetorical question:
Lahae (or the shortened form ahae) is made up of the question word hae and the
article la. It may fill the position of a questioned human NP and in addition
it can also be used attributively. Although la is the masculine article, lahae
can be used for both sexes; the word *wahae does not exist.
(184) o-waa-ane lahae boku amaitu? 'to whom did you give that book?'
2sR-give-him who book that
When lahae is the subject of a verbal clause, the verb form is very frequently
an active participle. Again, this corresponds to a cleft construction:
(189) o-butu te ndo lahae? 'with whom (pl) are you staying?'
2sR-stay loc PLUR who
Like o hae-no, lahae-no asks for the identity of a more specific entity than
lahae:
Sehae can possibly be analysed as consisting of the question word hae and
the prefixed numeral se- 'one', although the meaning of the combination is
not predictable from the meaning of the parts. It is the question word
8 CLAUSAL MODIFICATIONS 219
asking for amounts, numbers, extents and so on. When the referent is human,
sehae is inflected as a verb from the a-class. When the reference is to a
future state or action, the prefix na- is attached to sehae, as it is to
numerals.
(195) sehae litere pae ka-gholi-mu? 'how much rice have you bought?'
how.much liter rice NOM-buy-your
Used in rhetorical questions, sehae may also mean 'not much' (see 5.7.7).
In addition hamai can be used as a quantifier meaning 'every' (see 5.7.7) and
as an adverb meaning 'moreover':
These complex question words, consisting of the preposition peda 'like' and the
question words hae and hamai, are used for questioning means, manner, state and
quality. An alternative form, peda hamadi, is occasionally found. Especially in
combination with hae and hamai, peda is often inflected like a verb (see
6.2.4). The irrealis form na-meda is frequently shortened to nada.
The active participle [m]eda-no hae can be used attributively or as the head of
an NP and means 'which; what kind of', either referring to a choice between
alternatives or to the quality of the questioned noun:
Peda hae can also take the function of a rhetorical question which is closely
connected to a following statement. It is the speaker's comment on his previous
utterance, which is 'natural' and 'understandable':
8 CLAUSAL MODIFICATIONS 221
These three question words all mean 'when', but nefiemo and indefie can only
refer to a point in the past, whereas naefie refers to a future state or
action:
In Muna there are two interrogative verbs on the basis of the root afa. With
ae-inflection this root means 'do what?' The verb may be suffixed with the
indirect object pronominal suffixes, in which case the verb shifts
(irregularly) to the a-class, but the um-form is not necessarily employed:
When afa is in the ao-class, the verb has a different meaning: 'what is the
matter with?' In the third person singular it can usually be translated by
'why' (which is consequently done in the interlinear). Verbs following this
noafa are often but not always suffixed with the purpose suffix -ghoo. This is
also the case with the (inflected) preposition peda:
(233) noafa miina o-[m]ewa-ngi-ane-ghoo? 'why did you not resist him?'
why not 2sI-resist-TR-him-PURP
(234) noafa miina o-s[um]epa-ane-ghoo? 'why did you not kick him?'
why not 2sI-kick-him-PURP
8 CLAUSAL MODIFICATIONS 223
There are two adverbs in Muna, bhahi and bhara, both meaning 'perhaps, maybe,
possibly' that occur very frequently in interrogative clauses, so that they can
almost be termed question adverbs. Their function is to reinforce the question
character of the clause by expressing wonder or doubt (compare Indonesian
gerangan). These adverbs also occur in non-interrogative clauses as regular
adverbs. Examples:
pande-hao 'know'. It is not uncommon to find the adverb bhahi preceding the
embedded question. Although this bhahi seems to fill a kind of complementizer
position, it is an adverb indicating doubt and uncertainty.
For further discussion of juxtaposition and conjoining, see 9.2 and 9.3.
Echo questions are questions that are repeated by the hearer; for example 'Am I
interested in music?' as a response to the question 'Are you interested in
music?' Echo questions do exist in Muna, but they are relatively uncommon. When
the hearer is not sure whether he has heard the question correctly or is
extremely surprised at the question, he will usually answer o hae 'what?' or
peda hae 'how?', possibly with partial repetition of the question. These
answers are also found in reaction to unintelligible statements and commands.
The following examples are elicited material and possibly less natural:
The imperative mode is used to make orders to the hearer. This mode is
characterized by a special non-finite verb form (the imperative) and the
frequent use of a number of affixes and adverbs that serve to increase or
decrease the force or degree of politeness of the command.
As was shown in 4.9, the imperative is different for each of the three verb
classes in the following way:
class a: root
class ae: me- + root
class ao: mo- + root
Examples:
When the order is directed to more than one hearer, the second person plural
suffix -Vmu is added to the imperative. It follows all other suffixes.
With transitive verbs the same imperative forms are used. With definite direct
objects, the definiteness shift occurs.
When the speaker wishes to make a polite command he may choose the second
person polite subject markers. With the ae-class, the irrealis prefix tae-
seems to be fairly common, next to the realis te-. The suffix -mo (see below)
is normally added to this polite imperative:
Also, the 'unfamiliar' second person subject markers do- and de- may be used in
the imperative mode to address a hearer whose status and position is not yet
known to the hearer (see 4.1):
The second person free pronouns hintu 'you' and hintu-umu 'you (pl)' may be
added to imperatives. Depending on intonation and voice quality this conveys
either a very forceful command or merely contrastive emphasis ('not he, but
you'). These pronouns may either precede or follow the verb.
The imperative mode in Muna can be modified in certain ways to fit the
situation and to convey the speaker's emotion. Obviously, intonation and
intensity of speaking are extremely important as well. I will focus, however,
on the verbal character of the imperative here.
The following is a list of affixes (or affix combinations) and adverbs that
may modify imperatives. Since imperatives in English only allow for a small
range of formal modification, it is hard to get the impact of the imperatives
in the translation. In a number of cases, I have therefore provided Dutch
equivalents.
1. Suffix -mo
The suffix -mo (10.2.29) usually softens the command when spoken with a
neutral intonation. The command is not presented as an order, but almost
8 CLAUSAL MODIFICATIONS 227
When the command is said on a high pitch, the order may be more forceful. There
is an urgency which is absent in the previous examples:
The combination of ta- (10.2.48) and -mo results in a very weak imperative; the
command is almost presented as a recommendation or as if the speaker gives
permission. Dutch maar is a good equivalent of this affix combination:
3. Suffix -kaeta
When the speaker adds the demonstrative itu (see 5.5.6) to a command, it
signals mild irritation and impatience on his part. The hearer may not have
performed the action after the first command and the speaker has to remind him
again:
When the imperative form of the verb soba 'try' (alternant form insoba) is
added to a command, it may convey two things:
2. On a high pitch, soba may also convey mild irritation and impatience:
(282) insoba fetingke wamba-ku ini 'why don't you listen to what
try (IMP-)listen word-my this I say?' (Dutch: luister nou toch
eens)
The adverb pada 'already' signals an emphatic command which is not impolite,
nor is there impatience. The speaker is strongly convinced of the necessity of
the action being performed:
As can be seen in the first example, the second person subject markers are
occasionally retained. With polite negative imperatives this is always the
case:
Ko, koe and koe-mo can also be followed by non-second-person verbs, in which
case the meaning is 'let not'. The verb can be either in the realis or in the
irrealis:
The adhortative mode in Muna only applies to first person plural. The
adhortative expression mai-mo 'come on' often precedes adhortations:
Another way of forming the adhortative mode is by means of the purpose suffix
-ghoo. The difference between -ghoo and -ana/-mana seems to be that the
-ana/-mana adhortative suggests more readiness or willingness on the part of
the hearer to perform the action than with -ghoo. In the -ghoo adhortative the
action is presented almost as a purpose, which is not supposed to be performed
at once:
The Sentence
In this chapter I will deal with sentences, structures that involve more than
one simple clause. Traditionally a basic distinction is made between
coordination and subordination. It seems that such a distinction is difficult
to maintain for Muna, since there is only one clause type that is formally
marked as dependent, namely relative clauses. No other complex structures make
use of formal means (such as word order or special verb forms) to indicate
subordination. A difference that is made, however, is the distinction between
clauses that are juxtaposed (no conjunction, no intonation break) and clauses
that are conjoined. On the basis of the conjunctions used, conjoined clauses
can be further subdivided into, for example, contrastive, conditional and
temporal conjoined clauses.
These three linking types (subordination, juxtaposition and conjoining) will
be discussed in turn in 9.1, 9.2 and 9.3 - 9.19. In 9.20 direct and indirect
speech are discussed. The chapter ends with sections on the suffix -mo in
narrative discourse (9.21) and the clitic -a (9.22).
When the relativized NP is the subject of the dependent clause, the active
participle form of the verb is used. Plain active participles have the
232 A GRAMMAR OF THE MUNA LANGUAGE
a-class: um-form-/-no
ae-class: me-/-no
ao-class: mo-/-no
In certain examples the final suffix -no of the active participle may
optionally be left out (see 5.9.2). This is especially the case with the verb
mai-ghoo 'come from'. The suffix -no can always be added, however:
9 THE SENTENCE 233
When the active participle is suffixed with a direct object pronominal suffix
(-kanau, -ko, -e and so on, see 4.8.1), this suffix replaces -no:
When numeral verbs are used attributively, the active participle form is also
frequently used (see also 5.7.2):
When there is more than one verb in the relativized clause, all of them are in
the participle form; so and pata, if present, can occur only once:
Occasionally a relative clause precedes its head. This is not very common, and
in most cases a quantifier is part of the domain noun (see also 5.7.6). That,
however, is not a condition:
The active participle is also used when the subject of the relative clause
functions as a dependent ('possessed') NP within the main clause. The head noun
may keep or lose the possessive linker -no.
