The term “Dusun”
has many referents. The coastal Muslim from the western section
of the Malaysian Archipelago used the term to refer to various agricultural
peoples of Malaya, Kalimantan, and particularly Sabah (formerly
North Borneo). These Dusun peoples shared nothing in common in contrast
to other peoples in the area except perhaps a more stable village
structure and the propensity to plant groves of fruit trees around
their settlements. But this is conjecture.
Coastal Muslim from the eastern section of the Archipelago, mainly
Bajau (Sama), used the term “Ida’an” (alternatively
rendered “Idahan” or “Idaan”) to refer to
similar, non-Muslim peoples but particularly to those in Sabah.
This term appeared in the journals of early English voyagers to
the Borneo coast (see Forrest, 1779:368), and it continued to occur
with frequency in reports on northern Borneo into the late 1800's.
However, the term “Ida’an” sometimes was not restricted
to the Dusun-speaking populations alone, but included in addition
the Murut peoples (Low, 1848:343), the Kedayan of Brunei (Keppel,
1846:194-195), and so on.
Later English adventurers and colonizers, having closer contacts
with the coastal Muslim of the western area of the Archipelago,
such as the Brunei Malay, adopted the term “Dusun” to
refer to the indigenous peoples living on the west coast of Sabah
(e.g., Belcher, 1848, I:504-505; St. John, 1862:226). This folk
category of Dusun was also adopted by the organizers of the British
North Borneo Company (1890). It was used both in their descriptions
of the country and later as a category for the governmental censuses,
and, consequently, it became accepted for administrative purposes.
Thus the fact that many of the culturally heterogeneous peoples
of Sabah are frequently referred to as “Dusun” is due
more to an accident of history than to a conscious ethnological
This situation of course is not unique. Terms used for ethnic
identification that are derived from the folk classifications of
peoples foreign to those being identified might, for convenience,
be referred to as “exonyms.” And, more frequently than
not, exonyms become the accepted names for peoples instead of their
autonyms, which recognize the locally relevant distinctions. Difficulties
arise, however, when the semantic boundaries of such folk categories
are accepted uncritically and incorporated into the linguistic,
ethnological, and demographic literature.
The term “Dusun,” when used in the government censuses,
thus includes a culturally complex group of peoples. Since I have
detailed elsewhere the degree of this sociocultural heterogeneity
and discussed some of the resulting anthropological problems (Appell,
1966, 1967, 1968a, 1968b; Appell and Harrison, 1969), I will here
only briefly illustrate its scope with the extreme case of the so-called
Banggi Island Dusun. The isoglot2 of the Banggi islanders (Schneeberger,
1937) shares roughly only 28 per cent of cognates with the Rungus
Dusun, one of their closest neighbors (see Appell, 1967). In other
areas of culture the Banggi islanders also differ to a significant
degree, suggesting that they represent either a remnant population
or an intrusive one. This conclusion finds support in Dyen’s
(1965) lexicostatistical analysis of Austronesian languages. Languages
in Palawan, just to the north of Banggi Island, appear to be more
closely related to languages on the Borneo mainland than is the
Banggi Island isoglot. Dyen reports that his Dusun isoglot has 31.4
per cent of shared cognates with his Palawanic subfamily, and it
should be noted that the Dusun which Dyen used for his lexical list
was from Papar, which is at the southern end of the Dusun language
family distribution and is, consequently, one of the furtherest
removed from Palawan and Banggi Island.
Similarly, the cultural affiliations of the so-called Dusun peoples
of the Sandakan Residency to the main body of Dusun-speakers on
the west coast are not entirely clear. Along the lightly populated
interior drainage systems of the Kinabatangan and other rivers are
a group of peoples called Orang Sungei, who are included in the
governmental censuses with the main body of Dusun-speakers, although
in fact they may be derived in some cases from Murut populations;
in the Labuk River system are various reportedly Dusun-speaking
groups; and in the Segama River region are other alleged “Dusun”
groups. Clayre (1966), using lexical material, finds the languages
of two of these east coast groups — one in the lower reaches
of the Labuk River and the other in the upper reaches of the Kinabatangan
— closely related to those of the Dusun-speakers on the west
coast (see Map 1).
