ThesisPDF Available

Some Aspects of the language of the Orang Seletar

Abstract and Figures

This paper presents an analysis on some aspects of Seletar, the language of the Orang Seletar, namely the phonology, lexicon and polysemy. The Orang Seletar are one of Singapore’s indigenous sea peoples (Orang Laut) and one of the many groups of Orang Asli (‘indigenous people’) of Malaysia. Seletar is considered a dialect of Malay. Although some research has been done on the Orang Seletar, they often lack linguistic documentation and detailed analysis. Some interesting observations of Seletar were made in this study, for example neutralization of /p/ and /m/ in p-initial verbal roots, sesquisyllabic structure, various vowel-sequences, frequent contractions of Standard Malay lexical items, and polysemy unseen in Standard Malay. Seletar has some non-Malay lexical items which might indicate borrowings from Aslian and Borneo languages. However, many words seem uniquely Seletar. The paper concludes with a transcribed word list of 345 items based on the recordings of two speakers. Much further research is needed to analyse Seletar in detail. More than anything, this paper serves as an introduction to Seletar, which many overlooked as they deemed it ‘very similar to Malay’. The paper hopes to be a starting point to more detailed studies regarding the Orang Seletar, linguistic or otherwise.
Content may be subject to copyright.
Some Aspects of the language of the Orang Seletar
i
NANYANG TECHNOLOGICAL UNIVERSITY
SCHOOL OF HUMANITIES
Some Aspects of the language of the Orang Seletar
Student Name:
Tan Zhi Xuan (U1630040B)
Supervisor:
Assoc Prof Alexander Robertson Coupe
A Final Year Project submitted to
the School of Humanities,
Nanyang Technological University
in partial fulfilment of the requirements for the
Degree of Bachelor of Arts
in Linguistics & Multilingual Studies
2020
i
Declaration of Authorship
I declare that this assignment is my own original work, unless otherwise
referenced, as defined by the NTU policy on plagiarism. I have read the
NTU Honour Code and Pledge.
No part of this Final Year Project has been or is being concurrently
submitted for any other qualification at any other university.
I certify that the data collected for this project is authentic. I fully
understand that falsification of data will result in the failure of the
project and/or failure of the course.
Name: Tan Zhi Xuan
Signature:
Date: 31/03/2020
ii
Additional information
1. My FYP is an extension of my URECA project. No
If yes, give details and state how is this project different from your URECA
project:
2.  No
if yes, answer question
3. grant. No
4. Provide details of funding expenditure, (e.g. payment of participants: $10/hour;

Not applicable.
iii
5. You are reminded that an FYP is an independent project on an area of your own
interests. Any kind of outsourcing (including, but not limited to, transcription,
statistical analysis, etc.) would have a huge negative impact on your grade. If
your project is a part of or an extension of your supervisor
clearly i) what are your own intellectual contributions, and ii) what components of

audio/visual materials, etc.). If you have any doubts, please check with your
supervisor and/or the FYP coordinator before submission.
*Note that failure to declare may be treated as a case of plagiarism and it will
impact negatively on the assessment of your FYP.
Not applicable.
v
Contents
Contents ......................................................................................................................... v
List of Figures and Tables ...........................................................................................viii
List of Abbreviations .................................................................................................... ix
Acknowledgements ........................................................................................................ x
Abstract ......................................................................................................................... xi
1 Introduction ............................................................................................................ 1
Aim of the research project ............................................................................. 1
Outline ............................................................................................................. 1
The Orang Asli and the Orang Laut ................................................................ 1
1.3.1 Orang Asli ................................................................................................ 2
1.3.2 Orang Laut (of Singapore) ....................................................................... 3
1.3.3 Orang Laut (of Southeast Asia) ............................................................... 4
1.3.4 The language of the Orang Laut .............................................................. 6
Orang Seletar ................................................................................................... 7
1.4.1 Seletar (Place Name) ................................................................................ 8
1.4.2 Orang Seletar ........................................................................................... 8
1.4.3 The language of the Orang Seletar ......................................................... 10
2 Literature Review ................................................................................................. 12
Types of Literature ........................................................................................ 12
Previous Literature ........................................................................................ 12
2.2.1 Impressionistic descriptions ................................................................... 12
2.2.2 Word lists and Comparison of Vocabulary ........................................... 13
2.2.3 Technical Linguistic Analysis ................................................................ 14
2.2.4 Others ..................................................................................................... 15
Research Gap ................................................................................................. 15
3 Methodology ........................................................................................................ 16
Language Consultants ................................................................................... 16
Data Collection .............................................................................................. 16
3.2.1 Location ................................................................................................. 16
3.2.2 Data ........................................................................................................ 16
Equipment and Software ............................................................................... 17
3.3.1 Audio Recording .................................................................................... 17
Limitations of Study ...................................................................................... 17
4 Phonetics and Phonology ..................................................................................... 20
Syllable Structure .......................................................................................... 20
vi
Phonotactics .................................................................................................. 20
Sesquisyllabicity ............................................................................................ 23
Consonants .................................................................................................... 26
4.4.1 Oral Stops ............................................................................................... 26
4.4.2 Affricates ................................................................................................ 28
4.4.3 Nasals ..................................................................................................... 28
4.4.4 Trill ........................................................................................................ 28
4.4.5 Fricatives ................................................................................................ 29
4.4.6 Approximant .......................................................................................... 29
4.4.7 Lateral Approximant .............................................................................. 29
Vowels ........................................................................................................... 29
4.5.1 Monophthongs ....................................................................................... 30
4.5.2 Vowel Sequences ................................................................................... 33
5 Lexicon ................................................................................................................ 36
Malayic .......................................................................................................... 36
Malayic and Malay ........................................................................................ 36
Riau Sea Tribe Dialect Group ....................................................................... 37
5.3.1 Sound Innovations ................................................................................. 37
Lexical Items ................................................................................................. 39
5.4.1 Contracted Malay words ........................................................................ 39
5.4.2 Sumatran Malay lexical items ................................................................ 42
5.4.3 Non-Malay lexical items ........................................................................ 42
5.4.4 Others ..................................................................................................... 49
5.4.5 Remarks ................................................................................................. 49
6 Polysemy .............................................................................................................. 50
Wood/Tree ..................................................................................................... 50
Eye/Seed ........................................................................................................ 52
Soil/ Mud ....................................................................................................... 53
7 Conclusion ........................................................................................................... 55
Review of Paper ............................................................................................ 55
Future Research ............................................................................................. 55
8 Appendix .............................................................................................................. 56
Appendix A Orang Seletar Word list ......................................................... 57
Appendix B Comprehensive Word list ...................................................... 66
9 Additional Sources ............................................................................................... 86
Other Useful Sources .................................................................................... 86
9.1.1 Cultural Centre ....................................................................................... 86
vii
9.1.2 Photographs ............................................................................................ 86
9.1.3 Oral Interviews ....................................................................................... 86
9.1.4 Documentaries ....................................................................................... 87
References .................................................................................................................... 88
List of Figures and Tables
Figures
Figure 1-1 Map of the Aslian Languages (from Benjamin, 2012, p. 144) .................... 3
Figure 1-2 Orang Laut Distribution (Obsidian, 2014 based on Sopher, 1977, p. 54,
Plate III) ......................................................................................................................... 6
Figure 1-3 The distribution of the Orang Laut languages (from Anderbeck, 2012, p.
269). ............................................................................................................................... 7
Figure 1-4 The Orang Laut languages ........................................................................... 7
Figure 1-5 Orang Seletar Villages ............................................................................... 10
Figure 1-6 The Orang Seletar villages ......................................................................... 10
Figure 4-1 Seletar Syllable Structure ........................................................................... 20
Figure 4-2 Sound Diagrams for /dpan/ ....................................................................... 25
Figure 4-3 Sound Diagrams for /ptei/ ........................................................................ 25
Figure 4-4 Sound Diagrams for /nta/ ........................................................................ 25
Figure 4-5 Seletar Consonant sound inventory ............................................................ 26
Figure 4-6 Seletar Monophthongs ............................................................................... 30
Figure 6-1 The direction of polysemy of eye/face and seed/fruit ................................ 53
Tables
Table 1 Distribution of Consonants of Seletar ............................................................. 21
Table 2 Vowel Distribution of Seletar ......................................................................... 21
Table 3 Examples of Seletar words which are contracted Malay words ..................... 41
Table 4 Non-Malay words in Seletar ........................................................................... 42
List of Abbreviations
The following abbreviations are used in the text:
V vowel
N nasal
C consonant
syllable
# word boundary
1 first person
2 second person
3 third person
NEG negation
SG singular
PL plural
PROX proximal
DIST distal
DEM demonstrative
LOC locative
GEN genitive
PTCL speech particle
JOA Jabatan Orang Asli