When the relativized NP is the direct object of the dependent clause, the most
common strategy is to use the passive participle. The verb takes the prefix ne-
(variant ni-, see 10.2.33) and possessive suffixes indicate the agent (see 4.9
and 5.4.1). Such an agent need not be present in general statements. Again, the
negator is pata, while so marks the future:
A full nominal agent, if present, follows the third person singular suffix -no:
When the relativized NP is the indirect object of the dependent clause, the
passive participle is suffixed by the indirect object suffix -ghoo:
subject X - -
direct object - X X
indirect object - X-ghoo -
A quantifier may draw the locative noun to a pre-head position (see also 5.7.6
and example (19) in 9.1.1), but a post-nominal position is more common:
A relative clause may stand on its own without modifying a head noun. Such free
or 'headless' relatives are common when the understood head is 'people,
somebody' or when it is fully retrievable from the context, as in (47):
9 THE SENTENCE 237
(44)tamaka miina bhe [m]ondo-i-no 'but there was nobody who could
but not be able-TR-A.PART do it'
(45) do-bhari pata l[um]ulusu-no 'there were many who did not pass
3pR-many NEG pass-A.PART (the exam)'
mo-ngkubu-no
-short-A.PART
'he called all his fellow crocodiles. There were big ones and small
ones, there were long ones and short ones'
Free relatives are also found with passive participles without accompanying
head noun:
(49) garaa ini naando ne-ghongko-no 'but there was something she was
SURPR this be P.PART-suck-her sucking on'
The first question is more natural and more forceful, since the questioned
constituent is a complement in itself. Other examples:
(54) anoa kaawu [m]omaa-no 'he is the only one who has
he only eat-A.PART eaten'
9.2. Juxtaposition
Using the semantic relationship between the two clauses as a basis, the
following division can be made.
isa-hi-hi-no
older.sibling-PLUR-PLUR-his
'his child was still lying (there) surrounded by his older brothers
and sisters'
ka-pudhi
NOM-praise
'while living in this world one must not be led astray by praise'
(66) no-tende dahu no-hamba katogha 'the dog ran chasing the crow'
3sR-run dog 3sR-chase crow
The verb mai-ghoo 'come from', when juxtaposed to a motion verb, may lose
its subject marker and become almost preposition-like:
d. The juxtaposed clause consists of a numeral verb, also agreeing with the
first verb in subject and mood:
(85) no-wule-mo dahu no-lumpa 'the dog was tired from running'
3sR-tired-PF dog 3sR-run
1. Physical perception:
wora 'see'
fetingke 'hear'
namisi 'feel'
9 THE SENTENCE 243
4. Others:
With kona 'think, suppose' the juxtaposed clause as a whole may be cross-
referenced by -e 'it' on the verb:
(99) a-kona-e ome-gholi ghunteli 'I thought you had bought eggs'
1sR-think-it 2sR-buy egg
Certain clauses are directly juxtaposed to NPs. A primary division must be made
between independent and dependent clauses, based on the relation of the clause
to the NP.
Independent clauses function as the predicate of an equative clause of which
the subject is an NP. The NP and the juxtaposed clause are on the same level;
this is a special type of the equative clause. The subject NPs of such equative
clauses belong to a limited set of obligatorily possessed nouns indicating
speech or mental states and processes and a few others such as karadhaa 'work':
Some of these nouns may also follow the juxtaposed clause. This is especially
common with amba-no 'his word; he said' (see 9.20). In one example the subject-
noun namisi 'feeling' is found inside the predicate clause, which itself is
also complex:
se-mbali-ha-no wata
one-side-HA-POS log
Other examples are also found, in which the 'relativized' noun is the
subject, the object or a dependent NP within the juxtaposed clause. There is
no intonation break between the head noun and the clause:
Finally, a juxtaposed relative clause may follow the noun sabhabu-no 'the
reason'; the conjunction sampe 'until, so that' may introduce the clause:
d. Others, in which the semantic relationship between the head noun and the
juxtaposed clause varies, although some sort of dependency is involved:
When two clauses are conjoined without overt conjunction, the semantic
relationship shows a considerable variety (contrast, reason, clarification and
so on). Usually the two clauses show some sort of balanced structure, or, in
the case of clarification, the second clause is a partial repetition of the
first. In balanced structures, parts of the second clause may be 'gapped', but
the conditions for and the extent of gapping have not been investigated in
detail.
Finally, it should be mentioned that it is not always easy to decide whether
clauses that look like conjoining without conjunctions are in fact two
different sentences. In other words, whether a slight pause is a comma or a
full stop. In the following examples I assume the pause signals a break between
clauses:
sa-robhi-robhine-hi
ONLY-RED-woman-PLUR
The temporal adverb (participle?) bhaa-bhaano 'at first, for the first time'
occurs in such temporal clauses with the meaning 'when first'. The verb in the
second clause is not marked by -mo:
248 A GRAMMAR OF THE MUNA LANGUAGE
Temporal sequence can also refer to a point in the future. In that case Muna
does not make a distinction between a temporal and a conditional clause. The
first clause may be introduced by ane 'if', but need not be:
9.5. Alternative
(132) o-ghawa-e-mo itu maka miina-ho? 'have you got it or not yet?'
2sR-get-it-PF that then not.yet
For other uses of maka, see 9.7.2. The conjunction atawa (probably a borrowing
from Indonesian atau 'or') is also used for alternative conjoining.
9.6. Simultaneous
The conjunction bhe 'with' (also preposition, see 6.2.1) links two clauses that
occur simultaneously. The subjects of the two clauses must be the same and the
verb in the second clause cannot be suffixed with -mo.
9.7. Temporal
Temporal clauses allow for a wide variety of marking. Some of the linking
between the clauses is taken care of by affixes and some by conjunctions,
either dependent or free conjunctions. These three strategies will be discussed
and illustrated in turn. In most of the examples that follow, the clause
containing the temporally preceding action precedes the main activity clause.
That first clause is never marked with the perfective suffix -mo, whereas the
second clause is, except after such words as kansuru 'right away' (also when
inflected) and maka 'then' (see 9.21).
1. When the events denoted by the verbs in the two clauses occur in rapid
succession, the prefix sa- (10.2.44) is affixed to the verb in the first
clause, which is either nominally or verbally inflected. The subject of that
clause, if present, normally follows that verb. This rapid succession is
equivalent in meaning to the English construction 'as soon as'. When
followed by a vowel, sa- can be reduced to s-.
This construction is also possible for future time reference, in which only
the verbal (irrealis) inflection is used:
ta-sonso-sonso-ghoo-mo aitu
JUST-RED-follow-IO-PF that
'as soon as you hear (people) play the drum and play the gong, follow
that'
2. The prefix paka- (10.2.37) 'when first, when just', on verbal bases with
nominal or verbal inflection indicates a temporal relationship between two
states, or durative events in which the state/event in the main clause takes
place during the initial stages of the state denoted in the clause
containing paka-.
This prefix is related to the temporal adverb paka-paka 'at first; ini-
tially'.
3. The prefix ta- (10.2.48) on verbal bases indicates an 'until' clause. The
verb is obligatorily suffixed with -mo, while the suffix -hi also frequently
emerges. Clauses containing such ta-derivations may either precede or follow
the main clause.
1. One of the most common ways to mark temporal succession is through the use
of the conjunction kaawu 'after'. In a temporal clause containing kaawu the
verb must take the initial position in the clause, immediately followed by
kaawu. If a full subject is present, it must follow this kaawu. Again, the
verb in the temporal clause cannot be, but the verb in the main clause must
be, suffixed by -mo (except after the adverb kansuru).
kalangka-no ini
basket-her this
'when she got home, she immediately threw her basket away'
2. Temporal sequence is also marked by the conjunction maka 'then'. The verb in
the clause following maka cannot be suffixed with -mo:
(153) na-r[um]ato kapala maka na-[m]ere 'not until the ship arrives
3sI-arrive ship then 3sI-leave will he leave'
(154) nae-fua maka na-s[um]uli 'he will only return the day
3sI-two then 3sI-return after tomorrow'
In such clauses referring to the future, the temporal clause preceding the
conjunction maka may be marked with the futurity suffix -ho (10.2.13). This
-ho may be suffixed to verbs, nouns and adverbs. Verbs must be in the
realis:
(157) we ghabu-ho maka ao-maa 'I won't eat until later in the
loc kitchen-FUT then 1sI-eat kitchen'
3. The conjunction eano 'every time'. The second clause is not marked by -mo
because of the recurrent action:
6. The negator miina-ho 'not yet' (8.5.1) can also function as a conjunction:
'before the time of the match had come, the snail first told his
friends'
7. The nominal conjunction wakutuu 'time'. This noun (a loan from Indonesian)
can function as a conjunction, in which case it is followed by a juxtaposed
clause. Occasionally, the possessive suffix -no is added to wakutuu.
c. Ne-laa 'straight'.
kansuru do-filei
at.once 3pR-run.away
As discussed in 9.3, free conjunctions can connect both clauses and sentences.
Where sentence connections are illustrated, I will not always provide the
preceding context.
1. Kansuru 'at once, right away, immediately'. The verb following kansuru
cannot be suffixed with -mo.
9.8. Contrastive
2. Sembali (lit. 'one side'), and variants wasembali, masembali means 'except,
unless, on the contrary':
3. Tabea (variants: beano, tabeano, tabeahano) has a complex usage; it can only
partly be described as a conjunction, although that seems to be its primary
function. It links a negative clause to another clause specifying the
exception, or the conditions under which the first statement can be true
after all.
deki kapa-no
first cotton-his
'he cannot go up into the house, unless he first plants his cotton'
'but Wai-Wai refused, (except that) she wanted to return to her own
father's house'
the preceding part. In this case, it cannot be called a conjunction, and the
translation 'only; must' is appropriate.
9.9. Additive
'Bula Siku has given me the task to make a stable for his horse; in
addition I have to find him his grandfather's coin'
9.10. Surprisive
next state/event falls outside the logical and cultural rules of what can be
expected.
ta-no-si-ka-bisara-ha-mo
TA-3sR-SI-KA-speak-HA-PF
(203) garaa ini, niho-mo o-k[um]ala-mo? 'well, are you leaving only now?'
SURPR this just-PF 2sI-go-PF
raa-mata-mo
two-eye-PF
'he had said he would arrive at ten o'clock, but he came at two'
9.11. Conclusive
9.12. Clarificatory
The conjunctive noun maana-no 'that is to say' (lit. 'its meaning') may serve
to introduce a clarificatory clause:
na-ti-mo-tehi
3sI-ACC-CA-fear
9.13. Conditional
Conditional clauses, both real and unreal, are introduced by ane 'if'. Such
conditional clauses may refer to the present, the past, the future or to a
generic situation. An alternative for ane is barangka 'if'. Sometimes the two
are found together. In conditional clauses both the realis and the irrealis
mood is found. It is not yet clear which factors determine that choice.
kalei tatu
banana that
The main clause may simply consist of the question word peda hae 'how'. This
construction is used for suggestions or new ideas.
