Makassarese language

Makasar
basa Mangkasaraʼ ᨅᨔ ᨆᨀᨔᨑ 𑻤𑻰𑻥𑻠𑻰𑻭 بَاسَ مَڠْكَاسَرَءْ‎
Mangkasaraʼ written in the Makasar script and Lontara script
Pronunciation	[maŋˈkasaraʔ]
Native to	Indonesia
Region	South Sulawesi (Sulawesi)
Ethnicity	Makassarese
Native speakers	2.1 million (2000 census)
Language family	Austronesian Malayo-Polynesian South Sulawesi Makassaric Makasar
Dialects	Gowa Turatea Maros-Pangkep
Writing system	Lontara (present) Latin (present) Serang (Makassar Annals, religious purpose) Old Makassarese (historical)
Language codes
ISO 639-2	mak
ISO 639-3	mak
Glottolog	maka1311
Makasar language   other Makassar languages.

Makassarese (basa Mangkasaraʼ, IPA: [basa maŋˈkasaraʔ]), sometimes called Makasar, Makassar, or Macassar, is a language of the Makassarese people, spoken in South Sulawesi province of Indonesia. It is a member of the South Sulawesi group of the Austronesian language family, and thus closely related to, among others, Buginese, also known as Bugis. The areas where Makassarese is spoken include the Gowa, Sinjai, Maros, Takalar, Jeneponto, Bantaeng, Pangkajene and Islands, Bulukumba, and Selayar Islands Regencies, and Makassar. Within the Austronesian language family, Makassarese is part of the South Sulawesi language group, although its vocabulary is considered divergent compared to its closest relatives. In 2000, Makassarese had approximately 2.1 million native speakers.

Classification

Makassarese is an Austronesian language from the South Sulawesi branch of the Malayo-Polynesian subfamily, specifically the Makassaric group, which also includes the Konjo language (both Highland and Coastal varieties) and the Selayar language. The Konjo and Selayar language varieties are sometimes considered dialects of Makassarese. As part of the South Sulawesi language family, Makassarese is also closely related to the Bugis, Mandar, and Toraja-Saʼdan languages.

In terms of vocabulary, Makassarese is considered the most distinct among the South Sulawesi languages. The average percentage of vocabulary similarity between Makassarese and other South Sulawesi languages is only 43%. Specifically, the Gowa or Lakiung dialect is the most divergent; the vocabulary similarity of this dialect with other South Sulawesi languages is about 5–10 percentage points lower compared to the vocabulary similarity of Konjo and Selayar with other South Sulawesi languages. However, etymostatistical analysis and functor statistics conducted by linguist Ülo Sirk shows a higher vocabulary similarity percentage (≥ 60%) between Makassarese and other South Sulawesi languages. These quantitative findings support qualitative analyses that place Makassarese as part of the South Sulawesi language family.

Dialect

The language varieties within the Makassaric group form a dialect continuum. A language survey in South Sulawesi conducted by linguists and anthropologists Charles and Barbara Grimes separated the Konjo and Selayar languages from Makassarese. Meanwhile, a subsequent survey by linguists Timothy Friberg and Thomas Laskowske divided the Konjo language into three varieties: Coastal Konjo, Highland Konjo, and Bentong/Dentong. However, in a book on Makassarese grammar published by the Center for Language Development and Cultivation, local linguist Abdul Kadir Manyambeang and his team include the Konjo and Selayar varieties as dialects of Makassarese.

Excluding the Konjo and Selayar varieties, Makassarese can be divided into at least three dialects: the Gowa or Lakiung dialect, the Jeneponto or Turatea dialect, and the Bantaeng dialect. The main differences among these varieties within the Makassar group lie in vocabulary; their grammatical structures are generally quite similar. Speakers of the Gowa dialect tend to switch to Indonesian when communicating with speakers of the Bantaeng dialect or with speakers of the Konjo and Selayar languages, and vice versa. The Gowa dialect is generally considered the prestige variety of Makassarese. As the dialect spoken in the central region, the Gowa dialect is also commonly used by speakers of other varieties within the Makassaric group.

Distribution

According to a demographic study based on the 2010 census data, about 1.87 million Indonesians over the age of five speak Makassarese as their mother tongue. Nationally, Makassarese ranks 16th among the 20 languages with the most speakers. Makassar is also the second most-spoken language in Sulawesi after Bugis, which has over 3.5 million speakers.