[Map 1 inserted here.]
MAP 1. Location of Clayre’s Dusun dialects. (After Clayre
 with Residency boundaries added.)
However, if we exclude such divergent or unknown “Dusun”
groups as the Banggi islanders and those of the Sandakan Residency,
it is clear from an inspection of the available lexicons (see Cense
and Uhlenbeck, 1958, for a bibliography) that the remaining Dusun-speakers
living in the geographically contiguous areas of the West Coast
and Interior residencies speak closely related dialects or languages.
Clayre (1966), using more extensive but as yet unpublished linguistic
evidence, reaches a similar conclusion. These peoples form roughly
90 per cent of the Dusun demographic category used in the Government
Census and numbered approximately 130,000 in 1960 (Jones, 1962).
However, as I have previously pointed out (Appell, 1967, 1968a,
1968b; Appell and Harrison, 1969) even this limited class of Dusun-speakers
includes a number of very culturally diverse populations, the members
of which recognize no common cultural heritage and do not accept
the appellation “Dusun” as applying to them.
As originally used, the term “Dusun” meant “villager”
or people of the “orchards” (see Appell, 1965, 1967).
To some this has a pejorative connotation, and, since independence,
objections have been raised to its use. Consequently, I stress the
point that “Dusun” is an exonym and that those peoples
to whom it has been and still is being applied consist of a number
of culturally diverse ethnic units, and I restrict my use of the
term solely to linguistic purposes. A linguistic classification
is concerned with recondite concepts and distinctions, and more
usually than not it overrides the locally relevant cultural and
political distinctions. As such, it is particularly appropriate
for the use of exonyms. Consequently, I presently identify and classify
the ethnic groupings of northern Borneo by means of a binomial terminology,
the first part being the autonym of the group in question and the
last term being the name of the language family to which the group
belongs. This results in the identification of ethnic groupings
as follows: Rungus Dusun, Nulu Dusun, Lotud Dusun, Kadazan Dusun,
and so forth. In the Kudat District alone we have isolated fifteen
or more named ethnic groups of Dusun-speakers (see Map 2).3
At present the degree to which the members of this category of
Dusun-speakers are linguistically related is not completely clear.
While it is clear that isoglots geographically separated by one
or two speech communities are frequently not mutually intelligible,
we do not know whether these isoglots are mutually intelligible
with their neighbors from one end of the Dusun distribution to the
Clayre (1966) takes the position that Dusun includes several different
languages. She is the only person who has yet attempted a systematic
analysis of the various Dusun isoglots. There are serious limitations
to her data which I shall discuss, after summarizing her findings.
Using lexicostatistical techniques and an inspection of morphological
material, she concludes that there are two main divisions of the
Dusun language family: West Coast Dusun and East Coast Dusun. In
her West Coast Dusun she includes four languages (see Map 1): (1)
the Ranau, Bundu, and Tambunan isoglots; (2) Lotud; (3) Penampang;
and (4) Rungus. For the East Coast, she was able to examine only
three scattered isoglots. Because her data are sparse, she is not
sure whether the East Coast division includes one or two languages:
(1) the Labuk and Mangkok isoglots; and possibly (2) a Rungus-Bengkoka
axis forming a link between the West and East divisions.
[Map 2 inserted here.]
MAP 2. The locations of the major Dusun isoglots in the Kudat District.
Clayre has made a very important contribution to our understanding
of the Dusun languages and has established for the linguistic domain
what I have maintained for some time for other cultural domains:
the populations of Sabah, frequently lumped together under the term
“Dusun” by the outsider, consist of a number of socioculturally
complex and heterogeneous ethnic groupings. A better understanding
of the anthropology of Sabah can be obtained, therefore, if we consider
first these ethnic units and their differences before attempting
to establish the relationships of these groups to one another.