(213) peda hae ane da-t[um]angkahi deki? 'how about eating first?'
like what if 1pI-eat first
Similarly, negative real and counterfactual conditionals both take the negator
pa/paise:
nae-owa-ghoo kalangka?
3sI-bring-PURP basket
'if she were not stupid, why would she have brought a basket?'
(rhetorical question stressing the stupidity of the action)
The complex conjunction sa-pae/sa-paise means 'if not' and is equivalent to ane
pa. The main clause is sometimes introduced by the preposition ampa (6.2.3).
karada-no maitu
spear-his that
'the ruler of the village will be angry at him, if his spear is not
visible (= does not turn up)'
260 A GRAMMAR OF THE MUNA LANGUAGE
a-[m]eka-mate-ko
1sI-CAUS-die-you
'if you have not finished the stable for my horse in one day, I
will kill you'
o-ko-ana-ghoo mo-kesa-no
2sI-HAVE-child-IO -beautiful-A.PART
'now if only you let me go, I will ask God for you that you will
have a beautiful child'
9.14. Concessive
na-ti-kambisa-a?
3sI-ACC-scatter-CL
'why are those dry leaves not scattered, even though one throws
stones at them?'
When the concessive clause contains a numeral, the prefix ti- (variant te-,
10.2.50) precedes that numeral. Its function is to stress the unexpected
outcome ('as much/many as').
miina-ho na-[m]ande
not.yet 3sI-know
Finally, mahingga is followed by question words such as peda hae (peda hamadi)
'how' and ne hamai 'where' to mean 'however that be' and 'wherever':
9.15. Reason
'they planted its seeds, because this so-called wood-yam was rather
long and had many seeds inside'
'because they really wanted to have a child, they went to look for a
mango fruit'
bhangka-ndo no-kele-mo
boat-their 3sR-dry-PF
ko-wuna-no
HAVE-flower-A.PART
When the result is presented in a new sentence, with the reason preceding in
the previous sentence, the perfective suffix -mo is added to the
nominalization. If the negator pata is present, -mo is suffixed there.
9.16. Manner
ka-bhatatala-hao-no
NOM-curse-INT-his
'if one eats it one dies because of it, as is mentioned in the curse'
peda ne-mo-nea-ha-ghoo-no
like P.PART-CA-usual-HA-IO-his
'once a man went to look for his food as he was used to'
264 A GRAMMAR OF THE MUNA LANGUAGE
9.17. Purpose
Purpose clauses are marked as such by the suffix -ghoo (10.2.10) on the verb
(glossed PURP). This purpose -ghoo is different from, though possibly related
to, the indirect object suffix -ghoo. The verb is in the realis or irrealis,
depending on whether the purpose is generic or refers to a specific future
action.
When the purpose clause contains the preposition so 'for', this indicates a
close link between a noun in the main clause (not the subject) and the purpose
or the result. This noun functions as an instrument in the purpose clause:
With vowel-initial verbs, so is often cliticized to the verb, for example s-ae-
gholi-ghoo 'for me to buy' < so ae-gholi-ghoo (see 3.1.4).
This so also occurs in purpose clauses following the question words peda hae or
peda hamai 'how':
Negative purpose clauses are introduced by the conjunction koana 'lest'. There
is no -ghoo suffix on the verb, which must be in the realis mood:
9 THE SENTENCE 265
9.18. Dubitative
A dubitative clause indicates a purpose that is desired but far from certain.
The conjunction bhahi (also an adverb meaning 'perhaps'), glossed DUB
introduces such dubitative clauses. It is not always easy to adequately
translate them into English, although the older English construction
'that...may' comes very close.
9.19. Balanced
There are two types of balanced clauses. The first involves the so-called a
fortiori argument: if A is such and such, then certainly B. Both parts of
such balanced clauses are introduced by conjunctions. The first clause (the
266 A GRAMMAR OF THE MUNA LANGUAGE
minor clause) is marked by salangi, saneghoomo 'while' or ane 'if', whereas the
major clause is introduced by poolimo or poolighoomo (possibly from the verb
pooli 'can, be able'), often followed by dua 'too'.
dua inodi
also I
'if a newcomer like you can do it, how much more I myself'
inodi
I
nao-maa-e manusia
3sI-eat-it man
'when an animal eats her quid, it will die, how much more when a man
eats it'
The other type of balanced sentence is the so-called co-balanced type: 'the
more ... the more'. Both clauses are introduced by the conjunction nihompu 'the
more':
'the heavier his ears, the more he (= the rice) bows to his friends'
Direct quotations usually follow verbs of speech such as bisara 'say, speak',
feena 'ask', bhalo 'answer', but they may also precede these verbs. When the
verb of speech precedes the quote, the quotation noun amba-no 'his word' (or
amba-ku 'my word' and so on) is often found at the juncture point. It is not
obligatory. Occasionally the reduced variant mba-no is found.
9 THE SENTENCE 267
da-marasai'
1sI-suffer
ini o Wadho'
this ART Bajau
'then answered the people who had just arrived: "We are Bajau's"'
'she looked down under the house and said: "Maybe you girl", she
said, "I will propose to you", she said'
'the buffalo called him, he said: "Now", he said, "we will also
compete in kicking as hard as possible"'
'then the little old woman said: "Go home, girl, and tell your parents
that you will have no more happiness in this world"'
'the haji ordered the orphan child to go and call his teacher and to
say: "The haji in the plantation calls you"'
'then Kapitalao Lahataala said: "Go home and tell Kapitalao Dhunia:
'Kapitalao Lahataala has got a child'"'
9 THE SENTENCE 269
Notice that in (281) the speech between A (the little old woman) and B (the
girl) is direct and between B and C (the parents) indirect. The reverse is true
in (282), where the speech between B and C is also direct. Whether this
variation is always possible or whether it depends on other deictic phenomena
is not yet clear.
In narrative texts the perfective suffix -mo has its own special usage. Since
this usage can only be discussed with reference to complex sentences, it has to
be treated here rather than in the next chapter.
Primarily, -mo is suffixed to those verbs that denote a new completed state
or action. In narratives such verbs take the story one step further, and a
quick look at the use of -mo in a story reveals that the verbs suffixed with
-mo form the backbone or skeleton of the story. Hence we do not find -mo
attached to verbs in conjoined clauses or clauses providing background
information. In other words, -mo functions as a sequential or development
marker.
In certain cases -mo is optional, but the governing principles behind this
choice are not yet completely clear.
Some examples of -mo in consecutive clauses:
These examples consist of more than one sentence (boundary marked by <;>).
Sentences consisting of more than one clause have various restrictions on the
occurrence of -mo. It is not found in the following cases, even though a
completed state or action may be denoted:
(286) no-suli-mo no-bhasi ndoke 'he went home and called the
3sR-return-PF 3sR-call monkey monkey / he went home to call
the monkey'
(290) no-wule-mo dahu no-lumpa 'the dog was tired from running'
3sR-tired-PF dog 3sR-run
270 A GRAMMAR OF THE MUNA LANGUAGE
d. -mo is obligatorily absent after the conjunctions kansuru 'at once', maka
'and then', bhe 'with' and sampe 'until':
The verb a-fererene is not suffixed with -mo because it is inside a temporal
conjoined clause; a-ko-ghendu does not have -mo because it is not a
completed state. Other examples:
no-po-gunti
3sR-PO-cut
'he took his machete and hit the snake's neck until it was cut in
two'
9 THE SENTENCE 271
'in former times (lit. at the top) there lived Snail and Ant. The
parents of Snail and Ant went to clear the forest on the hills'
ka-ntara de-mboto
NOM-long 3pR-stay.awake
272 A GRAMMAR OF THE MUNA LANGUAGE
'once upon a time the jin and the mouse deer held a competition in
staying awake longest'
c. In a fast sequence of actions without conjunctions the second verb may lack
-mo:
'the old man lifted his spear, took the girl by the hand and then
they went to the place where the snake was'
The demand that the verb presents a new and completed action accounts for the
fact that -mo is not present in clauses providing background information and in
negative clauses.
Special mention must be made of clauses that are repetitions, rewordings or
further specifications of preceding clauses. Sometimes the verbs in such
clauses also take -mo, and sometimes they do not, the reasons for which are not
clear.
ka-kamo-kamokula
DIM-RED-old
'they went again to bring it; the little old woman went to bring it
again'
It does not have a clear meaning, but its usage can possibly best be described
under the following three headings.
(313) miina-ho siaghe da-bhari-a mie 'there were not yet many people'
not.yet too 3pI-many-CL person
(315) suano ka-ghosa-no pikore-a 'it was not the bird's strength'
not NOM-strong-POS k.o.bird-CL
(316) miina na-[m]ooli-a no-foni-si-e-a 'he was not able to climb it'
not 3sI-able-Cl 3sR-climb-TR-it-CL
'as for me, if I follow my desire, even until it gets light I will
not be bored chatting'
'as soon as his mother had left, A Ntaapo-apo went to play. The maize
that he had to cook he just left. When his mother returned from
working in the field, she found her child still asleep'
As was mentioned above, the clitic occurs at certain natural boundary points,
such as the following:
- at the end of a clause within a sentence;
- at the end of a preverbal temporal, locative or topic phrase;
- at the end of a preverbal subject or object;
- at the end of a clause-initial subject in an equative clause.
Some examples:
Derivational morphology
The following chart summarizes the main information about all the affixes and
affix-combinations to be discussed in 10.2. However, only the most common uses
are indicated; special uses and unproductive derivations are discussed in 10.2.
The meaning of some of the affixes is given in simple English equivalents.
Productivity of the affix is indicated by y(es) or n(o).
The categorial status of both the base and the derivation are provided. A
hyphen indicates that the derived form does not neatly fall into one of the
recognized categories (see also 3.5). When verbs are not subcategorized for
either morphology (a-, ae- or ao-class) or syntax (transitive versus
intransitive), this implies that a subdivision does not have to be made for the
affix under discussion and that the derivation has the same morphological and
syntactic features as the base.