JAKOA Jabatan Kemajuan Orang Asli
 
JHEOA Jabatan Hal-Ehwal Orang Asli

ABVD Austronesian Basic Vocabulary Database
PAn Proto-Austronesian
PMP Proto-Malayo-Polynesian
Acknowledgements
I am grateful to so many people who have helped in the conception, preparation and
drafting of this paper.
To T, M and R and other Orang Seletar who have contributed their time and participated
in this study, thank you. Your unique way of life is truly fascinating. Thank you for
sharing your experiences and stories. I look forward to working with you all again.
To my high school teacher Wen Yong, thank you for introducing me to the Orang
Seletar, and personally taking me to the village. Your passion for local culture and
geography inspires me.
To Prof Geoffrey Benjamin and Prof , who very kindly shared their
knowledge and sources with me, thank you.
To Francesco (Perono Cacciafoco), thank you for letting me do research on the Sequent
Occupance paper. It started my interest in the Orang Laut of Singapore, and
subsequently the Orang Seletar. Thank you for all the classes and the research
opportunities, I have learned so, so much.
To Alec, my FYP supervisor, thank you for teaching me so much these few years. The
study would not be possible without your constant guidance, suggestions,
encouragements and sharing of experiences of what it means to be a field linguist.
Thank you for always being patient. Thank you for taking your precious time to read
all the drafts. Any remaining errors in writing and the analysis are entirely my own. I
look forward to working with you more in the future!
To Tim, thank you for being here and your encouragements for this last leg of the race.
To my parents, thank you for the constant support. Without your open-mindedness and
curiosity, this study and these four years would not have been possible. I will see you
soon.
xi
Abstract
This paper presents an analysis on some aspects of Seletar, the language of the Orang
Seletar, namely the phonology, lexicon and polysemy. The Orang Seletar are one of
 (Orang Laut) and one of the many groups of Orang
     Seletar is considered a dialect of Malay.
Although some research has been done on the Orang Seletar, they often lack linguistic
documentation and detailed analysis. Some interesting observations of Seletar were
made in this study, for example neutralization of /p/ and /m/ in p-initial verbal roots,
sesquisyllabic structure, various vowel-sequences, frequent contractions of Standard
Malay lexical items, and polysemy unseen in Standard Malay. Seletar has some non-
Malay lexical items which might indicate borrowings from Aslian and Borneo
languages. However, many words seem uniquely Seletar. The paper concludes with a
transcribed word list of 345 items based on the recordings of two speakers. Much
further research is needed to analyse Seletar in detail. More than anything, this paper
serves as an introduction to Seletar, which many overlooked as they deemed it very
similar to Malay. The paper hopes to be a starting point to more detailed studies
regarding the Orang Seletar, linguistic or otherwise.
1 Introduction
AIM OF THE RESEARCH PROJECT
This research project aims to document the language of the Orang Seletar. As a result
of rapid urban development on the coast of Johor Bahru, Malaysia, the traditional
lifestyle of the Orang Seletar has been severely impacted. Several Orang Seletar
villages have faced and are facing issues regarding land rights and may be evicted from
their current settlements by private developers in the near future, should the state require
the lands (Nicholas, 2000). Aside from the uncertainty of land occupation, the
development projects at the coast have polluted the area. The sea and the mangrove
forests, traditionally inhabited by the Orang Seletar, have been losing ground to
concrete jungles. Pollution have caused the daily catch of the Orang Seletar, like fish
and crabs, to decrease severely, affecting their livelihood.
Because of the above stated issues, many within the community have chosen to
embrace mainstream society, and as a consequence have lost contact with their
indigenous culture, including religion, customs and language. With the threat of
extinction looming, documentation of their language is not only useful, but a necessary
task.
OUTLINE
The FYP is divided into 7 parts. This chapter explains who the Orang Asli, the Orang
Laut and the Orang Seletar are, and the current status the Orang Seletar language.
Chapter 2 discusses previous literature regarding the study of the language. The
methodology is explained in Chapter 3. Chapter 4 presents an analysis of the phonetics
and phonology, followed by an evaluation of the lexicon in Chapter 5. Chapter 6 is on
Polysemy of some Seletar words and the report concludes with Chapter 7.
The collected word list can be found in the Appendix. Some additional (non-linguistics)
sources are listed in the Additional Sources.
THE ORANG ASLI AND THE ORANG LAUT
The Orang Seletar have a multilayered identity. They are (i) one of 18 tribes of Orang
Asli (indigenous people) in Peninsular Malaysia, (ii) one of the groups of Orang Laut
(sea people) indigenous to Singapore and (iii) part of the numerous Orang Laut tribes
in Island Southeast Asia who make their living from the sea.
1.3 The Orang Asli and the Orang Laut 2
1.3.1 ORANG ASLI
Peninsular Malaysia has 18 groups of indigenous peoples. The 18 are categorized into
three subgroups, based on genetics and differences in social customs by the Department
of Orang Asli Affairs (JHEOA)
1
in Malaysia (JAKOA, 2020)
2
:
Negrito: Kensiu, Kintak, Lanoh, Jahai, Mendriq, Bateq
Senoi: Temiar, Semai, Semoq Beri, Jahut, Mah Meri, Ceq Wong
Proto-Malay: Kuala, Kanaq, Temuan Jakun, Semelai, Seletar
In this view, the Orang Seletar are Proto-Malays, who migrated to the Malay
Archipelago between 2500 and 1500 BC
3
(Ryan, 1976, pp. 4-5). The JHEOA
classification is not based on linguistic divisions, though.
Benjamin (2002, p. 22) categorizes the peoples on linguistic affiliation, and
identifies the ethnic groups
4
stated above into Mon-Khmer (Austroasiatic) and
Austronesian speakers. The Mon-Khmer speakers (in the context of Orang Asli, the
   have subgroups of Northern, Central, Jahut (a single
language) and Southern Aslian (Benjamin, 2012, p. 145). The Austronesian languages
spoken by the Orang Asli are Malay and Duano (Kuala) (Benjamin, 2012, p. 137).
Figure 1-1 shows the regions where the Orang Asli languages are distributed.

rather than the present-day locations, which are more restricted  
Orang Kanaq and Orang Seletar are Malayic dialects spoken by Orang Asli; Duano is
an unclassified Austronesian language
1
This is the one of the previous names for the department. Created in 1954, it was first called the
Department of Orang Asli (Jabatan Orang Asli, JOA). In 1963, it was renamed the Jabatan Hal-Ehwal
Orang Asli (JHEOA). In 2011, the department was renamed to the Department of Orang Asli
Development (Jabatan Kemajuan Orang Asli, JAKOA).
2
       kuih lapis (layer cake) folk-scholarly
ethonology (p. 18), which hierarchically ranks the population in terms of its evolutionary
    
thousands of years ago.
3
Where the Proto-Malays migrated from is disputed, some say Yunnan, others Taiwan (Nik Hassan &
Nik Abdul Rahman, 1998)
4
Benjamin (2002, p. 22) uses the labels given by the JHEOA, but states that these labels actually absorbs
other smaller groups not officially recognized by the department, namely the Temoq population (under

3 Introduction
Figure 1-1 Map of the Aslian Languages (from Benjamin, 2012, p. 144)
The Orang Seletar inhabit the coasts along the Straits of Johor. Due to their
proximity to the Johor coast and Singapore, the Orang Seletar also once roamed the
waters and islands of northern Singapore and considered to be one of the indigenous
peoples of Singapore (Benjamin, 2002, p. 31; Mariam, p. 2002, p. 274-280; Benjamin,
2016, p. 1; Hwang, 2018).
1.3.2 ORANG LAUT (OF SINGAPORE)
There were several groups of Orang Laut (sea people) present in Singapore before the
arrival of colonial powers. Other than the Orang Seletar, the indigenous sea people of
Singapore were the Orang Gelam, Orang Biduanda Kallang and Orang Selat,
collectively known as the Orang Laut (Hwang, 2018; Turnbull, 2009, p. 24). The Orang
Laut were said to have been present on the island since the 16th century, according to
Tomé Pirés, a Portuguese apothecary (Cortesão, 1944, p. 262). It is hard to tell the
1.3 The Orang Asli and the Orang Laut 4
groups apart from early records, as the earliest documentation by the Portuguese called
them collectively as Çelates
5
.
The Orang Laut were imperative to the formation of Singapore in the early
1800s (Andaya, 1975; Trocki, 2007). They lived near the palace of the rulers of the
time, surrounding Bukit Larangan (now Fort Canning Hill) and the Singapore River
(Wheatley, 1961, p. 81; Turnbull, 2009, p. 24). Archaeological excavations of the area
provide evidence of their prior domicile (Miksic, 1985, pp. 1-35; Kwa, 1985, pp. 121-
123).
Although they were regarded as crucial in forming early Singapore, today, the
Orang Laut are regarded as people of the past, no longer existing in the current discourse
of Singapore. All the places formerly inhabited by the Orang Laut e.g. Pulau Seking,
Tanjong Irau, Kampung Rokok have been destroyed, making way for urban
     into mainstream Malay society
(Mariam, 2002, p. 275).
1.3.3 ORANG LAUT (OF SOUTHEAST ASIA)
The Orang Laut of Singapore are actually part of a larger population that make their
living from the sea, and traditionally live in Island Southeast Asia. They have been
mentioned in literature as Sea Nomads, Sea Gypsies etc (this is true also for the Orang
Laut of Singapore). There exist many tribes or suku of Orang Laut throughout the
region. They differ in lifestyle, with some living on land and others leading nomadic
lives in boats (Anderbeck, 2012, p. 266).
The Orang Laut have been important players in the history of Southeast Asia
for three thousand years (Andaya & Andaya, 2001, p. 14). Their roles were tied to the
sea. During the time of the Srivijaya empire (7th -14th century), the Orang Laut helped
guide ships to important ports and patrolled the waters around the Malay peninsula and
the Riau archipelago (Andaya, 2018). During the time of the Malaccan and Johor
kingdom, they served as navy troops for Malay rulers and fishermen gathering sea
products for trade (Andaya, 1975; Andaya, 2008; Chou, 2003; Anderbeck, 2012, p.
266). They maintained strong patron-client relations with the rulers of Malacca and
Johor, but this relationship soured during the end of the 17th century, when the last
Malaccan sultan, Sultan Mahmud Shah II, was assassinated in 1699. This brought about
a divide within the different tribes of Orang Laut, and loyalties towards the new ruling
families waned (Trocki, 2007, p. 26).
The intervention of Bugis traders in the territory, followed by the appearance of
Illanun raiders in the 18th century also contributed to the decline of the Orang Laut
5
Çelates selat, which is also the namesake of the Orang Selat
and the Orang Seletar (Sopher, 1977, p. 337).
5 Introduction
(Barnard, 2007, p. 45; Trocki, 2007, p. 32; Anderbeck, 2012, p. 267). The late comers
had much more advanced technology in sea faring, and could offer more services to the
subsequent Malay rulers in their maritime expeditions when compared to the Orang
Laut (Barnard, 2007). The 19th century saw to the arrival of European powers to the
region, and with them, land-based industrial colonialism. Many groups chose to
abandon their maritime nomadic ways as a result (Sopher, 1977, p. 114), identifying as
, as it was far more economically advantageous (Lenhart, 1997, p. 586). The
current situation for Orang Laut at large is similar to those stated in §1.1 and §1.3.2,
many choosing fixed settlements on land and incorporation into mainstream society
(Hoogervorst, 2012).
Sopher (1977, p. 50, p. 54 - Plate III; see Figure 1-2) documented the
distribution of the Orang Laut and divides them into three groups: the Mawken (Moken
and Moklen), the Sama-Bajau, and the Orang Laut. The Moken and Moklen live on the
islands on the western coast of Thailand and Burma. The Sama-Bajau reside in the
northeast of Borneo, in the Sulu archipelago, and the islands of southern Philippines.
The Orang Laut can be further classified into the Urak Lawoi and the Orang Laut proper
(Andaya, 2008). The Urak Lawoi are sea people living along the northern part of the
Melaka Straits, in the islands and coasts of Thailand and Malaysia (Andaya 2008). The
Orang Laut reside in the Riau-Lingga Archipelago (Chou, 2003). As mentioned
previously, within this group of Orang Laut, there are many different tribes
6
(Andaya,
1975, p. 44; Chou, 2010, p. 26), including the Orang Kallang and Orang Selat
mentioned in §1.3.2 and also the Orang Seletar of this study.
6
Anderbeck (2012, p. 269) notes that ethnic affiliations are quite fluid.
1.3 The Orang Asli and the Orang Laut 6
Figure 1-2 Orang Laut Distribution (based on Sopher, 1977, p. 54, Plate III)
1.3.4 THE LANGUAGE OF THE ORANG LAUT
Sopher (1977, p. 176-183), on top of documenting the distribution of the Orang Laut,
also categorized their language into groups: Mawken, Bajau and Malayic.
7
As is
obvious from the name, the Mawken groups are the Moken and Moklen, and the Bajau
speakers the Sama-Bajau. These languages are Malayo-Polynesian languages, but non-
Malayic (Anderbeck, 2012, p. 267). Speakers of Malayic languages consist of the Urak
Lawoi and Orang Laut.
Anderbeck (2012, p. 267) adds to the Malayic group a further division into
Kedah, Riau Islands, Duano
8
and Sekak. Figure 1-3 shows an illustration of the
distribution of the varieties. group, while
the Orang Laut speak languages of the other three. The Kedah varieties are found in
southern Thailand to Kedah in Malaysia. The speakers of the Riau Islands varieties are
clustered in the Riau-Lingga Archipelago, between Sumatra, Singapore and west
Kalimantan. Duano is spoken by the Orang Duano (also Orang Kuala) on the west coast
of Johor, Riau and Jambi of the Sumatra. Sekak is spoken by Sea Tribes in the islands
of Bangka and Belitung. Figure 1-4 shows the classification made by Sopher (1977)
and then Anderbeck (2012, p. 267).
7
Malayic refers to languages deriving from the branch of Proto-Malayic (Anderbeck, 2012, p. 271).
8
Benjamin (2012, p. 137, 144, 151) leaves Duano as an unclassified Austronesian language.
7 Introduction
The language of the Orang Seletar is considered a Riau Islands variety.
Figure 1-3 The distribution of the Orang Laut languages (from Anderbeck, 2012, p. 269).
Figure 1-4 The Orang Laut languages
ORANG SELETAR
This section introduces the Orang Seletar: the possible origin of their name, their history
(described by early documentations), their current settlements, their livelihood and their
language.
Proto-
Austronesian
Formosan
Proto-Malayo-
Polynesian
Mawken
Bajau
Malayic
Kedah
Riau Islands
Duano
Sekak
1.4 Orang Seletar 8
1.4.1 SELETAR (PLACE NAME)
The earliest documentation of the word Seletar is in the Sejarah Melayu (Malay
Annals). The annals mentioned a place Sěletar,
9
through which Parameswara escaped
from Singapore to Muar in the chaos of a Javanese attack in the 15th century (Shellabear,
1967, p. 69):
-g Singapura. Maka Raja Iskandar pun
-
were defeated, and Sultan Iskandar Shah fled, going by way of Saletar and