One of the more interesting results of Clayre’s study is
that the Lotud have an isoglot that is significantly different from
other isoglots along the west coast, and they appear at present
not to fit into any continuum of progressive differentiation on
a geographical basis. That is, languages on either side of the Lotud
— the Penampang and Bundu languages — are more closely
related to each other than to the Lotud, which suggests that either
the Lotud isoglot or the Bundu and Penampang ones have not evolved
in their present loci. Interestingly enough, this linguistic disconformity
correlates with another feature of culture. Among the Lotud as well
as among the Rungus, heads taken in warfare were kept in the long-house,
whereas a separate head-house was used in both the Penampang and
the Bundu areas. The term for it, or for those associated with it,
bongkawan, is also found among certain groups in the intermontane
plains area (Appell and Harrison, 1969). This is not to say that
the Penampang and Bundu Dusun have in toto closer cultural affiliations
with the intermontane Dusun-speakers than with those along the west
coast. As yet we do not have that kind of evidence. On the other
hand, the limited evidence available indicates that the major watershed
in terms of social structure appears to be between those Dusun-speaking
groups who practice wet rice agriculture, such as the Lotud and
Penampang Dusun, and those having a swidden-based economy, such
as the Bundu and Rungus (Appell and Harrison, 1969).
Clayre’s work, however, has two limitations: her data come
from noncontiguous geographical areas and her ethnic identifications
in some cases are questionable. First, since Clayre’s material
comes from geographically isolated isoglots, we do not know whether
the intervening isoglots are mutually intelligible with those of
her sample. If they are, there could be shown for Dusun an orderly
progression or chain of dialects, which would form, according to
one view, a single language rather than a family of several related
languages, as now appears from Clayre’s data (see Dyen, 1960;
Voegelin et al., 1963).
Secondly, the problem of ethnic identification that arises in
Clayre’s work stems from the same external point of view characteristic
of the early explorers and the British colonial government. This
approach is also found in much of the ethnological and medical anthropological
research in Sabah, thus limiting its value (see Appell, 1968a for
a discussion of this point). As a consequence of this external point
of view, indigenous distinctions are overlooked and problems of
ethnic identity arise, particularly with regard to the two Dusun
isoglots collected by Clayre in the Kudat District, the area with
which I am most familiar.
Clayre (1966) identifies the two Dusun isoglots from the Kudat
District as “Rungus” and “Bengkoka.” On
her map, she shows Rungus occupying the whole Kudat Peninsula (see
Map 1), and she writes that Rungus is “the dialect of the
Kudat Peninsula.” Unfortunately this is wrong; like many visitors
to the Kudat area, Clayre may have been led to this misconception
by the local Dusun-speakers themselves. Rungus is the largest and
most visible ethnic group in the District, and frequently Dusun-speakers
from other ethnic groups in the area will identify themselves to
Europeans as “Rungus” for reasons of ease and simplicity.
I have discussed this problem a number of times (1963, 1965, 1966,
1967), and in particular I have shown (1968a) how this misconception
has invalidated a medical anthropological study of a so-called Rungus
village. In fact, four major isoglots are found on the Kudat Peninsula:
Rungus, Nulu, Gonsomon, and Tobilong (see Map 2).
The problem of identifying Clayre’s “Rungus dialect”
is compounded by the fact that her principal informant in the study
is listed as a missionary who was known among the Rungus proper
as speaking with a “Nulu accent.” The Rungus with whom
I have done extensive fieldwork state that the Nulu, Rungus, and
Gonsomon isoglots are mutually intelligible, differing only in some
small respects, particularly with regard to the personal pronouns.
I believe we can safely conclude that Nulu, Gonsomon, and Rungus
form a single language. However, I would refrain from terming it
the “Rungus language” as this leads to confusion between
this category, externally constructed and overriding local distinctions,
and the indigenous category of the Rungus people, which has significantly
different social boundaries. Since such linguistic categories are
exogenous, I believe that it makes better sense in the long run
and minimizes confusion if geographic terms or exonyms are used.