The following non-transparent abbreviations are used:
10.2.1. -ANA
Verbal suffix, productive. With first person plural inclusive inflection the
meaning is dual adhortative (see 8.8), while only the realis occurs. With other
persons a purpose clause is indicated in which the speaker conveys an attitude
of carelessness.
10.2.2. -E
Emphatic suffix on demonstrative (see 5.5.8) and vocative bases (see 7.12.1).
The vowel is often lengthened and pronounced on a high pitch.
10.2.3. E-
10.2.4. FE-
Further derivations:
no-fe(-)aghi 'spit'
no-fe(-)tapa 'ask'
no-fe(-)ntoho 'take a rest'
no-fe(-)redaa 'hope'
e. Base is a simple noun, the prefix occurs in combination with the suffix
-Ci; the derived verb is transitive:
10 DERIVATIONAL MORPHOLOGY 281
f. Base is a simple verb, the prefix occurs in combination with the suffix
-Ci; the derived verb is transitive:
ne-fe(-)buni 'hide'
ne-fe(-)tambo 'catch/hunt using bait'
ne-fe(-)api 'count'
ne-fe(-)tingke 'hear' (compare ne-tingke-tingke 'to overhear')
10.2.5. FEKA-
b. The prefix feka- is found in combination with the suffix -hi without any
apparent change in meaning compared with 1.a:
c. Both with simple and reduplicated bases the adverb can be inflected as a
verb, in which case it belongs to the a-class, is amenable to -um- and
intransitive:
10.2.6. FO1-
c. Base is a noun:
Further derivations:
10.2.7. FO2-
With certain verbs the class affixes (-me- and -mo-) surface:
c. The verbs fumaa (variant fomaa) 'eat', foroghu 'drink' and futaa (variant
fotaa) 'laugh' possibly show a fossilized prefix fo-. If that is the case,
the prefix has changed both in function and in form from either fo1- or
fo2- since none of the verbs is causative in meaning. On the other hand,
the first two are transitive, while all three are amenable to -um-.
284 A GRAMMAR OF THE MUNA LANGUAGE
This definitely points towards an analysis in which the forms are treated
as units.
10.2.8. FOKO-
10.2.9. FOKO-/-U
10.2.10. -GHOO
a. Base is a bound verbal root. These verbs can only occur suffixed with
-ghoo or with indirect object inflection (see 4.8.2):
no-waa-ghoo 'give'
no-ghele-ghoo 'forbid'
ne-asi-ghoo 'love, like'
Further derivations:
ka-waa-ghoo 'gift'
< waa-ghoo 'give'
ka-fo-ntaa-ntaa-ghoo 'what has been prepared for (X)'
< fo-ntaa-ntaa 'prepare' < ntaa-ntaa 'ready'
2. To indicate a purpose.
10.2.11. -HA
2. The suffix -ha occurs in combination with the active participle of simple
or derived verbs (-um-/-no, me-/-no, mo-/-no). The resulting meaning is
either partitive, temporal or locative.
10.2.12. -HI
ana 'child'
ana-hi 1. a.'child' b.'children'
2.'wife'
ana-hi-hi 1.'children'
2.'wives'
3. Suffix on verbal bases, with some variety of meanings. This -hi is not to
be confused with -hi as an allomorph of -Ci (10.2.16). This usage is
productive. The following meaning aspects can be distinguished:
b. The action or state denoted by the verb is one of a longer series (for
example, an enumeration in a description); with action verbs the use of
-hi implies that more was done:
a. The prefix feka- (10.2.5); -hi does not markedly affect the meaning,
though it possibly softens the factitive somewhat:
10.2.13. -HO
10.2.14. I-
10.2.15. -CAO
The suffix -Cao (C stands for a subclass of consonants, see 2.8.4) occurs on
verbal bases and is not productive. The resulting verb is almost always
transitive (class ae-), but sometimes intransitive (class a-). The derived verb
has an intensified meaning aspect compared with the basic verb, and often
denotes a violent and vehement action. -Cao-derived verbs are formally marked
by the fact that they take indirect object inflection, in which case the final
vowel in -Cao is sometimes lost (see 4.8.2).
The following list is not exhaustive, although most of the regularly used
derivations are found here. A difficulty with these verbs is that in several
cases different meanings were recorded, which were not always confirmed by
other speakers of the language. A certain degree of idiolectal meaning
variation seems to be connected with these verbs:
ne-fende-hao 'greet'
ne-dangku-lao 'hit/shake violently'
ne-furu-tao 'move very fast; exhaust, use up'
ne-tabhi-lao 'throw (clothes) over the clothesline so that they
hang out broadly'
ne-bhatatala-hao 'curse'
Further derivations:
292 A GRAMMAR OF THE MUNA LANGUAGE
no-ti-pande-hao 'known'
< pande-hao 'know' < pande 'know, clever'
do-po-angka-tao 'they respect each other'
< angka-tao 'respect' < angka 'appoint'
ka-tumbu-lao 'pole, house post'
< tumbu-lao 'plant (firmly) in the ground; hit firmly'
< tumbu 'pound, hit
10.2.16. -CI
Verbal suffix, with a number of allomorphs. The 'binding' consonant (C) may be
zero or any consonant, except for the voiced stops and all the prenasalized
consonants (see 2.8.4). There do not seem to be any semantic criteria that
determine the choice for a particular allomorph.
c. Base is a transitive verb. This is only a very small class with some
idiosyncratic meanings.
10 DERIVATIONAL MORPHOLOGY 293
a. A few verbs have more than one allomorph with different meanings:
10.2.17. KA-
These categories are not watertight divisions; often more than one meaning
is possible on a certain base, in which case the immediate context must
provide indications as to which interpretation is correct. Consider:
Apart from their use as 'normal' heads of NPs, ka-derivations occur in some
other construction types. Since these topics are all dealt with in other
parts of this grammar, they are only briefly discussed and illustrated here.
e. ka- occurs in combination with the suffix -a (does not occur on its own):
10.2.18. KA-/-HA
b. Meaning: the reason of the activity/state denoted by the verb (see 9.15).
The noun is obligatorily possessed.
It is also possible that the forms of this last category are made up of
the prefix ka- while the base is a reduplicated verb. With the available
information, this is hard to determine.
The compound nature of this derivation is most clearly shown in the fact
that the object of the main verb can be suffixed to the compounded verb:
10.2.20. KI-/-HA
10.2.21. KO-
d. Base is a 'sound' root (most of these are bound roots). There are dozens
of these derivations, for more examples see 7.12.3.
Further derivations:
10.2.22. KO-/-HA-
10.2.23. KO + REDUPLICATION
10.2.24. -MANA
Verbal suffix on verbal bases. Meaning: adhortative (plural); see -ana (10.2.1)
for adhortative singular. For adhortatives see 8.8. The subject is always first
person inclusive. Productive.
10.2.26. MANSO-
Verbal prefix on verbal bases; a-class, blocked for -um-. Meaning: 'often,
habitually'. The derivation characterizes persons or objects. Base is a simple
intransitive verb; transitive verbs cannot be the basis of this derivation.
Productive.
Further derivations:
2. Clipped participle prefix on verbal bases. See 5.9.2 and 10.2.51 for
clipped participles:
3. Class affix, that is, an auxiliary affix that occurs only in combination
with other affixes and has no independent function or meaning. The class
affixes are found in combination with the following affixes (only one
example is given for each category). The reason why they surface and the
variation that is found are not yet accounted for.
a. fe- (10.2.4)
b. fo2- (10.2.7)
c. ka- (10.2.17)
d. ne- (10.2.33)
e. piki- (10.2.40)
f. po- (10.2.41)
g. si-/-ha (10.2.47)
h. ti- (10.2.49)
4. Miscellaneous:
10.2.29. -MO
The suffix -mo (dialectally -mu) has a wide range of applications, but a
perfective or emphatic meaning component can usually be detected.
10.2.30. -VMU
The pluralizing suffix -Vmu occurs on several bases. The V stands for an echo
vowel, the last vowel of the base which is repeated in the suffix. -Vmu is the
shape of the morpheme in the Katobu subdialect (which forms the basis of this
grammar); in Tongkuno and Kabawo the shape is -omu.
308 A GRAMMAR OF THE MUNA LANGUAGE
The combination of the perfective -mo with the pluralizing -Vmu gives the form
-Vmoo in the Katobu subdialect, but -omuo in Tongkuno and Kabawo. The Katobu
form has contracted the vowels:
10.2.32. NA-
Passive participle prefix on verbal bases. Both ne- and ni- occur in standard
Muna, though ne- is the usual form in Katobu. This ne- should not be confused
with the subject marker ne- (third person singular for ae-verbs). Agents are
optionally marked by possessive suffixes. The participle functions in headed
and headless relative clauses (see 9.1).
Class affixes are found with only one ae-verb, but regularly with ao-verbs.
These ao-verbs are intransitive, but a participle can be formed because of
the indirect object suffix -ghoo referring to cause, reason or content:
10.3.34. NGKO-
Verbal affix on verbal bases; ae-class; blocked for -um-. Meaning: 'often,
habitually'; almost identical in meaning to manso-, but ngko- occurs both on
intransitive and transitive bases. Productive.
10.2.36. PA-
Nominal prefix on verbal bases. Indicates a profession: 'one who regularly does
X for his living'. Unproductive.
10.2.37. PAKA-
Prefix on verbal bases with verbal or nominal inflection. Meaning: links two
clauses by specifying their temporal relationship as one of initial overlapping
('when first, when just'). See also 9.7.1. Productive.
b. With verbal inflection (less usual); the prefix does not change the verb
class of the verbal base:
On complex bases:
10.2.38. PARA-
2. Base is a stative verb; the prefix para- alternates with mara-; in both
cases there is nasal accretion. The meaning is idiosyncratic.
Further derivation:
10.2.39. PE-
10.2.40. PIKI-
Further derivation:
10.2.41. PO-
Idiosyncratic meaning:
b. Base is a simple verb. Meaning: 'in parts, broken'; or: 'connected'. The
reference is dual.
no-po-dea 'shout'
no-po-gau 'speak, say'
no-po-kamunti 'whisper'
no-po-gira 'fight'
no-po-lele 'bring news'; also: ne-fo-lele 'spread news'
no-po-sintuwu 'ready, prepared'; also: ne-fo-sintuwu 'finish, make
ready'
no-po-toto 'be the same'
a. With -Ci:
b. With ta- ('until') and the auxiliary suffixes -ghoo and -mo. Class
affixes surface with ao-verbs. Meaning: 'until'.
d. With -ha:
Further derivations:
b. Base is a noun:
c. Base is a noun:
10.2.44. SA-
a. Base is a simple verb. Meaning: 'to V all the time; only V'.