10
Later mentions of Seletar come in multiple variations: the Seletar River marked as
R.Saleta in 1828 (Franklin & Jackson, 1828 - Plan of Singapore), Selita in 1865
(Cameron, 1865, pp. 86-88), Slitar (Hidayah, 2017), but most maps of Singapore after
1860s write Seletar as it is now.
1.4.2 ORANG SELETAR
It is unknown how long the Orang Seletar have lived in the Straits of Johor. As
mentioned above in §1.4.1, the earliest written record was 15th century. But views
regarding the origin of the place name, whether the ethnonym Seletar derived from the
place name Seletar (or the hydronym Sungei Seletar) or the other way around, is
disputed, and probably will never be answered. If the place name did get its origins
from the Orang Seletar, then the Orang Seletar probably have inhabited Singapore since
the 15th century, the time period in which the events stated in the Malay Annals took
place.
Probably the first time the Orang Seletar were mentioned as a group of distinct
people was in the 1819 Treaty of Friendship and Alliance between the East India
Company and Singapore, which listed the Orang Seletar population as 200 (Horton,
1997; Turnbull, 2009, p. 25). Earliest documented records describing the Orang Seletar
were in 1847 by British officials James Richardson Logan and John Turnbull Thomson.
Logan (1847, p. 302-
the rivers and creeks of Johor, noting their appearance and their shy nature. Thomson
(1847, p. 341-
prolonged contact with the people, in terms of their appearance, number, living
conditions, customs, language etc. Accompanying the descriptions, Thomson also drew
illustrations of the Orang Seletar (Thomson, 1847, p. 350 Plate I; Thomson, 1848).
Around 50 years after Thomson and Logan, Skeat and Ridley (Skeat & Ridley, 1900)
claimed that the people were already hard to find:
9

10
English Translation by Brown (1970, p. 41).
9 Introduction
Of these races it is not easy now to find any traces, as they have become
amalgamated with the Malays, adopting not only their language but also
their customs and religion. (p. 247)
Aside from Johor, the Orang Seletar were in Singapore. Carey (1976, p. 279)
describes a small group living in Singapore, one of three Orang Seletar settlements

75), confirms this with interviews with other Orang Seletar who have lived in Singapore
or have relatives who stayed in Singapore. The Orang Seletar who stayed in Singapore
amounted to about thirty people, but have since completely abandoned their traditional
lifestyle and assimilated to Malay (Ariffin, 1979, p. 17). The Orang Seletar resided in
the northern coast during their time in Singapore, especially around mangrove forests
in the area now known as Seletar and along the mouth of the Seletar River. The Seletar
island off the northern coast was also their home. All the place names above are
believed to be derived from the Orang Seletar. Up until the 70s, some of them resided
at the north-
75). The Temenggong of Johor, Sultan Abu Bakar, was said to have taken some of them
from Singapore to the Pulai River, in southwest Johor during the 1900s (Mariam, 2002,
p. 279; Savage & Yeoh, 2013, p. 290). Many of them still live around the area today.
     
waters between the two nations. In the 1980s, when Singapore started clearing the
mangrove forests in Seletar for the construction of an airbase, the Orang Seletar (aside
from those who chose to stay) moved across the waters to Johor to join others of their
tribe.
Traditionally, the Orang Seletar lived on boats, not on shore. However, that
changed around the 1950s (Ariffin, 1979), when the government wanted the originally
nomadic people to settle on land during the Malayan Emergency.
11
The earliest
settlement on land was in Kuala Sungai Redan in Pontian, but the Orang Seletar have
since occupied multiple areas along the Johor coast. The Orang Seletar of Johor today
still mainly work as fishermen. Other than fish, they also collect crabs, snails and clams
in the mangrove forests and riverbanks.
12
The current population is 1620
13
(Khairul,
11
The Malayan Emergency (Malay: Darurat Malaya), from 1948-1960, was a period where the
Federation of Malay, under British rule, fought a guerilla war with the Malayan Communist Party. In
order to prevent local people from aiding the communists, estimated around 500,000 people, most of
them Chinese, were interned in a resettlement program, to live in  where there are strict
curfews, no schools, reduced rice rations etc. The Orang Seletar were not interned in this resettlement
program, but the British did want to settle them in a fixed place, for surveillance purposes 
2017).
12
The habitat in which the Orang Seletar live in, mainly the mangrove forests, is very different from
Orang Laut proper, and Sopher (1977, p. 107) reported that the local Malays would call them Orang
Utan Seletar (Forest People Seletar), as opposed to Orang Laut (Sea People).
13
The Ethnologue lists the Orang Seletar population in Singapore as 790, taking their data from
. It
is not stated how these numbers were obtained. I believe all these numbers were meant to be Orang
1.4 Orang Seletar 10
2019), but the actual numbers might be higher, as the census probably does not consider
individuals living in urban communities. The Orang Seletar present in Johor Bahru
reside in eight villages: Kampung Telok Kabung, Simpang Arang, Sungai Temon, Bakar
Batu, Pasir Salam, Pasir Putih, Kuala Masai and Teluk Jawa 
Figure 1-5.
In recent years, both the sea and the mangroves are losing ground to land
reclamation projects, and as a result, the livelihoods of the Orang Seletar are heavily
affected. Despite having a long history with the Straits of Johor, the Orang Seletar (and
their    absorbed   MalayIndian
categorization both in Singapore and Malaysia, making them virtually unknown to
mainstream society.
1.4.3 THE LANGUAGE OF THE ORANG SELETAR
As mentioned in §1.3.1 and §1.3.4, the language of the Orang Seletar is an Austronesian
language, under the Malayo-Polynesian branch. It is Malayic and part of the Orang Laut
Riau Islands varieties (Anderbeck, 2012, p. 269).
The language of the Orang Seletar is called Bahasa Seletar (Ariffin, 1979;
, 2017). The speakers I have met agree that such is the name for their language,
but they also sometimes refer to it (to me) as Bahasa Asli 
Seletar in Johor, as they roughly correspond to the Orang Seletar population provided by the JHEOA in
1996 in Benjamin (2002, p. 22) which is 801, and the current one of 1615 by Khairul (2019).
Figure 1-6 The Orang Seletar villages
Figure 1-5 Orang Seletar Villages
11 Introduction
among themselves Bahasa Kon (Kon 
been borrowed from Mon-Khmer (Benjamin, 1997, p 109-111), see §5.4.3). In the rest
of the report, I will use Seletar to refer to the language of the Orang Seletar.
The Ethnologue  in Malaysia, due to
the disruption of intergenerational transmission (Eberhard, Simons and Fenning, 2019).

transmitted intergenerationally, and only the grandparents generation can speak the
language (Eberhard, Simons and Fenning, 2019). I have not met any Orang Seletar from
Singapore, so I cannot comment on this status. The Ethnologue also states the number
of speakers as 790,   
exactly how this number was obtained. JAKOA in Malaysia does not collect data
regarding number of speakers, only ethnic population. Thus, the number of speakers of
Seletar is unclear, but the ballpark figure is 2000.
Many Orang Seletar are bilingual in Malay and Seletar. Some who have chosen
Chinese partners are able to speak some Mandarin Chinese. Current issues regarding
land and livelihood are affecting the Orang Seletar community, not just in terms of their
lifestyle, but also their language. With more exposure to the mainstream society, the
Orang Seletar are increasing their use of Malay in order to communicate with
 Thus, Seletar is rapidly losing ground to Malay, which many consider is the
macrolanguage of Seletar (Pelras, 1972; Pelras, Pernia & Durrell-Khalife, 2002 ;
Benjamin, 2012).
The language does not have a writing system (Eberhard, Simons and Fenning,
2019). The older Orang Seletar I have met are illiterate. The younger generation that
have gone or are going to school can read and write Malay.
This first chapter has presented the aim of the research and introduced the Orang
Asli, Orang Laut and the Orang Seletar history, their language and
its current status were also described. The next chapter discusses previous research done
on Seletar.
2 Literature Review
TYPES OF LITERATURE
Literature regarding the Orang Seletar and their language is very limited and also very
scattered. Earlier sources are mostly anthropological notes, with only a small section
on language. The descriptions are not technical but concern general characteristics of
the language. More recent work focused more on collecting word lists from the people,
but their main focus is also anthropological. Some sources are non-academic, but due
to the limited amount of research, they are nevertheless valuable. More technical
linguistic work usually focuses on Orang Laut languages at large, and oftentimes lacks
Seletar data. No analysis of morphological or syntactical structures of Seletar can be
found.
PREVIOUS LITERATURE
I have divided past studies on Seletar into four parts: (i) Impressionistic descriptions
(ii) Word lists and comparison of vocabulary with Malay (iii) Technical linguistic
analysis and (iv) Others.
2.2.1 IMPRESSIONISTIC DESCRIPTIONS
As mentioned in §1.4.2, Thomson (1847) took detailed notes on the Orang Seletar,
including their language. He also included a list of Orang Seletar names, which he found
to be similar to the names of heroes in the Malay Annals. Here is an excerpt of how he
described the language:
Their language is the Malayan, and considerable pains was taken to elicit
any words foreign to that language, but without effect. Their dialect is the
same as that of the Orang Laut of Tulloh Blangah, but spoken with a
slightly more guttural accent, and they clip their words as much as the
natives of Keddah. As a proof of their possessing the same language as the
Malays, I may mention that the children were heard when playing to
converse in this language and were perfectly understood by the Malays
amongst our crew (p. 343).
As I before stated they speak the language of the Malays with much less a
degree of difference in pronunciation, than may be found in stepping from
one county in England to another  We find in their proper names an
astonishing degree of similarity to the names of Malayan heroes prior to
the conversion of the race as mentioned in the Sijarah Malayu and other
works (p. 346).
13 Literature Review
2.2.2 WORD LISTS AND COMPARISON OF VOCABULARY
Carey (1976) collected a word list of two hundred lexical items (but not published