The Makassarese language is primarily spoken by the Makassar people, although a small percentage (1.89%) of the Bugis people also use it as their mother tongue. Makassarese speakers are concentrated in the southwestern peninsula of South Sulawesi, particularly in the fertile coastal areas around Makassar, Gowa Regency, and Takalar Regency. The language is also spoken by some residents of Maros Regency and Pangkajene and Islands Regency to the north, alongside Bugis. Residents of Jeneponto and Bantaeng Regencies generally identify themselves as part of the Makassarese-speaking community, although the varieties they speak (the Jeneponto or Turatea dialect and the Bantaeng dialect) differ significantly from the dialects used in Gowa and Takalar. The closely-related Konjo language is spoken in the mountainous areas of Gowa and along the coast of Bulukumba Regency, while the Selayar language is spoken on Selayar Island, to the south of the peninsula.

Current status

Makassarese is one of the relatively well-developed regional languages in Indonesia. It is still widely used in rural areas and parts of Makassar. Makassarese is also considered important as a marker of ethnic identity. However, in urban communities, code-switching or code-mixing between Makassar and Indonesian is common. Some urban Makassar residents, especially those from the middle class or with multiethnic backgrounds, also use Indonesian as the primary language in their households. Ethnologue classifies Makassar as a 6b (Threatened) language on the EGIDS scale, indicating that although the language is still commonly used in face-to-face conversations, the natural intergenerational transmission or teaching of the language is beginning to be disrupted.

Phonology

The following description of Makassarese phonology is based on Jukes (2005).

Vowels

Makassarese has five vowels: /a/, /e/, /i/, /o/, /u/. The mid vowels are lowered to [ɛ] and [ɔ] in absolute final position and in the vowel sequences /ea/ and /oa/.

Makassarese vowels

	Front	Central	Back
Close	i		u
Mid	e		o
Open		a

The vowel phoneme /e/ tends to be realized as the open-mid vowel [ɛ] when it is at the end of a word or before a syllable containing the sound [ɛ]. Compare, for instance, the pronunciation of /e/ in the word leʼbaʼ [ˈleʔ.baʔ] 'already' with mange [ˈma.ŋɛ] 'go to'. The phoneme /o/ also has an open-mid allophone [ɔ] when it is at the end of a word or precedes a syllable containing the sound [ɔ], as seen in the word lompo [ˈlɔ̃m.pɔ] 'big' (compare with órasaʼ [ˈo.ra.saʔ] 'heavy'). Regardless of their position within a word, some speakers tend to pronounce these two vowels with a higher (closer) tongue position, making their pronunciation approach that of the phonemes /i/ and /u/.

Vowels can be pronounced nasally when they are around nasal consonants within the same syllable. There are two levels of nasalization intensity for vowels: strong nasalization and weak nasalization. Weak nasalization can be found on vowels before nasal consonants that are not at the end of a word. Strong nasalization can be found on vowels before final nasal consonants or generally after nasal consonants. Nasalization can spread to vowels in syllables after nasal vowels if there are no consonants blocking it. However, the intensity of nasalization in vowels like this is not as strong as in the vowels before them, as in the pronunciation of the word niaʼ [ni͌.ãʔ] 'there is'.

Consonants

There are 17 consonants in Makassarese, as outlined in the following table.

Makassarese consonants

		Labial	Alveolar	Palatal	Velar	Glottal
Nasal		m	n	ɲ ⟨ny⟩	ŋ ⟨ng⟩
Plosive	voiceless	p	t	c	k	ʔ ⟨ʼ⟩
	voiced	b	d	ɟ ⟨j⟩	ɡ
Fricative			s			(h)
Semivowel				j ⟨y⟩	w
Lateral			l
Trill			r

Makassarese consonants except the glottal stop and voiced plosives can be geminated. Some instances of these might result from Proto-Malayo-Polynesian schwa phoneme *ə (now merged into a), which geminated the following consonant (*bəli > *bəlli > balli 'to buy, price' (compare Indonesian beli), contrasting with bali 'to oppose').

The phoneme /t/ is the only consonant with a dental pronunciation, unlike the phonemes /n d s l r/, which are alveolar consonants. The voiceless plosive phonemes /p t k/ are generally pronounced with slight aspiration (a flow of air), as in the words katte [ˈkat̪.t̪ʰɛ] 'we', lampa [ˈlam.pʰa] 'go', and kana [ˈkʰa.nã] 'say'. The phonemes /b/ and /d/ have implosive allophones [ɓ] and [ɗ], especially in word-initial positions, such as in balu [ˈɓa.lu] 'widow', and after the sound [ʔ], as in aʼdoleng [aʔ.ˈɗo.lẽŋ] 'to let hang'. These two consonants, especially /b/ in word-initial positions, can also be realized as voiceless consonants without aspiration. The palatal phoneme /c/ can be realized as an affricate (a stop sound with a release of fricative) [cç] or even [tʃ]. The phoneme /ɟ/ can also be pronounced as an affricate [ɟʝ]. Jukes analyzes both of these consonants as stop consonants because they have palatal nasal counterparts /ɲ/, just as other oral stop consonants have their own nasal counterparts.