When the verbal base is reduplicated, the continuous aspect is even more
emphasized, bordering on 'always':
4. Miscellaneous:
10.2.45. SA-/-HA
10.2.46. SI-
10.2.47. SI-/-HA
a. Base is a dynamic intransitive verb. Only plural inflection. With ae- and
ao-verbs, class affixes surface. Non-reduplicated bases have dual
reference:
Reduplicated bases have non-dual plural reference; the class affixes are
treated as belonging to the verbal base and are taken along in the
reduplication:
a. With singular subjects (plural subjects of this form have the meaning
listed under 1):
b. The circumfix si-/-ha is found in combination with the affix ka, affixed
immediately after si-. Possibly this is an independent affix combination.
Meaning: sudden, unexpected and frightening action. Productive.
In the last two examples the second meaning is more usually expressed as
do-si-kao-ndawu-ha and do-si-kao-lodo-ha.
Further derivations on this base with ta- and -mo are common; the meaning
is still more intensified (more sudden, more surprising and more
frightening). The ambiguity for plural subjects is retained.
10.2.48. TA-
3. Base is the preposition peda 'like'; optionally further affixed with -hi
and -mo:
The reduplicated base can be prefixed with ka-; the meaning does not
change, but the derived verb is an a-verb:
10.2.49. TI-
no-ti-ulu 'lost'
no-ti-sore 'run aground'
no-ti-sele 'startled, frightened'
Further derivations:
When the participle receives direct or indirect object inflection, the object
suffix replaces -no:
The suffix -no is also absent in a number of fixed expressions (see 5.9.2). The
resulting 'clipped participles' can only function as modifiers to nouns; they
cannot head an NP, a possibility which is only open for full participles.
Examples of clipped participles:
10 DERIVATIONAL MORPHOLOGY 325
Finally, -no is also absent in constructions with the negator tapa (see 8.5.3).
10.3. Reduplication
Reduplication is a morphological process in which a base is partly or wholly
repeated. Formally, three types of reduplication can be distinguished: full
reduplication (10.3.1), partial reduplication (10.3.2) and supernumerary
reduplication (10.3.3). Notice that reduplication which obligatorily
accompanies certain prefixes (for example mba-, mpo-) is not treated here. In
such cases the affix + reduplication of the base is analysed as a unitary
process.
On derived bases:
Notice that the basis of reduplication here is the verbal root, whereas
in the examples under 1 the affix forms part of the base. The difference
is shown in the following examples:
*do-po-ha-po-hamba-hamba
2. Intensification (after bhe and miina bhe, see 6.2.1 and 8.5.1):
dhara-dhara 'pigeon'
panto-panto k.o. small gnat
moa-moa 'maleo bird'
rapo-rapo 'peanut(s)'
lame-lame 'potato(es)'
godu-godu 'pumpkin, gourd' (also kodu-kodu)
328 A GRAMMAR OF THE MUNA LANGUAGE
tula-tula 'story'
hole-hole 'cookies, cake(s)' (compare ne-hole 'to fry')
sio-sio-mo 'hopefully'
mansa-mansangea 'hopefully'
paka-paka 'initially' (paka- 'when first', 9.7.1 and 10.2.37)
panda-panda-no 'finally' (panda 'bottom')
ingke-ingke-ho 'as if'
kau-kau-no-mo 'look like'
(also kau-kau-mo)
bhaa-bhaa-no 'at first, in the beginning'
kira-kira 'about to'
Partial reduplication, that is, the repetition of the first syllable of the
root, is not a productive process. Contrary to full reduplication, the verb
remains in the class to which the basic verb belongs, and the derivation is
amenable to -um-. Partial reduplication is almost exclusively found on simple
verbal roots and denotes an ongoing, progressive aspect. The syllable that is
reduplicated receives an optional extra vowel copy in some derivations. When
verbs allow for partial reduplication, full reduplication is often also
possible, but used with varying degrees of naturalness. In the case of mpona
'long', full reduplication does not occur.
10 DERIVATIONAL MORPHOLOGY 329
This is also the case in the question word ho-hae 'whatever' (more common:
hae-hae)
ne-gho-ghora 'urinate'
no-tu-turu 'fall down (eyes), be sleepy'
Further derivation:
Supernumerary reduplication, that is, the repetition of more than two syllables
of the base, is a very limited process that only occurs productively with
personified nouns that have been affixed with the article la or a:
What is meant by compounding is the linkage between two nouns through nasal
accretion (see 2.8.2), an unproductive process that is limited to a few frozen
phrases:
Nasal accretion is also found in combination with certain prefixes, whereby the
derived noun or verb is linked to a preceding word, for example pae ngka-dea
'red rice'; no-filei ngka-pode-podea 'she ran away screaming'. These complex
structures can also be regarded as compounds.
Compounding also occurs in the following cases, where a separate word is
incorporated into a verb. This is the case when the two words form a phrase or
when the basic verb is modified. The fact that the resulting construction is in
fact one word can be shown by the affixation of -mo, of object suffixes and the
active participle suffix -no after the incorporated word. In some cases this
incorporation is accompanied by nasal accretion.
10.5. Transposition
Transposition (or zero derivation) is the process by which roots change from
one word class to another without any morphological change. The question which
word is the base and which one the derivation is a tricky one which I will not
address (compare English oil, to oil versus drink, to drink). Rather, I will
only describe the existence of such pairs.
A transpositional relationship is found between nouns and verbs, or between
verbs belonging to different verb classes on the other hand. Each category
10 DERIVATIONAL MORPHOLOGY 331
probably has a few hundred members and the following lists are therefore only
illustrative of the facts.
1. Transposition noun-verb:
b. Ae-verb - a-verb:
10.6. Abbreviation
The following six texts represent some of the prose literature of Muna; their
primary function here is to give coherent samples of language usage. The
interlinear text is followed by a free translation and a number of notes
commenting on certain aspects of the text (emendations of the original,
unexplained phenomena and so on). For ease of reference, sentences are numbered
consecutively, which is done in the free translation as well. The numbers of
the notes also refer to these sentence numbers. For the first text the notes
are rather elaborate, giving detailed information on the constructions used
with reference to the relevant sections in the grammar. This is done in order
to facilitate the understanding of the texts.
Text one is a personal story by Mustafa from Matakidi (Lawa) and was directly
written down by him. Text two is a fable (with foods as characters) written by
Siddo Thamrin from Watuputi (Katobu), taken from Yatim et al. (1984) and
checked with the author. Text three is an animal fable, written by La Mokui
from Kabawo. Text four is the origin myth of the Muna rulers as found in an
unpublished collection of folktales from Sulawesi Tenggara; the author is
unknown. Text five is a relatively long romance; it is the only oral text of
the six. The story was narrated by Wa Kandiida from Mabodo (Katobu). I have
decided not to tidy up this text; hence features of spoken language abound:
hesitations, false starts, self-corrections, seemingly useless repetitions and
so on. Text six, finally, is a folktale about the origin of a spring, as
written by Runa Ujian R. from Lawa.
334 APPENDIX
Notes
Free translation
1. Now I will first tell you about the dream that I had last night. 2. Last
night I went to sleep at twelve o'clock. 3. I had not been asleep for long when
I dreamt that my father sat down at my side and touched my hand. 4. Then he
asked me a number of questions. 5. He said:
6. 'What are you doing, my boy?'
7. 'I am just sitting down, father,' I answered.
340 APPENDIX
[m]antasi-no.
suitable-A.PART
idi ini?'
I this
dao-maa-kanau'.
3pI-eat-me
54. Pada kaawu no-pogau peda nagha no-ere-mo maka no-kala. 55. Bhe
finish after 3sR-speak like that 3sR-stand-PF then 3sR-go with
Notes
3. ne-angka-do; lit. 'the one who is appointed by them', that is, their
leader. This passive participle functions as a noun throughout this story.
The same is true of ne-fumaa 'what is eaten; food'.
5. do-po-owo-owili-mo; the reduplicated syllable has undergone vowel
assimilation (from do-po-owi-owili-mo).
7. The precise identification of some of the yams is not clear.
12. kantawu 'heap' here means 'meeting, those present'.
13. pe (here and throughout the story) is a dialectal variant of pa or pae. pe
is relatively uncommon.
20. mbadha 'body', a variant of the more common badha.
26. The function of nagha 'that' is not clear.
33. intaidi-imu can either mean 'all of us (inclusive)' or 'all of you
(polite)'. Both meanings are possible here.
35. o hae so nealaghoo, lit. 'what can be taken from' is an idiomatic
expression meaning 'what can be expected/hoped from'.
48. nao-maa is an irregular irrealis from the root fumaa 'eat'.
48. ta-na-omba-mo; the irrealis has lost a medial nasal (from ta-na-[m]omba-
mo); see 2.8.1.
51. na-h[im]iri 'will slice' is dialectal (parts of Katobu); standard Muna na-
h[um]iri, see 2.8.1.
52. ungkame refers to the complex activity of soaking the cassava pieces in
water while regularly treading and squeezing them so that the poisonous
substance is dissolved.
Free translation
1. Once upon a time there was no longer any unity among the different sorts of
food eaten by man. 2. In short, they were always quarrelling with each other.
3. The point of controversy was who was suitable to be appointed leader of the
foods; 4. in other words, to become king of the foods, to be put at the head
(of the table) when people needed something or when they had a party.
5. Well, after some time all the foods sent each other a message. 6. The
message was that a meeting would be held in order to facilitate the choosing of
the king of the foods. 7. When the time of the meeting arrived, all the
different foods gathered together: the maize, the yam, the sweet potato, the
cassava, the tuber, the rice, the wild cassava, the wild bean, in short, all
the foods were present. 8. Now among those present, the wild cassava always put
himself in the forefront. 9. He really wished that they would appoint him king
of the foods. 10. The rice, on the other hand, did not stir a limb. 11. He sat
quietly in a corner (lit. on the end of a beam).