I suggest that the language consisting of at least the three dialects
of Rungus, Nulu, and Gonsomon be termed the Marudu Dusun language,
since its representatives are found on both sides of Marudu Bay.
Eventually, after further linguistic research, Marudu Dusun may
be considered to include other Kudat isoglots, such as I have located
on Map 2, but as yet there is no evidence for this.
Before discussing Bengkoka Dusun, it is important to note briefly
the importance of considering dialects in any type of linguistic
typology or historical reconstruction. Dyen (1960:36) makes the
A problem might arise in dealing with a language that covers a
large geographical area and is highly dialectalized. It is conceivable
that in its spread different languages could have separated from
it at different points in its territory. If the extensive language
is A, and the offshoots are B and C, it is possible that B is similar
to some dialects of A, and C to others, and the family tree that
results might depend on the dialect of A that was chosen as representative.
Dyen then makes the point that the correct solution would require
that the relation of any two languages be determined by the percentages
scored by their most closely related dialects (see also Dyen, 1965
This problem of dialect identification and the relationship between
local dialects arises in Clayre’s Bengkoka Dusun category.
She states (1966:3) that Bengkoka is one of several minor dialects
spoken on the Melobong Peninsula, but which of the many dialects
it is cannot be identified from her description. On Map 2 I have
listed the various isoglots of the Melobong Peninsula as identified
by Dusun-speakers themselves. Melobong Rungus are very closely related
to the Rungus of the Kudat Peninsula, and the members of these two
Rungus sections state that they do not speak different isoglots.
It would be interesting to know whether Clayre’s Bengkoka
Dusun by chance includes lexical material from the Melobong Rungus.
Before concluding, it might be worthwhile to speculate on the
form future classification of the indigenous languages of Sabah
might take. At present there are four major groups: the Dusun languages,
the Northern Murut languages, the Southern Murut languages, and
Bisaya.4 Although there has been speculation that the Tidong languages
of the east coast are related to the Murut languages, field research
is needed to determine whether this group can in fact be considered
an indigenous language group or whether it is one that has moved
into Sabah in relatively recent times. Likewise I do not include
the language of the Banggi islanders in this classification since
the status of their language must still be determined.
My Northern Murut group is equivalent to the North Borneo Murut
category of Cense and Uhlenbeck (1958), and my Southern Murut is
equivalent to their Sarawak Murut. I have not followed their terminology
since the members of these two groups are not restricted to the
political divisions of Sabah or Sarawak, as is suggested by this
terminology. However, my own terminology should be rapidly outmoded
by the work of D. J. Prentice of the Australian National University,
who has recently returned from an extensive field investigation
of the various Murut groups in Sabah. The results of his significant
research will soon be published, and this will clear up a number
of important problems such as the linguistic and ethnic divisions
of the Northern Murut, the relationship of the Murut languages to
the Dusun language group, and the relation of Northern Murut to
Southern Murut. I here exclude the Southern Murut from my postulated
language family of Sabah since there appears to be a significant
linguistic and cultural boundary between the Northern and Southern
Murut, with the latter showing more close affinities with a large
group of peoples to the south than with the Northern Murut (see
Appell, 1968a and Cense and Uhlenbeck, 1958). However, the dimensions
of this linguistic discontinuity have never been adequately determined,
and the answer will have to await the results of Prentice’s
With regard to the Bisaya, Roger Peranio, of the State University
of New York at Stony Brook, has collected extensive linguistic data
among the Bisaya of Sarawak, and once this work has been analyzed
and published it should contribute to our understanding of the relationship
of the Sabah section of the Bisaya to the Dusun languages and to
those of the Northern and Southern Murut. The Dusun group of languages
is still largely unanalyzed, as we have already pointed out, and
this extensive group needs the same kind of detailed and careful
field study that Prentice has done for the Murut. In the meantime,
however, I would like to postulate one language family for the indigenous
languages of Sabah, to be called the “Ida’an language
family.” This term seems particularly appropriate not only
because it is an exonym but also because it has been applied at
one time or another to most of the members of this postulated family.