only six words differing from standard Malay (p. 279). These words were not listed.
But he does note the usage of ‘Kun’ when the Orang Seletar were referring to
-
Johor and Pahang (Carey, 1976, p. 279).
 Pelras, Pernia & Durrell-Khalife, 2002)
fieldwork of Duano led him to venture briefly into Orang Seletar territory, so that he
could compare the similarity between the two languages. He concluded that the two are
not closely related. His view of Seletar is that it is simply a dialect of Malay, but with
frequent contractions due to the dropping of intermediary parts of words. According to
his data, 85% of the words he collected are Malay, another 5% are Malay words but
semantically different, and 10% are non-Malay words. Among this 10%, he notes some

study are in the Appendix.
Ariffin (1979) conducted fieldwork in Kampung Simpang Arang for his
anthropological studies. A brief two-page section on language was written. He supports
the view of the close relatedness between Seletar and Standard Malay and lists a few
words that deviate from this similarity (see Appendix). He notes some peculiarities in
Seletar, namely the glottal stop ending of words and the dropping of consonant /k/ and
/r/. He also observes some multilingualism in the community, with some able to speak
fluent Malay and Mandarin.
Yusop (2011) might have the largest complied word list yet for Seletar. He has
collected 1590 words from speakers of Kampung Bakar Batu. After examining the list,
it is obvious that Malay and Seletar are very closely related. I have listed some words
from Yusop in the Appendix.
Blissett and Elzinga (2015) conducted fieldwork to collect word lists from two
endangered languages in Malaysia, Seletar and the Sungai Sugut language in Sabah,
and compared them to Standard Malay. For Seletar, they found that many words are
exactly the same with Malay, and others obviously deriving from phonological changes.
He observes a small group of words unique to the language. He notes the missing meN-
prefix and -an suffix of Malay in Seletar, as well as the strategy of negating verbs,
which is unique from Malay (but did not say what that is).
Nazarudin (2015, Table 5) lists 22 Seletar words associated with seafaring.
   -149) describes some aspects of Seletar. He gives a
description on its current status as a home language, which he says is used daily
throughout all ages, and highlights the bilingual abilities of the Orang Seletar. He also
2.2 Previous Literature 14
gives examples of pronunciation differences between Seletar and Malay, namely the
glottal stop endings, the dropping of /k/ and /i/ in the middle of words and /r/ at the start,
middle and end of words. Some sentences (transcribed using Malay orthography) in

Malay is given, followed by kinship terms. Though there is a particular section
dedicated to language of the Orang Seletar, throughout the book, there are words written
in italics representing terms relevant to the Orang Seletar. Some of these are exactly the
same as Malay, but semantically different from Malay, while others seem unique to the
language.
Hidayah (2017) in he
But these words were taken from Daud, Yunus and Sitti (1986), which actually studies

2.2.3 TECHNICAL LINGUISTIC ANALYSIS
Kähler (1960) studied the Malay dialects spoken by several Orang Laut tribes (Orang
Darat of Batam Island, Orang Akit on Rupat Island and Orang Hutan in Tebing Tinggi
Island and Orang Laut) in very detailed phonetic descriptions. Though Kähler did not
use Benjamin (2002, p. 27) says he
demonstrated similarities between the language spoken by the Jakuns and the Orang
Seletar and other Orang Suku Laut of the Riau Islands. As I cannot read German, I
cannot confirm this. Also, Anderbeck (2012, p. 279) says the location of the mentioned
Orang Lautähler 
Anderbeck (2012) did not work on Seletar specifically, but provides
classifications for the Orang Laut varieties from examining several Orang Laut
languages from a multitude of sources. He adds to the existing sources with words
collected from the Bintan Orang Laut in the Riau Islands (see Appendix). He lists the
sound innovations shared by the Riau Islands variety (see §5.3.1). From comparing the
different varieties, a conclusion is made that there is little evidence of a non-Malayic
-
which may represent an older stratum of Malay old Malay (p. 294). Aside from his
linguistic observations, he mentions Arnaud et al. (1997) having Orang Seletar word
lists, but I do not have access to this resource, and neither does he. He guessed that the
data may be the same as Pelras (1972) above.
Mohamad Yusof and Mohamad Nor (2014) illustrates the phonemic inventory
of Proto-Malay languages, compares them and tabulates their degree of relatedness and
their similarity to Standard Malay. Seletar was found to be 87.1% cognate with
Standard Malay, 87.1% with the language of the Jakun, 84.1% with the Temuan and
73.7% with Duano.
15 Literature Review
2.2.4 OTHERS
Samsur (2015a, 2015b, 2019) focuses on analysing the Orang Seletar language as an
interaction, with a discourse analysis kind of approach. He includes conversations in
Seletar and Malay (they use Malay when talking to the researcher), which are
transcribed using Malay orthography, some notes about the differences between Malay
and Seletar words, and a tiny list of Seletar pronouns.
Benjamin (in press) main focus was to assess Austronesian languages in
Singapore other than Malay. These include both Malayic languages, like the Orang
Selat language and Banjarese, and non-Malayic languages like Javanese and Buginese.
He suggests that closer inspection of the language of the Orang Seletar may give us
some insights into the Greater North Borneoigin theory of the homeland of Malayic
languages suggested by Blust (2010).
RESEARCH GAP
As seen from the review above, technical linguistic analysis on the Orang Seletar
language is severely lacking, and efforts should be taken not just to document isolated
words used by the Orang Seletar, but also natural language use, such as conversations,
instructional texts etc.
This research project aims to describe the language in more technical terms, with
regard to the phonetics and phonology, lexicon and the semantics (polysemy) of Seletar.
3 Methodology
This section discusses the language consultants, the data collection process and details
regarding the equipment and software used to record and analyse the data.
LANGUAGE CONSULTANTS
The language consultants for this study are three Orang Seletar (T, R and M) from
Kampung Simpang Arang. Both T and M are women, 63 and 35 years old respectively,
and R is a 41-year-old man. R and M are siblings, while T is the cousin of both R and
M.
All three consultants are native speakers of Seletar. Besides their native language,
they all speak Malay. M is married to a Chinese man, and she speaks a bit of Mandarin
Chinese. T is Muslim, while R and M are Christians. They all speak Seletar at home
and to their children. R is a fisherman, but he also hunts for wild boars. T also used to
work as a fisherman but has stopped going out to sea as frequently as before because of
her age. Instead, she looks for mud crabs and fish baits (which they sell to anglers) in
the mangroves. M operates a small business selling drinks and snacks to locals. T was
born on a boat and started living on land in the 1960s. T, R and M are illiterate.
DATA COLLECTION
3.2.1 LOCATION
I did not have access to the facilities ideal for recording in the village. Thus, the audio
and video recordings were made in front of the shophouse owned by M. As this was a
shop, during the sessions, there were interruptions from visitors and customers. The
consultants and I re-recorded where possible.
3.2.2 DATA
A word list was collected from T and R. M provided inputs. Each participant gave full
consent to being recorded prior to audio and video recording.
The language of elicitation used is Malay. I have learned Standard Malay in
school in Malaysia for 12 years and regularly use Malay when I am in Malaysia.
3.2.2.1 Word list
A 345-item word list was collected from the consultants. These included 210 words
taken from the Austronesian Basic Vocabulary Database (ABVD) (Greenhill, Blust &
Gray, 2008)         
(1972), Blis
added 17 words of my own to these. Two recordings were made, one from T and another
from R.
17 Methodology
For a clearer comparison, I added Proto-Austronesian word list (Blust, 1999a;
Blust & Trussel, 2020), Proto-Malayo-Polynesian word list (Blust, 1999b; Blust &
Trussel, 2020-Malayic list beside the ones I have collected.
Words were asked in Malay, and the consultants gave the corresponding words
in Seletar. During the elicitation, there were also a group of other Orang Seletar present.
Sometimes, they would remind or correct the consultants on the elicited words.
The word list is transcribed phonetically in IPA and can be found in the
Appendix.
EQUIPMENT AND SOFTWARE
To record the data, microphone, audio recorders and cameras were used during the
fieldwork.
3.3.1 AUDIO RECORDING
A Zoom H5 Handy Recorder was used to record the audio. The sampling rate was at
48kHz and 24 bits. The audio was recorded as Mono WAV files. For all the audio
recordings, a cardioid condenser wired lapel mic was used (Audio-Technica AT831b).
The acoustic analysis of the audio recordings was done using Praat.
LIMITATIONS OF STUDY
Almost all Orang Seletar are bilingual in their native language and Malay. When
speaking to outsiders, they tend to use Malay as the language of communication. As a
result, the consultants may have shifted from Seletar to Malay unconsciously during
elicitation. Seletar is very closely related to Malay in terms of its lexicon, with a lot of
words being exactly the same, and others with pronunciation differences (see Chapter
5). Thus, at times, it is difficult to tell whether the word elicited is a word used in Seletar
with the exact same pronunciation as that of Malay, or a Malay word, pronounced as a
Malay speaker would, but not used by the Orang Seletar.
This limitation is especially felt during the phonological analysis, when
determining whether certain vowels and consonants are contrastive or not. For example,
       adiʔ in Seletar. When asked whether the
substitution of [i] with [e] would change the meaning (adeʔ is the pronunciation of
            