Phonotactics

The basic structure of syllables in Makassarese is (C₁)V(C₂). The position of C₁ can be filled by almost any consonant, while the position of C₂ has some limitations. In syllables located at the end of a morpheme, C₂ can be filled by a stop (T) or a nasal (N), the pronunciation of which is determined by assimilation rules. The sound T assimilates with (is pronounced the same as) voiceless consonants except [h], and is realized as [ʔ] in other contexts. The sound N is realized as a homorganic nasal (pronounced at the same articulation place) before a stop or nasal consonant, assimilates with the consonant's /l/ and /s/, and is realized as [ŋ] in other contexts. On the other hand, in syllables within root forms, Makassarese contrasts an additional sound in the C₂ position besides K and N, which is /r/. This analysis is based on the fact that Makassarese distinguishes between the sequences [nr], [ʔr], and [rr] across syllables. However, [rr] can also be considered as the realization of a geminate segment rather than a sequence across syllables.

CV patterns

	V		o		'oh' (interjection)
	CV		ri		PREP (particle)
	VC		uʼ		'hair'
	CVC		piʼ		'birdlime'
	VV		io		'yes'
	VVC		aeng		'father'
	CVV		tau		'person'
	CVVC		taung		'year'
	VCVC		anaʼ		'child'
	CVCV		sala		'wrong'
	CVCVC		sabaʼ		'reason'
	CVCCVC		leʼbaʼ		'already'
	CVCVCV		binánga		'river'
	CVCVCVC		pásaraʼ		'market'
	CVCVCCV		kalúppa		'forget'
	CVCCVCVC		kaʼlúrung		'palm wood'
	CVCVCVCVC		balakeboʼ		'herring'
	CVCVCVCCVC		kalumanynyang		'rich'

The sounds /s l r/ can be categorized as non-nasal continuous (sounds produced without fully obstructing the flow of air through the mouth) consonants, and none of them can occupy the final position of a syllable except as part of a geminate consonant sequence. Basic words that actually end with these consonants will be appended with an epenthetic vowel identical to the vowel in the preceding syllable, and closed with a glottal stop [ʔ], as in the words ótereʼ /oter/ 'rope', bótoloʼ /botol/ 'bottle', and rántasaʼ /rantas/ 'mess, untidy'. This additional element is also referred to as the "VC-geminate" (echo-VC) sequence, and it can affect the position of stress within a word.

Generally, base words in Makassarese consist of two or three syllables. However, longer words can be formed due to the agglutinative nature of Makassarese and the highly productive reduplication process. According to Jukes, words with six or seven syllables are commonly found in Makassarese, while base words with just one syllable (that are not borrowed from other languages) are very rare, although there are some interjections and particles consisting of only one syllable.

All consonants except for /ʔ/ can appear in initial position. In final position, only /ŋ/ and /ʔ/ are found.

Consonant clusters only occur medially and (with one exception) can be analyzed as clusters of /ŋ/ or /ʔ/ + consonant. These clusters also arise through sandhi across morpheme boundaries.

	nasal/lateral					voiceless obstruents					voiced stops + r
	m	n	ɲ	ŋ	l	p	t	c	k	s	b	d	ɟ	ɡ	r
/ŋ/	mm	nn	ɲɲ	ŋŋ	ll	mp	nt	ɲc	ŋk	ns	mb	nd	ɲɟ	ŋg	nr
/ʔ/	ʔm	ʔn	ʔɲ	ʔŋ	ʔl	pp	tt	cc	kk	ss	ʔb	ʔd	ʔɟ	ʔg	ʔr

The geminate cluster /rr/ is only found in root-internal position and cannot be accounted for by the above rules.

Sequences of like vowels are contracted to a single vowel; e.g., sassa 'to wash' + -ang 'nominalizing suffix' > sassáng 'laundry', caʼdi 'small' + -i 'third person' > caʼdi 'it is small'.

Stress

The stress is generally placed on the penultimate (second-last) syllable of a base word. In reduplicated words, secondary stress will be placed on the first element, as in the word ammèkang-mékang /amˌmekaŋˈmekaŋ/ 'to fish (casually)'. Suffixes are generally counted as part of the phonological unit receiving stress, while enclitics are not counted (extrametrical). For example, the word gássing 'strong', if the benefactive suffix -ang is added, becomes gassíngang 'stronger than' with stress on the penultimate syllable, but if given the first-person marker enclitic =aʼ, it becomes gássingaʼ 'I am strong', with stress on the antepenultimate syllable (third-last).