This Ida’an family would thus include the three subfamilies
of Dusun, Northern Murut, and Bisaya.
Finally, I have appended a basic word list for the Rungus isoglot,
which was collected from several informants, all from villages in
the Matunggong River basin on the Kudat Peninsula. All of the informants
spoke only Rungus until, in their mid-teens, attempts were made
to learn Bazaar Malay. All maintain that they are “true Rungus”
in contrast to other ethnic units on the Kudat Peninsula. My linguistic
analysis is not yet complete, and, therefore, I have omitted several
entries where the most appropriate Rungus word is still in question.
Appendix: Rungus Dusun basic word list
| 1. all
| 2. and
| 3. animal
| 4. ashes
| 5. at
| 6. back
| 7. bad
| 8. bark (tree)
| 9. because
|13. to bite
|16. to blow (wind)
|18. to breathe
|19. to burn (intr.)
|22. cold (weather)
|23. to come
|24. to count
|25. to cut
||monibas (with knife)
|26. day (not night)
|27. to die
|28. to dig
|31. to drink
|32. dry (substance)
|33. dull (knife)
|36. earth (soil)
|37. to eat
|40. to fall (drop)
|42. fat (substance)
|44. to fear
|45. feather (large)
|47. to fight
|51. to float
|52. to flow
|54. to fly
|58. to freeze
|60. to give
|69. to hear
|73. to hit
| (in hand)
|76. to hunt (game)
|82. to kill
|83. know (facts)
|85. to laugh
|87. left (hand)
|89. to lie (on side)
|90. to live
|94. man (male)
|96. meat (flesh)
|112. to play
|113. to pull
|114. to push
|115. to rain
|117. right (correct)
|118. right (hand)
|127. to say
|128. scratch (itch)
|129. sea (ocean)
|130. to see
|132. to sew
|133. sharp (knife)
|135. to sing
|136. to sit
|137. skin (of person)
|139. to sleep
|141. to smell (perceive odor)
|147. to spit
|148. to split
|149. to squeeze
|150. to stab (or stick)
|151. to stand
|153. stick (of wood)
|156. to suck
|158. to swell
|159. to swim
|166. to think
|170. to throw
|171. to tie
|173. tooth (front)
|175. to turn (veer)
|177. to vomit
|178. to walk
|179. warm (weather)
|180. to wash
|182. we (incl.)
|194. With (accompanying)
1 Fieldwork among the Rungus Dusun of Sabah, Malaysia, during
1959-1960 and 1961-1963, was conducted under the auspices of the
Department of Anthropology and Sociology, Institute of Advanced
Studies, the Australian National University. The Australian National
University also supported an initial analysis of the data collected.
NSF Grant 923 supported further analysis of the data and the preparation
of material for publication. Support from an ACLS grant made the
preparation of this paper possible.
2 I coin here the term “isoglot” to refer to the speech
of a group of people who consider their language or dialect to be
significantly different from neighboring communities and thus have
an indigenous term by which to identify it. I find this term preferable
to “language” or “dialect” as these imply
a certain linguistic status vis-à-vis other languages or
dialects, whereas the term “isoglot” is neutral in this
regard. An isoglot therefore refers to the speech of a self-conscious
speech community. I also find this term preferable to other terms
that have been used and which may or may not stress self-consciousness,
such as “communalect” (Wurm, 1964) and “isolect”
(Hudson, 1967). There has been an interesting development in linguistic
terminology in which the pseudo-morpheme, -lect, has been used with
various prefixes to mean speech, for example, those just mentioned
and also “idiolect” (Hockett, 1958). This usage, to
my way of thinking, is a barbarism whether it is derived from “dialect”
or “lecture.” Dialect does not consist of the two morphemes
*dia- and *-lect, and lecture on the other hand is derived from
the Latin stem meaning “to read.” The term “isoglot”
includes two morphemes: iso-, derived from Greek and meaning “alike,”
“equal,” or “the same,” and -glot, also
from Greek and in a derivative sense denoting “language.”