Seletar does not contrast these two vowels in this environment (thus putting [i] and [e]
in free variation), or because the latter was a Malay pronunciation of the word, which
they also use when shifting to Malay? The questions would require further investigation
and, ideally, elicitation methods which would avoid the influence of Malay.
I tried to avoid these circumstances to the best of my ability. As M speaks a little
bit of Mandarin, when I felt that there was some confusion between Seletar and Malay,
I would ask the words to M in Mandarin, and she would provide me with the Seletar
3.4 Limitations of Study 18
words. But her understanding of Mandarin is very limited and thus cannot cross-check
the majority of the words in the word list.
Given the preliminary stage of the study and my very limited knowledge of Seletar,
Malay is a very important contact language between the Orang Seletar and me, and
cannot currently be avoided.
The methodology of this study has been discussed. The next chapter presents an
analysis of the phonetics and phonology of Seletar.
4 Phonetics and Phonology
The sound system of Seletar is discussed in this chapter. First, the basic syllable
structure is introduced in §4.1, followed by the phonotactics in §4.2 . The notion of
sesquisyllabicity is discussed in §4.3. The chapter ends with the description of the sound
inventory of Seletar, starting in §4.4.
The data is transcribed broadly using IPA.
SYLLABLE STRUCTURE
The Seletar syllable displays the structure found in Figure 4-1. Optional constituents
are enclosed in parentheses. An utterance can be formed by minimally one syllable. A
syllable consists of optional onsets plus a rhyme. The rhyme contains an obligatory
nucleus and an optional coda. The nucleus may contain a vowel sequence of two vowels
(see §4.5.2.2), or a single syllabic consonant.
Figure 4-1 Seletar Syllable Structure
PHONOTACTICS
The possible consonants allowed in the onset and coda slots are presented in Table 1.
The onset contains only one consonant (C1) and can be filled by any of the 19
consonants except the glottal stop (see §4.4 for the consonantal inventory). The coda
slot (C2) is only limited to unvoiced stops /p, t/, nasals /m, n, /, lateral /l/, fricatives /s,
h/ and the glottal stop //. /r/ is usually dropped in the onset and the coda positions, see
§5.3.1, but in some words it is retained (e.g. səratɔs bənar 
due to the unconscious shifting stated in §3.4. In some words, /r/ appears as allophone
[], see §4.4.4. The glottal stop usually occurs in the coda position, but it does
sometimes occur in medial position, see the discussion of §4.4.1.
(C1)
(C2)
21 Phonetics and Phonology
Table 1 Distribution of Consonants of Seletar
Onset
Coda
C1
C2
/p/
/b/
/m/
/t/
/d/
/n/
/r/ [ɣ]
()
()
/s/
/l/
/w/
/tʃ/
/dʒ/
/ɲ/
/j/
/k/
/g/
/ŋ/
/ʔ/
(of second syllable)
/h/
The nucleus can be formed by one or two vowels. The V1 slot can be filled by any of
the six vowels (see §4.5 for the vowel system). V2 can be filled by any vowel except
/e/. Table 2 illustrates the vowel distribution.
Table 2 Vowel Distribution of Seletar
V1
(V2)
/i/
/e/
/a/
/ə/
/u/
/o/
4.2 Phonotactics 22
Roots in Seletar are largely monosyllabic and disyllabic, though trisyllabic roots do
appear.
Monosyllabic roots:
V /e/ 
VC // 
VVC /al/ 
CV /u/ 
CVV /bou/ 
CVC /ki/ 
CVVC /kian/ 
Disyllabic roots:
V.VC /a.i/ 
V.CV /a.ti/ liver
V.CVV /e.kou/ 
V.CVC /a.bo/ 
VV.CVC /ai.sa/ 
VC.CVC /um.put / 
CV.VC /la.ot/ 
CV.CV /la.b/ 
CV.CVC /ko.kot/ 
CV.CVV /ko.tou/ 
CV.CVVC /s.mua/ 
CVC.CV /mim.pi/ ; 
CVC.CVV /ban.tai/ 
CVC.CVC  
Trisyllabic Roots:
CV.CV.CVC /t.li./ 
CV.CVC.CVC /k.lon.tot/ -
CVC.CV.CVC /t.g.lm/ 
23 Phonetics and Phonology
SESQUISYLLABICITY
           
(1973, p. 86) in his discussion on tones in Austroasiatic languages.
-[Austro-  ture, with
             
-
 -Khmer] languages also have a syllabic structure
intermediate between the truly monosyllabic [Sino-Tibetan] and truly polysyllabic

Prior to that, there was 
(1952) about twenty years earlier. She notes that the minor syllable is a stage in between
the extended monosyllable and the full, or major disyllable.
Although sesquisyllabicity has been discussed extensively since then in many
languages (e.g. Diffloth & Zide, 2003; Thomas, 1992; Kruspe, 2004; Burenhult, 2005;
Brunelle & Pittayaporn, 2012; Michaud, 2012, p.; Butler, 2015), the definition of a
sesquisyllable is vague and varies from scholar to scholar (Pittayaporn, 2015, p. 502).
Some consider a sesquisyllable to be any disyllabic word with a reduced number of
phonemic contrasts in initial syllables, while others insist on only a neutral vowel or
syllabic consonant in the initial syllable. Pittayaporn (2015) and Bulter (2015) try to
show how varied and complicated sesquisyllables can be.
The definition used here for a sesquisyllable is the one given by Michaud (2012,
p. 2), who defines          
optional nucleus, V: either a vowel, or a sonorant (nasal or liquid) serving as nucleus.
In the Austroasiatic domain, the most frequently encountered situation is one in which
there can be no vowel contrast in the presyllable: the nucleus consists simply in a schwa,

Sesquisyllabicity is prominent in Austroasiatic (Mon-Khmer) languages,
including Aslian languages. The presence of sesquisyllables in Seletar may indicate
some contact between Orang Seletar and Aslian speakers. Jakun, also Malayic and
spoken in areas near the Orang Seletar, also has this final syllabic stress, noted as earlier
as 1906 by Skeat and Blagden (1906, p. 773), who also postulated Mon-Khmer
influences.
Anderbeck (2012, p. 274), with his examination of Jakun vocabulary collected
by Seidlitz (2005) and Skeat and Blagden (1906), thinks that the Jakun people were
originally speakers of an Aslian language (specifically something similar to the extinct
Aslian language called Kenaboifor the concept
of masuk Melayu .
4.3 Sesquisyllabicity 24
Bradley (1980) observes something similar in Burmese. He notes that Burmese
in recent history has reduced many of its full syllables to minor syllables, as a
Many speakers originally
speaking Mon-Khmer languages are shifting to Burmese, but the Mon-Khmer
characteristics are still felt in the sesquisyllabicity of words.
This may be the case too for the Orang Seletar, that they were originally
speakers of (maybe) an Aslian language but has since shifted to Malay. The
sesquisyllabic structure of words may be a remnant of the Aslian substratum that once
characterized the language of the Orang Seletar, many centuries ago before their
lexicon, see §5.4.3, but
these are so few in number that they might not mean anything but some language
borrowing. More research is required to analyse this phenomenon in detail.
Some sesquisyllabic words are presented below. Figure 4-2, Figure 4-3 and
Figure 4-4 show the spectrographs and the waveforms of three sesquisyllabic words,
dəpan , pətei  and ntʃaʔ NEG. The sesquisyllabic words often have
voiced onsets, but they do occur in voiceless onsets as well, like pətei 
[dma] 
[bsal] 
[dkat] 
[ptei] 
[n.ta] NEG
[n.tua] -
An epenthetic schwa is transcribed between the onsets of the two syllables for
dəpan pətei For ntʃaʔ NEG the nucleus of the minor syllable is
occupied by a syllabic nasal.
25 Phonetics and Phonology
nta
ta
ptei
Figure 4-2 Sound Diagrams for /dpan/
dpan  ptei 
/nta/ NEG
Figure 4-3 Sound Diagrams for /ptei/
tei
Figure 4-4 Sound Diagrams for /nta/
4.4 Consonants 26
CONSONANTS
19 consonants were identified. These are listed in Figure 4-5 according to their place
and manner of articulation. The inventory shows six distinct places of articulation and
seven manners. The size of the consonant inventory is average (19-25 consonants, the
most typical consonant inventory), as categorized by Maddieson (2013).
Bilabial
Alveolar
Palato-alveolars
Palatal
Velar
Glottal
Oral Stops
p
b
t
d
k
Nasal
m
n
Trill
r
Fricative
s
h
Approximant
w
j
Lateral approximant
l
Affricate
t
d
Figure 4-5 Seletar Consonant sound inventory
4.4.1 ORAL STOPS
Oral stops occur in 4 places of articulation bilabial, alveolar, velar and glottal. The
stops are all unaspirated. Voicing is contrastive for bilabial, alveolar and velar stops.
The minimal pair here shows the contrast between voiced and voiceless bilabial
stops.
/p/ versus /b/
/pa.u/ 
/ba.u/ 
/p/ is sometimes realized as [m]. 

asked to repeat the words. This may be due to the influence of the Standard
Malay verbal prefix mə(N)-, which changes the initial voiceless bilabial stops
of the verbal root to bilabial nasals, e.g. pukol > məmukol. In Standard Malay,
initial voiceless /p/, /t/ and /k/ of the root word undergoes a homorganic nasal
substitution to /m/, /n/ and // respectively when either the active verbal prefix
mə(N)-, actor/instrument prefix pə(N)- or the nominal circumfix pə(N)- is
attached (Adelaar, 1992, p. 9).
For Seletar, the homorganic nasal substitution happens only for /p/, and without
the attachment of prefixes.
/p/ [p]~[m] / verbal roots with initial p
27 Phonetics and Phonology
/ptit/ [ptet]~[mtet] to 
 [puse~[muse 
However, /p/ and /m/ are also contrastive. The following minimal pair shows
this.
/p/ versus /m/
/pa/ 
/ma/ 
It seems that the contrast for /p/ and /m/ is suspended for verbal
roots with onset /p/, which can be realized as [m] with no meaning change. This
suspension of contrast between phonemes in certain environments is called
neutralization.
14
Neutralization introduces the concept of an archiphoneme,
which         
sounds using only the common distinctive features shared (Trubetzkoy, 1969),
and it is represented in capital letters.
The voiced and voiceless alveolar stops are shown here by a near-minimal pair.
/t/ versus /d/
/tga/ straight
/d/ to 
There are voiced and voiceless velar stops in Seletar, represented by a near-
minimal pair below.
/k/ versus /g/
/ki/ left
/gigi/ 
A glottal stop is very common in Seletar but its status is a bit complicated. Usually,
it appears in the coda of the final syllable, closing it. This is seen in both Malay and
Seletar, for example dudoʔ masaʔ  both languages. However,
Seletar has a glottal stop coda in final syllables of words which are open syllables
in Malay, for example kakiʔ təliŋaʔ kaki and təliŋa in
Malay. In these cases, the glottal stop may be considered to be a marker of word
prosody, signalling the end of a word (but in the data collected, not every open
syllable has a glottal stop following it). This distinction is necessary as it influences
the realization of /i/ in §4.5.1.
The closing of originally open syllables with glottal stop is a phenomenon among
Sea Tribe languages in the Riau Archipelago (Anderbeck 2012, p. see §5.3).
14
On neutralisation see Trubetzkoy (1969) and Akamatsu (1988).
4.4 Consonants 28
However, for some words (not recognizable as Malay), it is difficult to determine
whether the glottal stop is a phoneme closing a syllable, or just a word boundary
marker. For example, tɔʔ ŋəʔ dʒɔʔ .
Another instance where a glottal stop occurs is in the medial position. For example,
kɛʔɛŋ  and utaʔa . But these words may just be variant pronunciations,
as the consultants says kɛŋ and utaha when repeating the words.
4.4.2 AFFRICATES
Voicing is contrastive for affricates. There are two affricates present in Seletar: /t/ and
/d/, shown by near-minimal pairs.
/t/ versus /d/
/panta/ 
/pandit/ 
/ktit/ 
/kd/ 
4.4.3 NASALS
Seletar has 4 types of nasals bilabial, alveolar, palatal and velar. The contrast between
the nasals are presented by minimal and near-minimal pairs.
/m/ /mt/ 
/n/ /nps/ 
// /t/ 
// /t/ 
Aside from //, the nasals precede other kinds of vowels as well.
/m/ /mt/ 
/mata/ 
/n/ /nama/ 
// 
// // 
/baa/ 
 // 
es/ to cry
/t/ 
4.4.4 TRILL
Seletar has an alveolar trill /r/. /r/ has an allophone [], which is in free variation with
[r]. /r/ is usually dropped in Seletar words, for example, sut surut
in Standard Malay. But in some words, the trill is retained, maybe due to the consultants
29 Phonetics and Phonology
switching to Malay, as mentioned in §3.4. Where /r/ is not dropped, it sometimes
appears as [