Other morphemes counted as part of the stress-bearing unit include the affixal clitic, marking possession, as in the word tedóng=ku (buffalo=1.POSS) 'my buffalo'. Particularly for the definite marker ≡a, this morpheme is counted as part of the stress-bearing unit only if the base word it attaches to ends in a vowel, as in the word batúa 'the stone'—compare with the stress pattern in kóngkonga 'the dog', where the base word ends in a consonant. A word can have stress on the preantepenultimate (fourth-last) syllable if a two-syllable enclitic combination such as =mako (PFV =ma, 2 =ko) is appended; e.g., náiʼmako 'go up!' The stress position can also be influenced by the process of vocalic degemination, where identical vowels across morphemes merge into one. For example, the word jappa 'walk', when the suffix -ang is added, becomes jappáng 'to walk with', with stress on the ultimate (last) syllable.

The stress on base words with VC-geminate always falls on the antepenultimate syllable; for example, lápisiʼ 'layer', bótoloʼ 'bottle', pásaraʼ 'market', and Mangkásaraʼ 'Makassar', because syllables with VK-geminate are extrametrical. However, the addition of suffixes -ang and -i will remove this epenthetic syllable and move the stress to the penultimate position, as in the word lapísi 'to layer'. Adding the possessive clitic suffix also shifts the stress to the penultimate position but does not remove this epenthetic syllable, as in the word botolóʼna 'its bottle'. Meanwhile, the addition of the definite marker and enclitics neither remove nor alter the stress position of this syllable, as in the words pásaraka 'that market' and appásarakaʼ 'I'm going to the market'.

Grammar

Pronouns

Personal pronouns in the Makassar language have three forms, namely:

• free forms;
• proclitics that cross-reference S and P arguments ('absolutive');
• and enclitics that cross-reference A arguments ('ergative').

The following table shows these three forms of pronouns along with possessive markers for each series.

Personal pronouns

	Free pronouns (PRO)	Proclitic (ERG)	Enclitic (ABS)	Possessive marker (POSS)
1SG	(i)nakke	ku=	=aʼ	=ku
1PL.INCL/2POL	(i)katte	ki=	=kiʼ	=ta
1PL.EXCL	†(i)kambe	—	*=kang	†=mang
2FAM	(i)kau	nu=	=ko	=nu
3	ia	na=	=i	=na

The first person plural inclusive pronouns are also used to refer to the second person plural and serve as a form of respect for the second person singular. The first person plural pronoun series ku= is commonly used to refer to the first person plural in modern Makassar; pronouns kambe and possessive marker =mang are considered archaic, while the enclitic =kang can only appear in combination with clitic markers of modality and aspect, such as =pakang (IPFV =pa, 1PL.EXCL =kang). The plural meaning can be expressed more clearly by adding the word ngaseng 'all' after the free form, as in ia–ngaseng 'they all' and ikau–ngaseng 'you all', or before the enclitic, as in ngaseng=i 'they all'. However, ngaseng cannot be paired with proclitics.

Nouns and noun phrases

Characteristics and types of nouns

Nouns can be defined syntactically as words with the potential to form or head noun phrases (including relative clauses), and act as arguments in a predicate and thus be cross‒referenced with clitic pronouns. They have the potential to be both possessor or possessed (head and modifier) in possessive constructions, with this relationship being marked morphologically with affixal enclitics on the possessed noun phrase. They may also be marked for definiteness with the affixal enclitic ≡a. Like the other major classes they may also function as predicates. Below illustrates all of these major points::

(-)

Anaʼnai karaenga

anaʼ

child

≡na

≡3.POSS

=i

=3

karaeng

king

≡a

≡DEF

anaʼ ≡na =i karaeng ≡a

child ≡3.POSS =3 king ≡DEF

he is the son of the king.'

In addition, nouns can also be specified by demonstratives, modified by adjectives, quantified by numerals, become complements in prepositional phrases, and become verbs meaning 'wear/use [the noun in question]' when affixed with the prefix aK-.

Nouns that are usually affixed with the definite clitic ≡a and possessive markers are common nouns. On the other hand, proper nouns such as place names, personal names, and titles (excluding kinship terms) are usually not affixed with definiteness and possessive markers but can be paired with the personal prefix i- like pronouns.

Some common nouns are generic nouns that often become the core of a compound word, such as the words jeʼneʼ 'water', tai 'excrement', and anaʼ 'child'. Examples of compound words derived from these generic nouns are jeʼneʼ inung 'drinking water', tai bani 'beeswax' (literal meaning: 'bee excrement'), and anaʼ baine 'daughter'. Kinship terms that are commonly used as greetings are also classified as common nouns, such as the words mangge 'father', anrong 'mother', and sariʼbattang 'sibling'. Another example is the word daeng which is used as a polite greeting in general, or by a wife to her husband.

The other main noun group is temporal nouns, which usually appear after prepositions in adjunct constructions to express time. Examples of temporal nouns are clock times (such as tetteʼ lima '5.00 [five o'clock]'), estimated times based on divisions of the day (such as bariʼbasaʼ 'morning'), days of the week, as well as dates, months, and seasons.