12. When everybody was present, and there was nobody left to wait for, those
present discussed the question of who was most suitable to become the leader of
the foods, so that they would not always be quarrelling. 13. The wild cassava
said: 'It seems to me that it will not be hard to search. 14. Personally I
think we can just look around. 15. Whoever has broad leaves, a big fruit, and
is thorny and hairy, that is the one who must be appointed. 16. That is to say,
he will be loved, he will be feared.' 17. Now this wild cassava himself has
broad leaves. 18. His stalk is thorny and hairy, and his fruit in the earth is
big. 19. Somebody else answered: 'Now I do not know about the others, but for
me personally, it is not the body, not the stature that we have to look at. 20.
Someone may have a beautiful body, a handsome figure, a sturdy stature, but if
his character is not good, if his disposition is bad, 21. and furthermore, if
his mind is dull, if he is rather stupid, if he is only looking for praise,
well then obviously it is senseless to appoint him as leader of the foods. 22.
INTERLINEAR TEXTS 347
Tamaka miina bhe ka-ndawu dua kenta topa anagha, rampano no-sia-e
but not be NOM-fall also fish dry that because 3sR-bite-it
no-hamba anoa.
3sR-chase he
Notes
Free translation
1. Once a crow stole a dried fish. 2. The dried fish that he stole was not too
big. 3. He was therefore able to fly with it. 4. While he was flying with it,
he was seen by a dog. 5. The dog said to himself: 'What a delicious dried fish
that crow has; if only it were for me.' 6. When he had spoken thus he
immediately started to run, following the route of the crow. 7. Whenever he
alighted in a tree top, the dog guarded him below. 8. When the crow flew again,
the dog ran after him, following the route the crow was taking. 9. He really
wished that the dried fish would fall from the crow's beak. 10. But the dried
fish never fell, because the crow held it very firmly in his beak while he flew
as fast as possible. 11. The faster the crow flew, the faster the dog ran. 12.
After this had lasted for some time, the crow knew that he was being chased by
the dog.
13. After he had flown for a long time the crow got tired, and the dog was
also tired of running. 14. So the crow landed on a rather high tree; the dog
also stopped, his tongue hanging out with exhaustion (lit. his tongue about to
be pulled out with exhaustion). 15. He rested and looked up at the crow while
he wondered how he could get the dried fish from that crow. 16. Very soon he
had an idea. 17. The dog spoke loudly: 'What a handsome crow is this, what
shiny feathers he has got, I think his voice must also be very beautiful when
he sings.' 18. When the crow heard the dog speaking, he laughed: 'Kaa...
kaa...kaa.' 19. The dried fish fell from his beak to the ground. 20. The dog
picked it up and ate it. 21. Only then did the crow realize what had happened
(lit. did he think): 'I have suffered a loss because of the dog's praises.' 22.
The crow was very sorry that the dried fish he had stolen had been eaten by the
dog.
23. The purpose of this story is to teach the younger generation, so that
they know that something illegal cannot be enjoyed. 24. And so that they also
know that while living in this world one must not be led astray by praise.
INTERLINEAR TEXTS 351
TEXT 4: MIE BHETENO NE TOMBULA (The man who emerged from a bamboo stalk)
kamokula-no Tongkuno.
elder-POS T.
we Wabahara.
loc W.
Notes
Free translation
1. The old people say that in former times someone lived here on Muna who had
emerged from a bamboo stalk and whose name was Zaidhul Dhalaali. 2. In that
period there were not yet many people on Muna. 3. This story took place in
Tongkuno. 4. The number of people in Tongkuno at that time was about forty
families. 5. Their leader was called the elder of Tongkuno.
INTERLINEAR TEXTS 355
6. Once the elder of Tongkuno ordered four of his people to go and cut bamboo
in a forest which is called Lambu Bhalano (= Big House). 7. This bamboo was to
be used as binding material. 8. So these four people went to the Lambu Bhalano
forest. 9. They arrived in the forest and as soon as they saw a cluster of
bamboo, they chose the one that had the biggest stalk and started to cut it.
10. But when they cut at the bottom, the bamboo gave a cry of pain, he said:
'Ouch, my leg.' 11. They cut higher up, but the bamboo gave again a cry of
pain: 'Ouch, my belly.' 12. They cut still higher up, but the bamboo kept
uttering cries of pain: 'Ouch, my head.' 13. Since the bamboo kept giving cries
of pain when they cut it, the four men stopped cutting and together they pulled
the bamboo out, roots and all. Then they carried it on their shoulders and
brought it up to Tongkuno. 14. When they arrived in Tongkuno, they immediately
took the bamboo they carried to (lit. in front of) the elder of Tongkuno.
15. Then the four men told the elder of Tongkuno the reason why they had
pulled the bamboo out with roots and all: that it had kept crying out in pain
when they cut it. 16. They had not finished their story, when all of a sudden
the bamboo spoke again; he said: 'Do not disturb me, I am your king, make me a
mosquito net and put me inside.'
17. When the elder of Tongkuno heard the speech of the bamboo, he believed
him at once; and immediately he ordered the women in the village of Tongkuno to
make a mosquito net. 18. Very soon the net was ready. 19. When the net he had
ordered was ready, they brought the bamboo up to the house of the elder of
Tongkuno and put him inside the net. 20. After seven days the bamboo had
disappeared, there was only an exceedingly handsome man in the mosquito net.
21. This is the one who is called 'Born of Bamboo', and treated as lord day and
night by the elder of Tongkuno and by the people of the village of Tongkuno.
22. Not long after this had happened, people in other villages also heard
about it. 23. As a result, presents arrived from all sides, so that one can say
that 'Born of Bamboo' was treated as lord by the whole island of Muna.
24. After forty-four days messengers from Duruka arrived who came up to bring
a message to the elder of Tongkuno. They said: 'At the beach of Duruka, which
is called Wabahara, somebody has been washed ashore, a beautiful girl who rode
(the sea) in a basin; her name is "Lifted from the Basin".' 25. When the elder
of Tongkuno heard this, he brought the message to 'Born of Bamboo'. 26. At once
'Born of Bamboo' ordered the elder of Tongkuno to go and fetch the girl who
rode in the basin. He said: 'Go down and get me the woman who rides in that
basin; that is my wife; her name is Tandiabhe.' 27. Quickly the elder of
Tongkuno ordered people to go down and fetch Tandiabhe from Wabahara. 28. When
they arrived in Tongkuno with her, Tandiabhe went into the mosquito net, and
she lived with 'Born of Bamboo'.
29. Months passed, years passed and they had children: Sugi Manuru, Sugi
Laende and Sugi Patola. 30. They remained in Tongkuno, while the whole island
of Muna brought them honour and supported them. 31. The house of the elder of
Tongkuno where they lived was called a palace; because 'Born of Bamboo' was
considered to be an extraordinary person, he was called king of Muna. 32.
Therefore, all the later kings of Muna are descendants of 'Born of Bamboo'.
33. This was the story of the one who is called 'Born of Bamboo'.
356 APPENDIX
Hamai itu', amba-no, 'tabea a-[m]ala-e-mo inodi kupa'. 28. 'Pa ingka,
where that word-her unless 1sI-take-it-PF I money FUT.not ENIM
ala-mo. 29. Ingka bhe kadu-no kupa-no ini a-mai'. 30. Pasighoono
(IMP-)take-PF ENIM with bag-POS money-his this 1sR-come then
ini. 43. Amba-no: 'A-k[um]ala deki tora inodi ela'. 44. Amba-no: 'Uumbe'.
this word-his 1sI-go first again I TAG word-her yes
isa-hi-hi-no.
older.sibling-PLUR-PLUR-her.
me-kala-kala itu eda'. 54. Amba-no Ndaipitu amba-no: 'Uumbe'. 55. Amba-no:
IMP-RED-go that girl word-POS N. word-her yes word-his
63. Garaa kaasi no-tuampe dua ne wawo-no pimpi. 64. No-tuampe ne wawo-no
SURPR pity 3sR-stuck also loc top-POS cliff 3sR-stuck loc top-POS
ne-inta-intagi A Titibholo.
3sR-RED-wait ART T.
ibu-ku.' 73. No-angka no-ala-e nagha maka no-tei-e welo soronga, maka
wife-my 3sR-go 3sR-take-her that then 3sR-put-her in box then
Titibholo. 75. Nae-feabha hae pada, miina bhe mie telo lambu
T. 3sI-greet what EMPH not be person in house
we tehi Sampuru'. 80. 'Ee, ka-raku-ku bhe manu nagha, ingka aini-ha-kanau,'
loc sea S. INJ NOM-hate-my with cock that ENIM this-HA-me
93. Kotughu daano dua o kae-mbali ngkodau ini. 94. Awatu naando-e watu
true really also ART NOM-become old this that be-it that
Notes
Free translation
1. This child..., Titibholo, his father and mother had died. 2. So he lived
with his grandmother. 3. Until he was grown up he lived with his grandmother.
4. Then, when he was grown up, he went sailing. 5. He went to make a living in
Java. 6. He made a living in Java and after he had got a lot of money he
returned, poor Titibholo. 7. When he arrived he gave his grandmother an order:
'Find me somebody to whom I can propose, mother.' 8. She said: 'Where is there
someone you like, boy?' 9. He said: 'At the village chief's of course; he has
got seven girls there.' 10. She said: 'Which one do you want to propose to?'
11. He said: 'The eldest, if possible.' 12. (She said...) His (grand)mother
went with a bag of money. 13. She went and arrived there; the village chief
(the son?) said: 'How far you have strayed, granny!' 14. She answered: 'I have
indeed really lost my way.' 15. He said: 'What could it be that you need,
granny?' 16. She said: 'Why speak in riddles? Isn't it obvious I have come to
propose, to propose for Titibholo?' 17. He said: 'Has he arrived then?' 18. She
said: 'Yes, he has arrived.' 19. He said: 'Which one does he like? 20. She
said: 'The eldest, if possible.' 21. 'Good grief,' said the eldest, 'why should
Titibholo propose to me? Incredible!'
22. 'OK; just choose,' said their father. 'Look here, if the eldest does not
want him, maybe one of the others will.' 23. She went past them one by one
offering them the bag of money, but all six of them refused Titibholo. 24. She
looked down under the house and said: 'Maybe you, girl; I will propose to you.'