/r/ [r]~ [] in free variation
Examples,
/s-ribu/ [s-ribu] one-
/krana/ [krana] 
/brat/ [bat] 
/rambut/ [ambut] 
4.4.5 FRICATIVES
Seletar has voiceless fricatives /s/ and /h/. There also appears to be a voiced velar
fricative [], which seems to be an allophone of /r/ and in free variation with [r], see
§4.4.4 above.
The voiceless alveolar fricative /s/ and voiceless glottal fricative /h/ are
presented here.
/s/ /sika/ 
/lus/ 
/h/ /hido/ 
/duh/ 
4.4.6 APPROXIMANT
Two approximants are present in Seletar, the palatal approximant /j/ and labio-velar
approximant /w/.
/j/ /ajam/ 
/w/ /awan/ 
4.4.7 LATERAL APPROXIMANT
A voiced alveolar lateral approximant is present in Seletar.
/l/ /ludh/ 
VOWELS
Seletar has six phonemic monophthongs. There are also various vowel-vowel
sequences. There is a clear distinction between the height of the vowels (close, mid,
open), as well as backness (front, central, back). The front and central vowels are
unrounded, while the back vowels are rounded. Length is not a contrasting factor.
4.5 Vowels 30
4.5.1 MONOPHTHONGS
Figure 4-6 illustrates the monophthongs in Seletar. Vowels in square brackets are
allophones to the phonemes on their left.
Front
Central
Back
Close
i [e]
u
Close-mid
e []
[]
o []
Open-mid
Open
a []
Figure 4-6 Seletar Monophthongs
There are 2 front unrounded vowels, /i/ and /e/. /i/ has allophone [e] when it
occurs in the closed ultimate syllable (by all consonants allowed in coda
position, except the glottal stop)
15
of disyllabic words. In other environments,
i.e. non-final syllable position, /i/ and /e/ are contrastive.
This phenomenon regarding the high and mid vowels is observed in Standard
Malay, where (i) there is in various degrees a non-phonemic lowering of high
vowels occurring in final closed syllables beginning with a consonant; (ii) in
penultimate syllables (and only here), high and mid-vowels can be in phonemic
contrast
/i/ [e] / ] C __ C #
(except prosodic )
[i] / elsewhere
Examples,
/kulit/ [kulet] 
/putih/ [puteh] 
/kaki/ [kaki] 
/ati/ [ati] 
/pisau/ [pisau] knife
15
The status of the glottal stop is discussed in the end of §4.4.1.
31 Phonetics and Phonology
/i/ versus /e/ (in monosyllabic words and non-final syllables)
Examples,
 
 
/timbul/ 
/temba/ 
/e/ seems to be in free variation with [].
/e/ [e]~[] in free variation
Examples,
/es/ [es] tomorrow
/ekou/ [kou] 
/len/ [len] 
/teo/ [to] look
There are central vowels // and /a/. // is contrasted with /a/ with a minimal
pair, but only in initial syllable position.
// /lma/ 
/a/ /lama/ 
// seem to be in free variation with [].
// []~[] in free variation
Examples,
/d/ [d] 
/d [d 
/ludh / [ludh] 
/sndp/ [sndp] 
/a/ has allophone [] when it occurs in the open, ultimate syllable of a disyllabic
word, or in final syllables ending with word prosodic . However, there are
exceptions like gəlap    lp] with final
syllable schwa although the syllable is a closed one, see below. Further
investigation is required to better understand this phenomenon.
/a/ [] / ] C __ #
prosodic
[a] / elsewhere
Examples,
4.5 Vowels 32
/dada/ [dad] 
/lama/ [lam] 
/du.a/ [du.] 
/bsal/ [bsal] 
/pand [panda 
Exceptions: /glap/ [glp] 
/blah/ [blh] 
/bulan/ [buln] 
/bukan/ [bukn] NEG
/tlam/ [tlm] 
However, the analysis of /a/ having allophone [] presents a problem in
biuniqueness. /a/ cannot have allophone [], which itself is in free variation with
[]. This complication is 
/a/ realized as [] in final open syllables is a phenomenon belonging to a
particular type of Standard Malay in Malaysia,    y
(Clynes & Deterding, 2011). Those that retain the /a/ as [a] in final open
-(Clynes & Deterding, 2011).
From the data, it seems that the consultant switches between the two varieties,
probably unconsciously, as stated in §3.4. It is currently hard to determine which
variety Seletar follows, the a- or schwa-variety. Because of this, there is
difficulty in writing a proper rule for realization of /a/. I have kept both the rules
for realization for both /a/ and //, although they violate the biuniqueness
principle. More research is required to determine their status.
There are 2 rounded back vowels, /u/ and /o/.
/u/
Examples,
/udin/ 
/kuku/ 
/tidu/ 
/o/ seems to be in free variation with []
/ o / [o]~[] in free variation
Examples,
/tumbo/ [tumbo] 
/guno  
/kokot/ [kkt] 
33 Phonetics and Phonology
/oe/ [] 
The close-mid vowels seem to have a front-back symmetry. The front
unrounded /e/ is symmetrical to the back rounded /o/ in that both [e] and [o] are
in free variation with their open-mid allophones [] and [] respectively. It has
been observed that vowel systems tend to be symmetrical (e.g. Trubetzkoy,
1969; Lass, 1984). The Seletar vowel system is quite symmetrical, with a typical
 shape, and also fits 1984, p. 94) observation that in vowel
systems in the matching of half-
open vowels (i.e. front pairs match back pairs) and isolation elsewhere (e.g. only
one open vowel).
4.5.2 VOWEL SEQUENCES
In Seletar, it is quite frequent to have a sequence of two vowels occurring next to each
other without an intervening consonant. These include vowel sequences with hiatus, i.e.
vowels belonging to different syllables, and vowel sequence without hiatus.
4.5.2.1 Vowel Sequence with Hiatus
The two juxtaposing vowels are separated into two syllables by a hiatus. Each of these
vowels forms the nucleus of its independent syllable. In the data, the hiatus is
transcribed using a period between the vowels. For example,
/la.ot/ 
/m.ah/ 
/ta.un/ 
4.5.2.2 Vowel Sequence without Hiatus
With the current data, it is unclear whether vowel sequences without hiatus occurring
in Seletar can be considered as diphthongs, as two separate vowels pronounced in rapid
succession, or as a single vowel followed by a glide. More research is required to
determine the status of the vowel sequence.
There are nine vowel sequence observed in the data. All vowel sequences only occur in
either monosyllabic words or the final syllable of di- or trisyllabic words, except /ai/.
/ai/ can occur in the first syllable, seen in aisaʔ . aisaʔ may be a compound with
ai SG as the first element, but further investigation is required. If this is the case, then
all vowel sequences are only allowed in either monosyllabic words or final syllables.
The final syllable can be open or closed.
Allophones are presented in square brackets.
ia /kian/ to 
iu /tium/ 
4.5 Vowels 34
ei /pasei/ 
ai  river
/aisa/ 
au /pisau/ 
ao /gao/ 
ua~[u] /smua/ [smua] 
/dua/ [du] 
ou /tlou / 
oi /siboi/ 
/tgoi/ 
Interestingly, there seems one word with vowel sequence of three successive vowels
tʃiao  The word is dated, according to the consultants, and has
been replaced by kajoh . To stick to only vowel-vowel sequences, I
decided to transcribe the word as / tao /.
This concludes the description of the phonetics and phonology of Seletar. The
next chapter assesses the lexicon of Seletar, and whether it resembles Malay or not.
5 Lexicon
MALAYIC
The Malayic languages are languages descended from a common proto language, Proto
Malayic. Malayic languages include Malay, Minangkabau, Kerinci, Banjarese,
Jakartanese etc. and various languages of inland western Borneo (see Smith, 2017 for
Malayic languages of Borneo). The Malayic languages have been classified by Adelaar
(1992 and 2005) by the co-occurrence of 14 phonological developments from Proto-
Malayo-Polynesian
16
:
1. *j > d
2. *z > j
3. *w > Ø
4. *R (and *r) > r
5. *q > h
6. *h > Ø (except between vowels, or if the following vowel is a schwa)
7. *-iw > -i
8. *uy > i
9. Split of *-ay to -ay and -i
10. Split of *-aw to -aw and -u
11. Cluster reduction
12. Nasal become homorganic to following stop
13. Final voiced stops became devoiced
14. Homorganic nasal accretion between initial schwa and following stop
The majority of Proto-Malayic lexemes are disyllabic, but there are also monosyllabic,
trisyllabic and even tetrasyllabic reconstructions. (Adelaar, 1992, p. 102).
MALAYIC AND MALAY
    -Malayic, while