Derived noun

Derived nouns in Makassarese are formed through several productive morphological processes, such as reduplication and affixation with pa-, ka-, and -ang, either individually or in combination. The following table illustrates some common noun formation processes in Makassarese:

Formation of derived nouns

Process		Productive meanings	Samples	Note
reduplication		diminution or imitation	tau 'people' → tau-tau 'statue, doll'
suffix pa-	pa-NOMINAL/VERBAL ROOT	actor, creator, or user	jarang 'horse' → pajarang 'rider'; botoroʼ 'gamble' → pabotoroʼ 'gambler'
	pa-VERB BASE	instrument	akkutaʼnang 'ask' → pakkutaʼnang 'question'; anjoʼjoʼ 'point' → panjoʼjoʼ 'index finger, pointer'
	pa-DERIVED VERBAL BASES	instrument	appakagassing 'fortifiying' → pappakagassing 'tonic, fortifying medicine or drink'
	pa>VERB BASE<ang	place or time	aʼjeʼneʼ 'bathe' → paʼjeʼnekang 'bathing place, bathtime'; angnganre 'eat' → pangnganreang 'plate'
	paK>ADJECTIVE<ang	someone who is easily ADJ, inclined to be ADJ	garring 'sick' → paʼgarringang 'sickly person'
confix ka>...<ang	ka>ADJECTIVE<ang	ADJ-ness	kodi 'bad' → kakodiang 'badness'
	ka>REDUPLICATED ADJECTIVAL<ang	peak of ADJ	gassing 'strong' → kagassing-gassingang 'greatest strength'
	ka>BASIC VERB<ang	state or process	battu 'come' → kabattuang 'arrival'
suffix -ang		instrument	buleʼ 'carry on shoulders' → bulekang 'sedan chair'

There are some exceptions to the general patterns described above. For example, reduplication of the word oloʼ 'worm' to oloʼ-oloʼ results in a broadening of meaning to 'animal'. The affixation of pa- to a verb base does not always indicate an instrument or tool, for example paʼmaiʼ 'breath, character, heart' (as in the phrase lompo paʼmaiʼ 'big-hearted') which is derived from the word aʼmaiʼ 'to breathe'. The affixation of pa>...<ang to the verb base ammanaʼ 'to give birth' results in the word pammanakang meaning 'family', although it is possible that this word was originally a metaphor ('place to have children').

Writing systems

1. Makasar script; 2. Lontara script; 3. Serang script; 4. Latin script

Although Makassarese is now often written in Latin script, Makassarese has been traditionally written with Lontara script and Makasar script, which once was used also to write important documents in Bugis and Mandar, two related languages from Sulawesi. Further, Makassarese was written in the Serang script, a variant of the Arabic-derived Jawi script. Texts written in the Serang script are relatively rare, and mostly appear in connection with Islam-related topics. Parts of the Makassar Annals, the chronicles of the Gowa and Tallo' kingdoms, were also written using the Serang script.

Latin based system

The current Latin-based forms:

Majuscule forms (uppercase)
A	B	C	D	E	G	H	I	J	K	L	M	N	Ng	Ny	O	P	R	S	T	U	W	Y	(ʼ)
Minuscule forms (lowercase)
a	b	c	d	e	g	h	i	j	k	l	m	n	ng	ny	o	p	r	s	t	u	w	y	(ʼ)
IPA
a	b	c	d	e	g	h	i	ɟ	k, ʔ	l	m	n	ŋ	ɲ	o	p	r	s	t	u	w	j	ʔ

There have been numerous methods of writing Makassarese in roman script. Some only appeared in short word lists or explanations of the writing system, such as Raffles (1817), Crawfurd (1820), and Thomsen (1832) (see Noorduyn 1993). Only those which became more common or are used in particularly significant works are examined here. Not surprisingly, attempts to write Makassarese in roman script have been heavily influenced by colonial languages, namely Dutch and Indonesian. At present there is no generally accepted standard ̶ or more accurately, there are several competing romanisation methods.

There are two main Dutch based orthographies. One, developed by Matthes (1859), was in use until the Second World War. The second one was developed by Cense's (1979) and is only found in his dictionary.