3 This method of binomial terminology may not be universally applicable
in Sabah. In certain lightly populated areas of the interior, social
identification through group membership is not in terms of a multivillage
ethnic group but in terms of the name of the village in which the
individual resides. Thus, no autonyms exist above the village level
by which an individual may establish his ethnic identity in contrast
to other ethnic groups. This appears to be a function of population
density, and I have discussed elsewhere some of the factors involved
(Appell, 1968b). It would be interesting to know in these unusual
cases to what degree there is dialect differentiation between neighboring
4 Harrisson (1962) has summarized what is presently known about
Appell, G. N. 1963. Myths and legends about the Rungus of the Kudat
District. Sabah Society Journal 4:9-15.
_____. 1965. The nature of social groupings among the Rungus Dusun
of Sabah, Malaysia. (Unpublished Ph.D. dissertation.) Canberra,
Australian National University.
_____. 1966. The ethnography of the Dusun-speaking peoples of
Sabah. Sarawak Museum Journal 14:376-388.
. 1967. Ethnography of northern Borneo: critical review of some
recent publications. Oceania 37:178-185.
_____. 1968a. A survey of the social and medical anthropology
of Sabah: retrospect and prospect. Behavior Science Notes 3:1-54.
_____. 1968b. Social anthropological research in Borneo. Anthropologica
(Ottawa). In press.
Appell, G. N., and Robert Harrison. 1969. The ethnographic classification
of the Dusun-speaking peoples of northern Borneo. Ethnology 8(1).
Belcher, Edward. 1848. Narrative of the Voyage of H.M.S. Samarang
during the years 1843-46. 2 vols. London, Reeve, Benham & Reeve.
British North Borneo Company. 1890. Handbook of British North
Borneo. London, William Clowes and Sons.
Cense, A. A., and E. M. Uhlenbeck. 1958. Critical survey of studies
on the languages of Borneo. Koninklijk Instituut voor Taal-, Land-
en Volkenkunde, Bibliographical Series 2. ‘s-Gravenhage, Martinus
Clayre, B. M. 1966. A comparison of some dialects of Dusun. Sabah
Society Journal 3:3-12.
Dyen, Isidore. 1960. Comments to “Lexicostatistics So Far”
by D. H. Hymes. Current Anthropology 1:34-39.
_____. 1965. A lexicostatistical classification of the Austronesian
languages. International Journal of American Linguistics, Memoir
19 (Volume 31, No. 1).
Forrest, Captain Thomas. 1779. A voyage to New Guinea and the
Moluccas, from Balambangan. London, G. Scott.
Harrison, Tom. 1962. “Bisaya” in North Borneo and
elsewhere. Sabah Society Journal 2:6-14.
Hockett, Charles F. 1958. A course in modern linguistics. New
Hudson, Alfred B. 1967. Padju Epat: The ethnography and social
structure of a Ma’anjan Dajak group in southeastern Borneo.
(Unpublished Ph.D. dissertation.) Ithaca, Cornell University.
Jones, L. W. 1962. North Borneo: Report on the census of population
taken on 10th August, 1960. Kuching, Government Printing Office.
Keppel, H. 1846. The expedition to Borneo of H. M. S. Dido for
the suppression of piracy; with extracts from the journal of James
Brooke Esq. of Sarawak. London.
Low, Hugh. 1848. Sarawak; its inhabitants and productions, being
notes during a residence in that country with H. H. the Rajah Brooke.
London, R. Bentley.
St. John, Spenser. 1862. Life in the forests of the Far East.
Schneeberger, W. F. 1937. A short vocabulary of the Banggi and
Bajau language. Journal of the Malayan Branch Royal Asiatic Society
Voegelin, C. F., et al. 1963. Obtaining an index of phonological
differentiation from the construction of non-existent minimax systems.
International Journal of American Linguistics 29:4-28.
Wurm, S. A. 1964. Papers in New Guinea linguistics No. 2. (Linguistic
Circle of Canberra Publications Series A, Occasional Papers No.
4.) Canberra, Australian National University.