mean the language is Malay or a dialect of Malay, for example Iban is considered
Malayic but is not derived from Malay, though it is related to Malay (Adelaar, 1992,
vi).
The internal subgrouping of Malayic languages is a matter of some debate (see
Adelaar, 1992, 2004; Blust, 2010). Thus, the classification of whether a language is a
dialect of Malay or Malayic is often not agreed upon. The uncertainty regarding the
16
The phonological developments are written in Austronesian conventions, where *z=d; *R=r;
*r= (Smith, 2017, p. 26).
37 Lexicon
homeland of Malay and Malayic also adds to the confusion between Malay and
Malayic.
Anderbeck (2012, pp. 283-284) consider all Sea Tribe (Orang Laut) languages
as Malayic according to the phonological developments listed above in §5.1 (namely
criteria 1,3,5 and a nearly-universal PMP * > a in final closed syllables).
RIAU SEA TRIBE DIALECT GROUP
From here onwards, words collected from the word list are discussed, concerning
-
Malay (but possibly Malayic) words present in the word list. All transcriptions in this
chapter are written in IPA, unless stated otherwise.
5.3.1 SOUND INNOVATIONS
Anderbeck (2012, p. 297) lists several features innovating from Proto-Malayic that
characterizes the Riau Sea Tribe varieties as a dialect group. These are:
*a raising after voiced obstruents in penultimate or ultimate position of the
syllable
Frequent *-V closing (usually with )
(less universal) NC > N: medial nasal-voiced obstruent consonant clusters
show a reduction or elimination of the obstruent component, e.g. tiŋgi >tiŋi
*r
17
devoicing (to h, or ) in initial, intervocalic, consonant cluster, or final
position
Loss of initial and final *r
Final open *a raised to , ,
An analysis of the lexical items collected reveals that Seletar fits quite well with the
above features.
*a raising
Anderbeck (2012, p. 285) observes the raising of *a is usually to /e/. Two words
from the list show *a raising in penultimate syllables. Some words do exhibit *a
raising in the ultimate syllable, to // and allophones []. For example:
Penultimate *a
*dahit > det to sew
*dari > dei finger
17
Proto-Malayic *r was a velar or uvular fricative (Adelaar, 1992, p. 86). In the reconstructed Proto-
Malayic phoneme inventory, Adelaar (1992, p. 102) puts *r in the velar place of articulation. I have
retained the *r and not written it as * or.
5.3 Riau Sea Tribe Dialect Group 38
Ultimate *a
*ludah > ludh 
) >  
*libar > leb 
Frequent *-V closing (usually with ʔ)
This is a frequent phenomenon in Seletar. For example,
*kaki > kaki 
*mata > mata 
*kutu > kutu 
(less universal) NC > N
Only one word has medial nasal-voiced obstruent consonant cluster, but the
voiced consonant is not eliminated.
 
*r devoicing (to h, χ or )
There is one instance of *r devoicing to /h/, in utaha utara in
Malay. However, this is a loan from Sanskrit. Instead, Proto-Malayic *r is usually
dropped in reflexes in Seletar. For example:
Initial
*rumput > umput 
Intervocalic
*prut > put 
Final
*tidur > tidu 
Loss of initial and final *r
Loss of *r occurs very frequently in Seletar, both in initial and final position. Other
than that, the loss of *r also occurs intervocalic positions, as seen from the point
above. Some reflexes of *r are /i/ and /l/. For example:
Initial
*rumah > uma 
Intervocalic
*darum > dom needle
39 Lexicon
Final
*tlur > tlou egg
*pasir > pasei 
*akar > akl 
Final open *a raised to ə, ɘ, ɨ
Final *a usually are raised to // in Seletar. The phenomenon occurs only
sometimes, not always. For example,
) >  
*lawa / *laba() > lab 
*dua() > du 
Final *a is not raised to // for the following,
*mata > mata 
*tuha() > tua 
*apa() > pa 
LEXICAL ITEMS
Anderbeck (2012, p. 297) also observe the following in the lexicon of the Riau Sea
Tribe varieties:
Lexically mainstream
kian / kiun / kiuh 
According to the Seletar word list collected, the language is indeed lexically
mainstream, as in most of the words are Malay, albeit with frequent contractions, see
Table 3. Of the 345 words collected, only around 40 are not of Malay origin, see Table
4
original Malay form making them quite unrecognizable, for example dʒɔʔ and pəmpaʔ.
I consider them now as non-Malay, but this will require deeper investigation to confirm.
Seletar seems to only have duplet kian and kiuh. kian is  
(Malay datang / ke sini), currently in use. However, kiuh ke sana) is
part of an older form of the language, according to the consultants (in their words
bahasa dulu). Currently, gun  instead of kiuh.
5.4.1 CONTRACTED MALAY WORDS
As Pelras (1972) and Blissett and Elzinga (2015) observed, in Seletar, most words are
almost exactly the same as in Malay, albeit with contractions. Here are some examples:
5.4 Lexical Items 40
ɲɛt reŋit, a type of
mosquito. This is also seen in Bintan Orang Laut as ŋet (Anderbeck, 2012, p.
292)
tɔʔ to  Proto-Malayo-*tawa, Proto Malayic
*tawaʔ and kətawa in Standard Malay. tɔʔ may be a possible contraction from
*tawa, with the labial-velar approximant [w] dropping out.
The most common sound change is the dropping of intervocalic /r/, /l/, /k/ and
/h/, for example
 Proto-Malayic *kəriŋ Seletar kiŋ
-Malayic *dahan Seletar dɤn
Proto-Malayic *makan Seletar man
-Malayic *(d-)aləm Seletar dam
41 Lexicon
Table 3 Examples of Seletar words which are contracted Malay words
Meaning
Seletar word
Malay word18
Proto-Malayic
(Adelaar, 1992)
he/she (3SG)
a19
di
- (POSS)
*ia
mosquito
t
am
*amuk
woman
tina
prmpuan / btin
(female)
person



to laugh
t
ktaw
*tawa
left
ki
kiri
*kA-iri / *kiba20
to walk / pathway
dan
dalan
*((mb)Ar-)dalan
to swim

br
*(mb)A-r
stomach / belly
put
prut
*prut
blood
dah
darah
*darah
head
pala
kpal
*hulu()
eat
man
makan
*makan
to lie down
b

to stand
di
diri
*diri
to sew
det
dahit
*dahit
to steal
tui
turi

bird



branch
dn
dahan
*dahan
dry

kr
*k
inside
dam
dalam
*(d-)alm
far
duh
dauh
*dauh
angry
mah
marah
straight
lus
lurus
18
Words here are spoken by a native speaker of Singapore Malay.
19
ɲaʔ: possibly from Malay -nya [-], which in Standard Malay is a suffix referring to third person
possessive. It was originally used as a full word in an older form of Malay (Benjamin, personal
communication).
20
*A in Proto-Malayic represents // or /a/ (Adelaar, 1992, p. 53).
5.4 Lexical Items 42
5.4.2 SUMATRAN MALAY LEXICAL ITEMS
Anderbeck (2012), in his analysis of several Orang Laut languages, observed very few
           
collected, which only has five words used in Sumatran Malay, and the first 3 are
mentioned by Anderbeck (2012, p. 292).
kəpaʔ  standard Malay sajap.
bataŋ pokoʔ.
kəat to potoŋ. Seen in Jambi Malay kərat.
ŋalaʔ ubi. Seen in Riau Malay məŋgalo
21
and Haji
məŋgɑlo (Anderbeck, 2007, p. 36).
tinaʔ perempuan. tinaʔ is from Proto-Malayo-Polynesian
*tina betina is the form in
Standard Malay, but usually used to refer to female animals. When used for
humans, it is an insulting term (Blust & Trussel, 2020). Betina is commonly
8, p. 107).
5.4.3 NON-MALAY LEXICAL ITEMS
- does not mean that the word is non-Malayic, but only means that the word
is not found in Standard Malay and (likely) other dialects of Malay, e.g. Sumatran
Malay. Some of the listed words are indeed non-Malayic (e.g. aih SG, ta.oh 
and kɔn , which are Aslian). For other words, I cannot say for sure what their
origins are, but many have similarities with words from Malayic languages of Borneo.
The reconstructed Proto-Malayic forms are based on lexicons from Standard
Malay, Minangkabau, Banjarese, Middle Malay, Iban and Jakartanese (Adelaar, 1992).
Where there is no reconstructable form for Proto-Malayic, a Proto-Malayo-Polynesian
(PMP) form is shown to illustrate the similarity or difference. As the Malayic proto-
forms do not take into account other languages of Borneo, I have carried out comparison
with various Borneo languages using data from Smith (2017).
Table 4 Non-Malay words in Seletar
Meaning
Seletar word
Malay word
Proto-Malayic
(Adelaar, 1992)
REPTILES
lizard
t
tita
*cecek (PMP)
monitor lizard
d
biawa
*biawak
21
Ubi menggalo (in Indonesian orthography) is a type of cassava plant native to Riau, commonly eaten
by the Orang Sakai of Riau (Parsudi, 1995, p. 109)
43 Lexicon
MONKEYS
monkey
p
moet
gibbon
ta.oh

INSECTS
flying termite (alates)
kns
kelekatu
stingless bee
batau
kelulut
weaver ant

kerengga
butterfly
klba
kupu-kupu /
ram-ram
OTHER ANIMALS
dog

and
*asu
tiger

harimau
*hArimaw
insect bat
klntt
klawar
*kAlu
pig
isum
babi
*babi
animal
sibi

BODY PARTS
hand
kkt


neck
g
lehe
*lihr
body / life

badan
*awak
thigh
pmpa
ph
*paha()
PEOPLE
person
kn


friend
kn
kawan
VERBAL (VERBS
ADJ, ADV)
soft
mt
lmbut
*lmbut
murky, cloudy
bn
kroh
clear, transparent
d
drneh
to weave
but
anam
*aam
to come
kian

*dat
to hide
sndp
smbui
*buni
skinny
tgi
kurus
*kurus (PMP)
a little
ktt
sdiket / siket
now
k
kini / s
before / long ago
min
dahulu
*di qulu (PMP)
5.4 Lexical Items 44
OTHERS
I (1SG)
am
aku / saj
*aku
you (2SG)
a.ih
kamu
*kau()
they (3PL)
a.ih
mrek
*sida
what
ana
ap
*apa
who
aisa
siap
*si-apa / *sai
why

m / knap
no, not (NEG)
nta / bukn
tida / ta
*-da
here (PROX)
sika
sini
there (DIST); go there
gun / kiuh (dated)
san
basket
gag
bakol
*bakul (PMP)
Some words here have already been pointed out and discussed in the literature,
namely aŋkɔʔ isum kɔn  and kɔkɔt 
aŋkɔʔ : Pelras (1972) notes that there are similar words in Borneo languages
to aŋkɔʔ, which was given as uko  
state what these specific languages are. Anderbeck (2012, p. 293) notes that the
       kojoʔ, and related to
Kalimantan Malayic ukuj.
Taking this direction, f
the languages of Borneo (around 80), I found some similarity.
45 Lexicon
Language
Word for ‘dog’
(Smith, 2017)
Ngorek
aho
Mpraa
haw
Long Naah
aho
Taboyan
koko
Paser
koko / asu
Benuaq
koko
Tunjung
koko
Seberuang
ukuj
Mualang
ukoj
Lun Dayeh (Long
Bawan)
oko
isum : Pelras thinks the word is similar to ilum in Long Gelat. 
(2017) list, this proves to be true, though the word might 
original collection in the 70s. Here is a list of similar words in Borneo languages:
Language
Word for ‘pig’ (Smith, 2017)
Long Gelat
dum
Modang
dim
Gaai
dim
Kelai
dim
kɔn : this may be the most well-discussed in literature, first by Carey (1976,
p. 279) then Benjamin (2001, pp. 116-117). Both scholars propose the origin of the
word to be Mon-kun:
Old Mon, Middle Mon kon  person to whom one stands in
loco parentis’
It also appears in Southern Aslian as koɲ Here,
       . Usually for a traditionally
segmentary group, those who are friends are in the same ethnic group.
kɔkɔt : Pelras (1972) only notes that both Seletar and Duano have this word.
Anderbeck (2012) observes that Duano and Jakun (collected by Skeat and Blagden,
5.4 Lexical Items 46
1906) share this too. The extinct Biduanda Kallang also has this word. One Borneo
language, i.e. Kendayan has this word 
Language
Word for ‘hand’
Duano (Seidlitz, 2007)
kukt
Jakun (Skeat & Blagden,
1906, p. 408)
kokot
Jakun (Seidlitz, 2005)