Comparison of Makassarese orthography

	u	c	ɟ	ŋ	ɲ	-ʔ	(stressed vowels)
Matthes (1859)	oe	t͠j	d͠j	n͠g	n͠j	◌́, ◌̉	◌̂
Cense (1979)	u	tj	dj	ŋ	ñ	-ʼ	◌̀
Indonesian based (1975)	u	c	j	ng	ny	-k	(not written)
Locally preferred	u	c	j	ng	ny	-ʼ

Indonesian based system

In 1975 a seminar and workshops on standardising the writing of regional languages were held in the Language Institute in Ujung Pandang, in which a method of romanisation was proposed. This is found only in academic publications about the Makassarese language, such as the Makassarese‒Indonesian dictionary (Aburaerah Arief 1995) and various collections of folk texts published by the Balai Pustaka and the Departemen Pendidikan dan Kebudayaan (eg. Aburaerah Arief & Zainuddin Hakim 1993; Parewansa et al. 1992; Sahabuddin Nappu 1986; Sahabuddin Nappu & Sande 1991; Syamsul Rizal & Sahabuddin Nappu 1993; Zainuddin Hakim 1991). It is also used in school teaching materials. It uses the orthography of modern Bahasa Indonesia, which is for the most part reasonably adequate. The palatal stops are represented by c and j, and the only remaining digraphs are the palatal and velar nasals ̶ ny and ng respectively. These nasals when geminated are inconsistently represented as nny or nyny [ɲɲ] and nng or ngng [ŋŋ]. The palatal nasal which arises from assimilation is just shown as n (eg. moncong). Geminate consonants are represented as such, but sequences of glottal and voiced stops, for example [ʔb] are variously shown as kb and bb.

The main problem with this system is the representation of the glottal stop as k. Usually this is easily understood: syllable-final k = [ʔ], so that lekleng = [leʔleŋ], for example. However, when a glottal stop is syllable-initial, as when the a has been dropped from verbs such as [aʔlampa] ʻgoʼ, leaving [ʔlampa], this would have to be written as klampa (or ʻklampa showing the space left by the a). Although this can be understood, because [kl] is not a possible sequence in Makassarese, it is misleading. More seriously, when a glottal stop is intervocalic, such as when the verbal prefix aC- is attached to a vowelinitial root like [anaʔ] ʻchildʼ, resulting in [aʔanaʔ], the result is either the misleading akanak (as in Imran 1976), or some way must be found to show that this glottal stop does not become k. For example, Arief (1995) writes ak-agang [aʔagaŋ] ʻhave friendsʼ, showing the morpheme boundary, though this is never otherwise indicated.

But the most compelling argument against using k to represent the glottal stop is that they are not the same consonant. Many native speakers seem to agree with this three different people, when asked to read texts written using this system out aloud, began by pronouncing the ks, stopped in confusion, and started again pronouncing them as glottal stops. In my experience, most people prefer to use an apostrophe ⟨ʼ⟩, or leave out the glottal stops altogether (a shop in the centre of Makassar has a large sign reading BAJI PAMAI, which is pronounced [baɟiʔ paʔmaiʔ]).

Another drawback is that stress is never shown, and stress is the only way to distinguish between words with the applicative suffix -i and the enclitic =i. A quirk of the system is that it often uses an apostrophe to mark ʻmissingʼ (elided) letters in verb prefixes and encliticised demonstratives (much as it is used in English words such as donʼt). This can result in constructions such as iaminne ʻthis isʼ being represented as iamiʻnne or iami ʻnne. Since most Makassarese use the apostrophe to represent the glottal stop iamiʻnne looks like [iamiʔnːe].

Old Makassar and Lontara script

See also: Makasar script and Lontara script

Sample of a handwritten book, written in Makassarese using the Makasar script, of a diary of the Princes of the Sultanate of Gowa. The passimbang punctuation signs, typical of this script, are drawn and colored in red, as well as a few proper names and some inserts in Arabic.

Makassarese was historically written using Makasar script (also known as "Old Makassarese" or "Makassarese bird script" in English-language scholarly works). In Makassarese the script is known as ukiriʼ jangang-jangang or huruf jangang-jangang ('bird letters'). It was used for official purposes in the kingdoms of Makasar in the 17th century but ceased to be used by the 19th century, being replaced by Lontara script.

In spite of their quite distinctive appearance, both the Makasar and Lontara scripts are derived from the ancient Brahmi script of India. Like other descendants of that script, each consonant has an inherent vowel "a", which is not marked. Other vowels can be indicated by adding diacritics above, below, or on either side of each consonant.

Base consonant letters

	ka	ga	nga	pa	ba	ma	ta	da	na	ca	ja	nya	ya	ra	la	wa	sa	a	ha
Old Makassarese	𑻠	𑻡	𑻢	𑻣	𑻤	𑻥	𑻦	𑻧	𑻨	𑻩	𑻪	𑻫	𑻬	𑻭	𑻮	𑻯	𑻰	𑻱	-
Lontara script	ᨀ	ᨁ	ᨂ	ᨄ	ᨅ	ᨆ	ᨈ	ᨉ	ᨊ	ᨌ	ᨍ	ᨎ	ᨐ	ᨑ	ᨒ	ᨓ	ᨔ	ᨕ	ᨖ

The process by which the Makasar script became obsolete and was replaced by the Bugis was probably gradual, and may have been influenced by several factors, among them the decline of the power of Gowa and subsequent demoralisation of Makassarese chroniclers; the concomitant increase in Bugis power and influence; and the simpler (though less attractive to my eye) nature of the Bugis script. Cummings has speculated that for some time the Makassarese script was viewed as ʻmore arcane, rarified, and hence more spiritually powerful than the (Bugis) scriptʼ and also that Makassarese used it ʻin certain texts to distinguish themselves and their past from Bugisʼ (Cummings 2002:44). This could explain the scriptʼs continued existence for some time after the fall of Gowa, but the fact remains that there are now no Makassarese who can read it.