25. The youngest child, Wa Ndaipitu said..., she came. 26. She said: 'OK,
granny. 27. But then,' she said, 'I must take that bag of money.' 28. 'That's
all right, take it. 29. I have come with this bag of money (on purpose).' 30.
Then Wa Ndaipitu came up into the house and took the money.
31. When she came home, Titibholo asked: 'Which one wants me? Who is it that
wants me?' 32. She said: 'The youngest.' 33. 'Hey,' he said, 'that's amazing.'
34. He said: 'In that case I will now go away again to make some money,' he
said, 'because she is not grown up yet.' 35. 'Go ahead,' she said.
36. He went away and when he had been away six months, he returned again,
Titibholo. 37. He arrived with various articles of clothing: a dress, earrings,
a necklace and sandals. 38. When he arrived, they took them to her father, to
the village chief. 39. The old woman, his (grand)mother took them there. 40.
But my goodness, her older sisters were very offended. 41. They thought that
those goods were actually for them, so they hated Titibholo. 42. After that
Titibholo went to ask Wa Ndaipitu's permission to leave. 43. He said: 'I am
going off again, all right?' 44. 'Yes,' she said. 45. He went away and after
six months' absence he returned again. 46. He arrived with clothes, with this
and that, things she was still lacking, plates, what not, pitchers, everything
INTERLINEAR TEXTS 363
she was short of. 47. The ill feeling of her elder sisters kept mounting. 48.
When he had gone sailing three times, he married her, he married Wa Ndaipitu.
49. When they were married - they married at her father's house - they went to
his mother's house. 50. They went to his mother's house and lived there.
51. When their own house was ready, Titibholo sailed off again; he went again
to Java. 52. After visiting Java he would return again. 53. Before he left he
forbade her to go out. He said: 'Do not go out, girl.' 54. Wa Ndaipitu said:
'All right.' 55. He said: 'If your older sisters invite you to go swinging at
the dhangki mango, do not go with them.' 56. She said: 'Yes.' 57. But as soon
as Titibholo had left, they came to invite her. 58. They called on her, all six
of them, they kept on inviting her, but at first she refused. 59. But when one
came alone she agreed. 60. They went to swing. 61. While they swung, she (one
of the sisters) pushed her as hard as possible, so that she might fall in the
Sampuru sea, she thought. 62. When she fell..., Three times she pushed and then
she fell (off the swing). 63. But fortunately she got stuck (in a tree) on the
cliff. 64. After she got stuck, she moved from tree to tree until she reached
the top of the mountain. 65. She went up and lived there, the poor thing; she
lived on that mountain, waiting for Titibholo.
66. After a long time, a month had passed and she had lived on the mountain
top for forty days, the friends of Titibholo passed (on the sea), his friends
preceded him. 67. When one boat passed, she asked; when another passed, she
asked. 68. Then she said: 'Maybe you are Titibholo.' 69. They said..., they
answered from the boat: 'Hey, Titibholo is coming behind.' 70. She waited again
and waited. 71. When she saw the last boat, she called out in greeting: 'Hey,
maybe you are Titibholo.' 72. Titibholo turned round, thinking: 'But isn't that
my wife?' 73. He went up to her, took her and put her in a box; then he
returned home with her. 74. When he arrived, Titibholo did not greet anybody.
75. Who was there to greet? There was nobody in the house, except for the older
sisters. 76. He lifted all his goods into the house, his boxes and goods, he
carried it all into the house. 77. They came out to meet him, and they served
him water so that he could take a bath, Titibholo, in front (of the house).
78. Then a cock crowed. 79. The cock crowed and said: 'Cock-a-doodle-doo, Wa
Ndaipitu has been thrown into the Sampuru sea by her older sisters.' 80. 'Hey,
how I hate that cock; can't you see that here I am,' said her older sisters.
81. How they lied. 82. They thought they were fooling Titibholo. 83. But no.
The cock crowed again: 'Cock-a-doodle-doo, Wa Ndaipitu has been thrown into the
Sampuru sea by her sisters.' 84. After he had crowed three times, the cock
stopped: 'Cock-a-doodle-doo, Wa Ndaipitu has been thrown into the Sampuru sea
by her sisters.' 85. 'Ah, bother that cock, I am right here.'
86. After Titibholo had carried all his goods into the house, he went away to
look for (a place) to drink palm wine. 87. When he had drunk, late at night
past ten o'clock, he came home. 88. He took Wa Ndaipitu out so that they could
cook and have something to eat. 89. Then all the sisters fled together, they
jumped down (from the windows), fleeing from Titibholo. 90. For Wa Ndaipitu was
not dead as they supposed. 91. They had even changed house, they thought that
the goods were now theirs. 92. But she was not dead yet.
93. This is the true story of an old event. 94. And there, far away, there is
still that swing, near the dhangki mango, by the Sampuru sea.
364 APPENDIX
14. Pada aitu garaa welo liwu naando se-mie ni-angka nea-no
finish that SURPR in village be one-person P.PART-appoint name-his
Imamu Bhala Taghi, maka no-ko-kaembali dua. 15. Pasino Imamu Bhala Taghi
I. B. T. then 3sR-HAVE-miracle also then I. B. T.
anagha o gansa bhe polulu-no. 17. No-pooli-e dua Imamu Bhala Taghi
that ART pick.axe with axe-his 3sR-able-it also I. B. T.
peda-mo hale buea. 20. Imamu Bhala Taghi bhe mie welo liwu
like-PF floor crocodile I. B. T. with person in village
na-[m]o-fono liwu.
3sI-CAUS-full village.
21. Garaa naando-mo se-mie dua kamokula welo liwu no-pogau ne kantawu
SURPR be-PF one-person also elder in village 3sR-speak loc heap
Notes
Free translation
1. The person who first saw the Laghumbo spring was Wa Katogha ('The crow'). 2.
Wa Katogha was a widow, 3. who lived on her own in a hut. 4. Wa Katogha saw an
olo bird shaking water from its feathers. 5. She followed this bird and saw
that he put his bill in a hole among the stones. 6. She looked closely at it,
and what a surprise, this hole was full of water. 7. Wa Katogha was very happy
that she had discovered this spring, and right away sh e went home to get
her bamboo container.
8. This went on every day for the whole dry season. 9. The people in the
village naturally wondered where Wa Katogha got her water. 10. She did not have
any banana trees, nor did she have bamboo trees, and neither did she go to
Arobhai. 11. Then they were on the look-out for her and saw that she took her
bamboo container and went towards the hills. 12. The people in the village
spied on her while she put her container in the hole among the stones. 13. When
they knew that Wa Katogha had found a spring in a hole among the stones, the
INTERLINEAR TEXTS 367
village elders held a meeting and decided that they would make a hole in the
side of the mountain so that it would be easier for the whole village to get
water from it.
14. Now there was in this village a leader whose name was Imamu Bhala Taghi
('The priest with the big belly'); an exceptional man. 15. This Imamu Bhala
Taghi made the hole in the side of the mountain. 16. He worked with a pick-axe
and a normal axe. 17. Imamu Bhala Taghi managed to make a hole, but surprise!
the water flowed from inside the hill flooding the whole village. 18. And with
it came a multitude of crocodiles; they came from inside the mountain and moved
outside. 19. When they came outside it looked like a floor of crocodiles. 20.
Imamu Bhala Taghi and the people of the village were at their wit's end
wondering what they could do about the water and the crocodiles that kept
coming out of the mountain, filling up the whole village.
21. Then one of the village elders spoke to those present: 'Let's hold a
meeting and decide to kill Wa Katogha, so that possibly the water may be
stopped. 22. The reason why Wa Katogha should be killed is because it is she
who first discovered this spring.' 23. But the people were filled with pity (at
the idea) that the poor Wa Katogha would be killed, since if it were not for
her, they would never have got the water. 24. Then Imamu Bhala Taghi and the
village elders held a meeting and decided to kill a goat as a substitute for Wa
Katogha. 25. The first goat that came in sight they slaughtered. 26. The goat
was slaughtered by Imamu Bhala Taghi, and then put at the mouth of the stream.
27. The goat was carried with the stream inside the mountain, and at the point
where the goat stopped bleating, the water also stopped. 28. The reason it is
called Laghumbo (multitude) is because a multitude of crocodiles came out with
the water. 29. The water reaches as far as a hundred and twenty metres (inside
the mountain).
Bibliography
Adelaar, K.A.
1983 'Malay consonant-harmony: an internal reconstruction', in: J.T.
Collins (ed.), Studies in Malay dialects, Part I, pp. 57-67. Jakarta:
Universitas Atma Jaya, Badan Penyelenggara Seri NUSA. [NUSA 16.]
Adriani, N. (with A.C. Kruyt)
1914 De Bare'e-sprekende Toradja's van Midden-Celebes. Vol. 3: Taal- en
letterkundige schets der Bare'e taal en overzicht van het taalgebied:
Celebes - Zuid-Halmahera. Batavia: Landsdrukkerij.
Anceaux, J.C.
1952 The Wolio language; Outline of grammatical description and texts.
's-Gravenhage: Nijhoff. [KITLV, Verhandelingen 11; Second edition
1988.]
1987 Wolio dictionary (Wolio-English-Indonesian) Dordrecht/Providence:
Foris.
Berg, R. van den
1987 'Beberapa aspek morfologi bahasa Muna', Lontara 34:43-52.
1988 Muna dialects and the Munic languages: towards a reconstruction.
Paper presented to the Fifth Conference on Austronesian Linguistics,
Auckland. [To appear in Te Reo.]
Berg-Klingeman, L.F. van den
1987 'Klausa relatif Bahasa Indonesia dan Bahasa Muna', Lontara 34:5-25.
Blust, R.A.
1980 'Austronesian etymologies', Oceanic Linguistics 19:1-189.
Bouman, E.P.
1933 Memorie van overgave van de afdeling Boeton en Laiwoei. [Unpublished
typescript. Available at ARA.]
Brascamp, E.H.B.
1918 'Uit het koloniaal archief; XVI: De ontdekking van djatibosschen op
Pangesana in 1727', Tectona 11:723-43.
Bybee, J.L.
1985 Morphology; A study of the relation between meaning and form.
Amsterdam: Benjamins.
Chrtien, C.D.