Biduanda Kallang (Skeat
& Blagden, 1906, p. 408;
extinct)
kkt22
Kendayan
kokot
The following words have not yet been discussed in literature. Here, I will
present some similar words found in other languages. The Borneo languages data
are from Smith (2017).
dʒɔʔ  bɘʤɘwɐʔ (Seidlitz,
2007, p. 12). In Musi (Palembang Malay)   is biantʃak (Blust &
) and in Lampung bejawak (Ainun, 2016, p. 144). Maybe
the word comes from the contraction of something similar? Borneo languages also
have some similar words.
22
Not in IPA but spelt based on Romanised Malay used in the Straits Settlements (Skeat & Blagden,
1906, p. 796). The macron represents a long vowel. The acute accent represents the stressed vowel.
47 Lexicon
Language
Word for ‘monitor lizard
Duano
bw
Musi (Palembang Malay)
biantak
Lampung
bejawak23
Jankang
bodo
Ribun
budowo
Benyadu
padawak
Bakumpai
bdai
Ngaju
bdawak
ta.oh -Aslian as *tawɔɔh 
with reflexes, for example, in Semelai tawɔ (Kruspe, 2004, p. 9) and Semnam,
tawɔːh (Burenhult & Wegener, 2009).
kəlɔntɔt Borneo language Kejaman has kəlutuə̯ ŋ .
kələbaʔ kelebek. It is not an Arabic borrowing, as

(. Many Borneo languages have similar words
Language
Word for ‘butterfly
Beketan
kr
Punan Lisum
kr
Ukit
kl
Vo
bl
Tunjung
kllma
West Penan
kbava
Aoheng
k
sibɔi : Borneo language Punan Bah oi  and Sekapan ui 
exhibit some similarity to the second syllable of sibɔi.
gɔŋgɔŋHowever,
this word may also be connected to Malay kerongkong [kəroŋkoŋ] ,
23
This word is not written in IPA, but in Lampung orthography mentioned in Ainun (2016, p. 13). The

 and e. My guess is they are // and // respectively from
th/ and /k/.
5.4 Lexical Items 48
Proto-Malayo-Polynesian *kaRuŋkuŋ  uvula    

Language
Word for ‘neck’
Hovongan
t
Hliboi Bidayuh
gagu

Tunjung

Basap

pəmpaʔ, pun paa pupor;
Narum pupawɣ; Kiput pupun. Could pəmpaʔ originally be a compound, with the
second element a contraction from standard Malay pəhə?
mɛt lɛʔɛmɛt 
anaʔ : anaʔ seems closer to Proto-Malayo-Polynesian *-anu than Proto-
Malayic *apa.
kian 
kian / kiun / kiuh 
may be bi-morphemic, embedded within the Malay ke 
Evidence for this is Standard Malay sekian 
Malay word after all. He also notes that Duano has something like this, but not
completely: kia/kəna/kəsut.
gun / kiuh5.4, kiuh is a dated usage of the current
word gun. Anderbeck (2012, p. 292) pointed out that Jakun documented by
Seidlitz (2005) has tʃun z (2005) it is written as tʃʊdn.
aih 2SG  ajih  
(2012, p. 274). He states that this is a probable cognate with the extinct language
Kenaboi
24
word jei, and that Aslian languages have something like ajih 

kɛtɛt saketeʔ 
mɯin       Borneo language Kanowit mudəy, from
standard Malay mudə 
24
The classification of Kenaboi is disputed; some say it is Aslian, or a mix of Aslian and Austronesian
(Hajek, 1998) while others consider it an isolate (Skeat & Blagden, 1906).
49 Lexicon
ŋəʔ Kayanic and Punan languages in Borneo have words within a
compound with initial nasals (naʔ, neʔ, meʔ), and Kendayan has ŋahe, while
Seberuang and Mualang have ŋapa 
from standard Malay məŋapə, leaving only the coda of the məŋ- and the final open
vowel.
For the rest of the words in the list I was not able to find any corresponding words
in either the neighbouring Duano, Jakun or in the Borneo languages.
5.4.4 OTHERS
Seletar has English loans. From the word list, there are two kɔnə, from English
    endʒin      paŋkɛŋ tidu. paŋkɛŋ is
        paŋkɛŋ tidu means

5.4.5 REMARKS
Most words in Seletar are similar to mainstream Malay, albeit with some pronunciation
differences. Contraction of the Malay forms is a common phenomenon in Seletar.
However, a part of the lexicon seems to resemble Malay varieties of Sumatra, rather
than Peninsular Malaysia varieties. There are also non-Malay words in Seletar, though
their origin would require further research. I have found similarities in surrounding
languages (Jakun and Duano), Borneo languages and Aslian languages, though how
these words entered Seletar is still a mystery.
This concludes the chapter on Lexicon. In the following chapter, I give some
examples of polysemy in Seletar.
6 Polysemy
Although many words in Seletar resemble that of Malay in form exactly, from the data
I have gathered, the semantics of some words differ. Specifically, these words exhibit
polysemy that is not present in their Malay counterparts.
Here I present 3 polysemous words in Seletar. The words can be found in both
Malay and Seletar. Words on the left are in Malay orthography, while those on the right
are the Seletar pronunciations in IPA.
Kayu [kaju]
Mata [mata]
Tanah [tanah]
WOOD/TREE
Kayu [kaju]  [kaju], kayu not only means
 Moreover, b  are referred to with batang
 kayukayu is polysemous,
     t can also be represented with a compound
batang kayu, or with just batang itself, which in Malay, refers to something long and
cylindrical. The polysemy seems logical, as trees are the source of wood, and wood,
before additional woodwork, was a huge part of the tree (usually the most salient part
of the tree: the trunk). Cognitively, they are in a part-whole relationship.
The question is, wh

primarily studied by Witkowski, Brown and Chase (1981), who surveyed 66 languages
  44 out of 66 of the languages



many languages involve the addition of another word, i.e. an overt mark, to the


more marked and less salient one. Blust (1974, p. 5) reconstructed the Proto-
*puqun ni kaSiw GEN 
*puqun            eferential

51 Polysemy
Batang can
be added to kayu Batang, in this case, is the overt marker. This overt
marker is optional, as kayu 
         
polysemy with societal complexity. They observed that societies with this polysemy
usually are small-scale, traditional societies, while those separating the meanings tend
to be from larger nation states. As societies always develop from small to large, it is
         
polysemy in the past, but over time, as societies grew larger, lost the polysemous
meaning. The authors also suggest that languages with overt marker
through a process of losing the polysemy, eventually the  part is dropped, and
the overt marker will designate 
basic naming level of the society, where significance of life-form categories (e.g. tree,
bird, fish) take precedence over generic categories (e.g. wood, oak, robin) as the society
grows in complexity (Witkowski, et al., 1981, p. 9). With the increased saliency of
        present     
different terms.
The Orang Seletar are a minority group and are, indeed, a small society, and to
some extent still practice the hunter-gatherer lifestyle. It has been observed that the
original way of living on boats is no longer feasible, as noted in §1.4.2, and the society
is slowly incorporating mainstream culture. It seems that the language, to a small extent,

in Seletar, where sometimes batang (overt marker), instead of batang kayu, is used to
kayu may be dropped entirely. But, currently, batang, batang
kayu and kayu    
Seletar are bilingual in Malay might also cause t
kayu and pokok [poko], may influence
the polysemous use of kayu in Seletar.
Witkowski et al. (1981) also suggests that the shift and loss of polysemous
  he results of the increasing sophistication of woodworking
technology. For smaller societies, the wood itself is not changed drastically when used
to build shelters, transport, tools etc. and still resembles the tree in many forms. Thus,

society and technology, the original source of wood the tree is changed so much
 -   
reference.
This somewhat applies to the Orang Seletar, who have given up their traditional
wooden pa.uʔ (sailboat) for fiberglass boats. However, in the villages I have visited
6.2 Eye/Seed 52
(Kampung Sungai Temon and Kampung Simpang Arang), the use of wood for stilted
houses and platforms is still very traditional, in the sense that, visually, the affinity to
trees is still fairly obvious.
EYE/SEED
Mata in Seletar [mata] and Malay [mat
polysemy, and seed is indicated
by the term benih [bneh]. The polysemy might come from the similarity in physical
appearance between the seed and the eye. Both are round and small relative to the face
and the fruit. There might also be some similarity in how the seed and the eye are
perceived cognitively, as suggested by Brown and Witkowski (1983):

while face and fruit comprise the periphery of eye and seed. Thus formally
speak
(p. 76).


& Witkowski, 1983, p. 77). There also might be some metaphorical extensions that
resulted 

   mata? Brown and
Witkowski (1983, pp. 79-82) suggest that the eye is naturally salient, in that infants
were shown to pay more attention to human eyes very early in life, in comparison to
other facial features. The authors support the view of the natural saliency of the eye
with evidence from reconstructed proto-forms of various protolanguages exhibiting
-  

because of their

             
on top of the
          

l link between

53 Polysemy
The words kayu and mata used in Seletar introduced me to a larger set of
languages sharing the same polysemic feature, and that some cultures, genetically and
geographically unrelated, may share similar cognitive processes when it comes to
assigning meanings to words. The two words presented up till this point are not to
           
show similarities in certain small-scale cultures in their adaptation to their
environments.
SOIL/ MUD
Tanah [tanah] in Malay and 
tanah met [tana m

25
the interviewee called the
mudlands [tana], and stressed that the Malays call it differently as lumpur [lumpo(r)]
. He also said the tanah is where they make their living. Pelras (1972, p. 150) also
        tana      

As mentioned above, tanah  In this case, the word

meanings is a type of soil
added with water.