Ambiguity

Both scripts do not have a virama or other ways to write syllable codas in a consistent manner, even though codas occur regularly in Makassar. For example, in Makassar is baba ᨅᨅ which can correspond to six possible words: baba, babaʼ, baʼba, baʼbaʼ, bamba, and bambang.

Given that Lontara script is also traditionally written without word breaks, a typical text often has many ambiguous portions which can often only be disambiguated through context. This ambiguity is analogous to the use of Arabic letters without vowel markers; readers whose native language use Arabic characters intuitively understand which vowels are appropriate in a given sentence so that vowel markers are not needed in standard everyday texts.

Even so, sometimes even context is not sufficient. In order to read a text fluently, readers may need substantial prior knowledge of the language and contents of the text in question. As an illustration, Cummings and Jukes provide the following example to illustrate how the Lontara script can produce different meanings depending on how the reader cuts and fills in the ambiguous part:

Lontara script	Possible reading
	Latin	Meaning
ᨕᨅᨙᨈᨕᨗ	aʼbétai	he won (intransitive)
	ambetái	he beat... (transitive)
ᨊᨀᨑᨙᨕᨗᨄᨙᨄᨙᨅᨒᨉᨈᨚᨀ	nakanrei pepeʼ ballaʼ datoka	fire devouring a temple
	nakanrei pepe' Balanda tokkaʼ	fire devouring a bald Hollander
ᨄᨙᨄᨙ	pepe	mute
	pepeʼ	fire
	pempeng	stuck together
	peppeʼ	hit

Without knowing the actual event to which the text may be referring, it can be impossible for first time readers to determine the "correct" reading of the above examples. Even the most proficient readers may need to pause and re-interpret what they have read as new context is revealed in later portions of the same text. Due to this ambiguity, some writers such as Noorduyn labelled Lontara as a defective script.

Serang script

The influence of Arabic following Islamization in 1605, and also the presence of a Malay trading community using the Arabic-derived Jawi script, gave the option of using the Arabic script to represent Makassarese. This script was known as serang. It is a more complete way of representing the spoken language than the indigenous scripts, because it allows the representation of syllable-final consonants, but it never became widespread, with only a few manuscripts in such a script represented in collections. One of the major examples is the diary of the courts of Gowa and Talloʼ, which was transliterated and translated into Dutch by Ligtvoet (1880). However, the use of Arabic script (writing the Arabic language) is common in manuscripts to represent Islamic names, dates, and religious concepts.

Additonal letter compared to Arabic

Sound	Isolated form	Final form	Medial form	Initial form	Name
/c/	چ‎	ـچ‎	ـچـ‎	چـ‎	ca
/ŋ/	ڠ‎	ـڠ‎	ـڠـ‎	ڠـ‎	nga
/ɡ/	ࢴ‎	ـࢴ‎	ـࢴـ‎	ࢴـ‎	gapu
/ɲ/	ڽ‎	ـڽ‎	ـڽـ‎	ڽـ‎	nya