1965 'The statistical structure of the Proto-Austronesian morph', Lingua
14:243-70.
Couvreur, J.
1935 Ethnografisch overzicht van Moena. Raha. [Unpublished typescript.
Available at KITLV e-128.]
BIBLIOGRAPHY 369
Elbert, J.
1911 Die Sunda-Expedition des Vereins fr Geographie und Statistik zu
Frankfurt am Main; Festschrift zur Feier des 75jhrigen Bestehens
des Vereins. Vol. 1. Frankfurt am Main: Minjon.
Esser, S.J.
1938 'Talen', in: Atlas van Tropisch Nederland, kaart 9b. Batavia:
Topografische Dienst in Nederlandsch-Indi.
Hale, K.
1973 'Deep-surface canonical disparities in relation to analysis and
change: an Australian example', in: T. Sebeok (ed.), Current trends
in linguistics, Vol. 11: Diachronic, areal and typological lin-
guistics, pp. 401-58. The Hague: Mouton.
Hanafi
1968 Hubungan kata ganti orang dengan kata kerdja dalam bahasa Muna,
ditindjau dari segi linguistik deskriptif. [Unpublished thesis IKIP
Makassar. Available at KITLV mf15341986.]
Hopper, P. and S. Thompson
1980 'Transitivity in grammar and discourse', Language 56:251-99.
Iskander, D.
1979 'A second specimen of the Matanna water snake, Enhydris matannensis
(Blgr.), from Raha, Muna, Indonesia', British Journal of Herpetology
5:849-50.
Kosasih, S.A.
1983 'Tradisi berburu pada lukisan gua di Pulau Muna, Sulawesi Tenggara',
in: Rapat evaluasi hasil penelitian arkeologi I, pp. 46-63. Jakarta:
Pusat Penelitian Arkeologi Nasional.
1984 'Hasil penelitian lukisan-lukisan pada beberapa gua dan ceruk di
pulau Muna (Sulawesi Tenggara)', in: Rapat evaluasi hasil penelitian
arkeologi II, pp. 55-66. Jakarta: Pusat Penelitian Arkeologi Nasi-
onal.
Lanayo
1979 Pogala salah satu drama tradisional masyarakat Muna dalam fungsinya
sebagai alat pendidikan. [Unpublished thesis IKIP Ujung Pandang.
Available at KITLV mf52331986]
Macdonald, R.R.
1976 Indonesian reference grammar. Washington: Georgetown University
Press.
Mattalitti, M. Arief, M. Sikki, J.S. Sande, A. Usmar and Padala
1985 Kamus Muna-Indonesia. Jakarta: Departemen Pendidikan dan Kebudayaan,
Pusat Pembinaan dan Pengembangan Bahasa.
Mills, R.F.
1981 'Additional addenda', in: R.A. Blust (ed.), Historical linguistics in
Indonesia, Part I, pp. 59-82. Jakarta: Universitas Atma Jaya, Badan
Penyelenggara Seri NUSA. [NUSA 10.]
Mingneau, M.
1974 Missie en ontwikkelingswerk op het eiland Muna; Transmigratieproject
'La Kapera', Muna, Z.O. Celebes, Indonesi. [Unpublished typescript.]
Mller, S.
1839-44 'Schetsen uit de oostelijke streken des Indischen Archipels, voor-
namelijk van Celebes, Boeton, Amboina en de Banda-eilanden', in:
Verhandelingen over de natuurlijke geschiedenis der Nederlandsche
overzeesche bezittingen, land- en volkenkunde, pp. 83-128. Leiden:
Temminck.
370 BIBLIOGRAPHY
Noorduyn, J.
1987 The languages of Sulawesi. Paper presented at the Colloquium of
European Indonesianists at Passau.
Piliha
1981 Perbandingan pemakaian kata ganti diri bahasa Muna dengan pemakaian
kata ganti diri bahasa Indonesia. [Unpublished thesis IKIP Ujung
Pandang. Available at KITLV mf52401986.]
Quirk, R. and S. Greenbaum
1973 A university grammar of English. London: Longman.
Reichling, A.
1935 Het woord; Een studie omtrent de grondslag van taal en taalgebruik.
Nijmegen: Berkhout.
Sande, J.S., J.F. Pattiasina, M. Sikki, M. Arief Mattalitti and A. Kadir Mulya
1986 Morfosintaksis bahasa Muna. Jakarta: Departemen Pendidikan dan
Kebudayaan, Pusat Pembinaan dan Pengembangan Bahasa.
Sidu Marafad
1976 Perbandingan pemakaian kata ganti diri bahasa Muna dengan pemakaian
kata ganti diri bahasa Indonesia. [Unpublished thesis IKIP Bandung.
Available at KITLV mf15331986.]
Silzer, P.J.
1984 Ambai: an Austronesian language of Irian Jaya, Indonesia. [Ph.D.
thesis, Australian National University.]
Sirk, Yu. Kh.
1988 'Research needs: the Wotu language', Baruga; Sulawesi Research
Bulletin 2:10-2.
Sneddon, J.
1982 'Sulawesi', in: S.A. Wurm and Shiro Hattori (eds), Language atlas of
the Pacific area, sheet 44. Canberra: Pacific Linguistics C-66.
Uhlenbeck, E.M.
1949 De structuur van het Javaanse morpheem. Bandoeng: Nix. [Ver-
handelingen van het Koninklijk Bataviaasch Genootschap van Kunsten
en Wetenschappen 78.]
Verstappen, H.Th.
1957 'Een en ander over het rifpantser van het eiland Muna (Z.O.
Celebes)', Tijdschrift van het Koninklijk Nederlandsch Aardrijks-
kundig Genootschap 74:441-9.
Vonk, H.W.
1937 Nota betreffende het zelfbesturend landschap Boeton. [Unpublished
typescript.]
Vuuren, L. van
1920 Het Gouvernement Celebes; Proeve eener Monographie. Vol. 1. Wel-
tevreden: n.n.
Walker, A.T.
1982 A grammar of Sawu. Jakarta: Universitas Atma Jaya, Badan Penye-
lenggara Seri NUSA. [NUSA 13.]
Yatim, Nurdin
1981 Bahasa Wuna. Jakarta: Departemen Pendidikan dan Kebudayaan, Pusat
Pembinaan dan Pengembangan Bahasa.
Yatim, Nurdin, Rabiana Badudu, Abdul Kadir B. and N. Arifin
1984 Morfologi kata kerja bahasa Muna. Ujung Pandang: Proyek Penelitian
Bahasa dan Sastra Indonesia dan Daerah Sulawesi Selatan. [Unpublished
typescript.]
Index
morphophonemics 2.8
names 5.2.3, 10.6
nasal accretion 2.8.2
nasal substitution 2.8.1
negation 8.5
negators 8.5
nominal inflection 4.9
nominal phrase ch.5
nominalization 5.9.2, 7.7, 10.2.11, 10.2.17,
10.2.18
nominalized relative clauses 9.1.3
non-local prepositions 6.2
nouns 3.3.1, 5.2
numerals 3.3.4, 5.7.1, 5.7.2
numeral verbs 5.7.1, 5.7.2, 7.9.2
object agreement 7.3.3
object complementation 9.2.3
object focus 7.3.4
object incorporation 4.7
object suppression 4.6.1, 7.3.2
oblique indirect objects 7.9.2, 10.2.10
optative 10.2.13
ordinals 5.7.1
partial reduplication 10.3.2
participles 3.5, 4.9
particles 3.3.9, 5.6
partitive 4.6.3, 10.2.11
passive 7.3.4
passive participles 5.7.1, 8.5.3, 9.1.2, 10.2.33
path 6.1.3
pausal clitic 9.22
perfective 9.21, 10.2.29
personal pronouns 3.3.3, 5.3, 8.7.2
personification 10.3.1
petative 8.1, 10.2.16
phonemes 2.1
phoneme frequency 2.3
phonology ch.2
phonotactics 2.7
plural(ity) 5.2.3, 5.4.2, 5.6.5, 10.2.12,
10.2.30
polarity 8.0
polite forms 4.1, 4.8.4, 8.7.1
possessive inflection 4.9
possessive suffixes 4.8, 5.3
pragmatics 7.1.4
predicative 10.2.11
prepositions 3.3.7, ch.6
preverbal IO/DO 7.9.4
profession 10.2.36
prohibitive 8.7.4
pronouns 3.3.3
proper nouns 5.2.3
purpose 6.2.2, 10.2.10
purpose conjoining 9.17
INDEX 375
quantification 5.7.7
quantifiers 3.3.5, 5.7.6
questions 8.6
question tag 8.6.1
question words 8.6.2, 8.6.3, 8.6.5
realis 4.5
reason 7.9.1, 10.2.18
reason conjoining 9.15
recipient 6.1.3, 7.9.1
reciprocals 8.4, 10.2.41
reduced pronouns 5.3.3
reduplication 5.2.3, 10.3
referent 7.9.1
referential demonstratives 5.5.5, 5.5.6
reflexive 5.4.1, 7.9.2
relative clauses 5.8.1, 9.1
relative juxtaposed clauses 9.2.4
repetitive action 10.2.16
result 7.1.4
resultative 6.2.1
right-dislocation 7.1.3
root structure 2.4.2
sentences ch.9
simultaneous conjoining 9.6
sound verbs 7.12.3, 10.2.20
source 6.1.3
specificness 6.1
stress 2.5, 2.6
subjects 4.1
subject complementation 9.2.3
subject harmonization 9.2.2
subject inflection 3.3.2, 3.5, 4.1
subordination 9.0, 9.1
supernumerary reduplication 10.3.3
surprisive conjoining 9.10
syllable 2.4.1
syllable structure 2.4.1
temporal adverbs 7.11.1
temporal affixes 9.7.1
temporal conjoining 9.7
temporal conjunctions 9.7.2, 9.7.3
temporal periphery 7.10.1
time 6.1.3, 10.2.11, 10.2.18
topic 7.1.4, 7.8
topic establishment 7.1.4
topic of contrast or comparison 7.1.4
topic shift 7.1.4
transitive clauses 7.3
transitive verbs 3.3.2
transitivity 4.6.1, 4.6.2
transitivizing 8.1
transposition 4.4, 4.6.3, 10.5
valency 8.0
verbs 3.3.2
verb classes 4.3
376 INDEX