Sample text

Sample text
Old Makassar script	𑻨𑻳𑻰𑻴𑻭𑻶𑻥𑻳𑻷𑻥𑻮𑻱𑻵𑻠𑻲𑻷𑻱𑻢𑻴𑻠𑻳𑻭𑻳𑻠𑻳𑻱𑻭𑻵𑻨𑻷𑻰𑻳𑻬𑻡𑻱𑻷𑻱𑻵𑻣𑻶𑻯𑻨𑻷𑻠𑻥𑻦𑻵𑻬𑻨𑻷𑻰𑻳𑻬𑻡𑻱𑻷𑻮𑻭𑻳𑻠𑻥𑻦𑻵𑻬𑻨𑻷 𑻨𑻲𑻱𑻵𑻭𑻥𑻶𑻷𑻥𑻮𑻱𑻵𑻠𑻲𑻷𑻥𑻢𑻵𑻷𑻭𑻳𑻥𑻮𑻠𑻴𑻮𑻴𑻥𑻱𑻴𑻦𑻳𑻷
Lontara script	ᨊᨗᨔᨘᨑᨚᨆᨗ᨞ᨆᨒᨕᨙᨀᨀ᨞ᨕᨂᨘᨀᨗᨑᨗᨀᨗᨕᨑᨙᨊ᨞ᨔᨗᨐᨁᨕ᨞ᨕᨙᨄᨚᨓᨊ᨞ᨀᨆᨈᨙᨐᨊ᨞ᨔᨗᨐᨁᨕ᨞ᨒᨑᨗᨀᨆᨈᨙᨐᨊ᨞ ᨊᨊᨕᨙᨑᨆᨚ᨞ᨆᨒᨕᨙᨀᨀ᨞ᨆᨂᨙ᨞ᨑᨗᨆᨒᨀᨘᨒᨘᨆᨕᨘᨈᨗ᨞
Serang script	نِسُوْرُمِ مَلَائِكَكَةْ أَڠُّكِرِيْرِكِ أَرِيِنَّ سِيَࢴَاءً إِمْفُوَانَّ كَمَتِيَانَّ سِيَࢴَاءً لَنْرِ كَمَتِيَانَّ نَنَئِيْرَمَّ مَلَائِكَكَة مَاڠِيْ رِ مَلَكُ الْمَوْتِ‎
Latin Script	Nisuromi malaekaka anngukiriki arenna, siagáng empoanna kamateanna, siagáng lanri kamateanna, na naerammo malaekaka mange ri Malakulmauti.
Translation	The angels were ordered to record his name, the circumstances of his death, and the cause of his death, then the angels took him to Malakulmauti.

Some common words and phrases in the Makassarese language, transcribed in the Latin script, are as follows (⟨ʼ⟩ represents the glottal stop).

Writing system examples

Lontara	Romanized	Indonesian	Translation
ᨈᨕᨘ	tau	orang	people
ᨄᨎᨗᨀᨗ	paʼnyiki	kelelawar	bat
ᨕᨑᨙ	areng	nama	name
ᨕᨊ	anaʼ	anak	child
ᨔᨙᨑᨙ᨞ ᨑᨘᨓ᨞ ᨈᨒᨘ᨞ ᨕᨄ᨞ ᨒᨗᨆ᨞	seʼre, rua, tallu, appaʼ, lima	satu, dua, tiga, empat, lima	one, two, three, four, five
ᨅᨕᨗᨊᨙ	baine	perempuan, istri	female, woman, wife
ᨅᨘᨑᨊᨙ	buraʼne	lelaki, suami	male, man, husband
ᨈᨅᨙ	tabeʼ	permisi, maaf	excuse me, sorry
ᨈᨕᨙᨊ, ᨈᨙᨊ, ᨈᨑᨙ	taena, tena, tanreʼ	tiada	none, nothing
ᨒᨙᨅ	leʼbaʼ	telah	already
ᨔᨒᨆᨀᨗ ᨅᨈᨘ ᨆᨕᨙ	salamakkiʼ battu mae	selamat datang	welcome
ᨕᨄ ᨕᨈᨘ ᨆᨕᨙ ᨀᨅᨑ?	apa antu mae kabaraʼ?	apa kabar?	how are you?
ᨅᨍᨗᨅᨍᨗᨍᨗ	bajiʼ-bajiʼji	baik-baik saja	I am fine
ᨊᨕᨗ ᨕᨑᨙᨊᨘ?	nai arenta?	siapa namamu?	what's your name?
ᨒᨀᨙᨑᨙᨀᨗ ᨆᨕᨙ? ᨒᨀᨙᨀᨗ ᨆᨕᨙ? ᨒᨀᨙᨆᨕᨗᨀᨗ?	lakereki mae?, lakekimae?, lakemaeki?	kamu mau ke mana?	where are you going?
ᨀᨙᨑᨙᨀᨗ ᨆᨕᨙ ᨕᨆᨈ	kerekiʼ mae ammantang?	kamu tinggal di mana?	where do you live?
ᨔᨗᨐᨄᨆᨗ ᨕᨘᨆᨘᨑᨘᨈ?	siapami umuruʼta?	berapa usiamu?	how old are you?
ᨔᨒᨆᨀᨗ ᨑᨗ ᨆᨂᨙᨕᨈᨙ	salamakkiʼ ri mangeanta	selamat sampai tujuan	have a safe trip
ᨔᨒᨆᨀᨗ ᨑᨗ ᨒᨄᨈ	salamakkiʼ ri lampanta	selamat tinggal	goodbye
ᨅᨈᨘ ᨑᨗ ᨀᨈᨙ	battu ri katte	tergantung padamu	it depends on you

See also

• Makassan contact with Australia

References

• Tabain, Marija; Jukes, Anthony (2016). "Makasar". Illustrations of the IPA. Journal of the International Phonetic Association. 46 (1): 111–99. doi:10.1017/S002510031500033X, with supplementary sound recordings.

External links

• Lontara Alphabet in Unicode