1.

Introduction
The present book is a des'ripti/e grammar of Makalero@ one of the 1# national languages of East Timor re'ognised as su'h in the 'ountryFs 'onstitution* 4t is spoken in the subdistri't of 4liomar@ in the south(east of the 'ountry* The lo'ation of East Timor in the 'onteIt of insular 5outh(East Asia is sho-n in Map 1*

Malaysia Malaysia

Philippines Papua Papua +e- Cuinea +eEast Timor Cuinea Australia East Timor
73

4ndonesia 4ndonesia

Map 1: East Timor in 5outh(East Asia 2adapted from 5'happer !

With an area of 1"@:1 km_ and 1ust under a million inhabitants@ East Timor eIhibits Huite a striking language di/ersity* The linguisti' lands'ape of Timor as -ell as the neighbouring islands of Alor and Pantar is depi'ted in Map !* The Makalero( speaking area is darkened*

Map !: The languages of Timor@ Alor and Pantar 2adapted from =ull !

:3

.=APTE; 1

` 1*1 2p* !3 gi/es an outline of the history of the young 'ountry of East Timor* 4t in'ludes some details on the 4liomar subdistri't@ although relati/ely little is kno-n of its earlier history* ` 1*! 2p* "3 and ` 1*3 2p* 83 present ba'kground information on the geographi'al setting of the Makalero area and on its speakers and their 'ulture* ` 1*: 2p* 73 dis'usses the language and the language situation* The geneti' 'lassifi'ation of Makalero is outlined in ` 1*" 2p* 1!3@ follo-ed by a sket'h of the languageFs stru'ture in ` 1*# 2p* 1#3* ` 1*8 2p* !83 gi/es an o/er/ie- o/er pre/ious -ork* ` 1*9 2p* !93 details the 'orpus underlying the present analysis and the methodology used* ` 1*7 2p* 3!3@ finally@ dis'usses the orthography used for Makalero*

1.1 A short history of East Timor and Iliomar

The island of Timor is often 'ited as being on the possible migration route for man into Australia and Melanesia 2Clo/er 179#: :3* Earliest e/iden'e of settlement may be as old as : @ years* 4t is today inhabited by speakers of both Austronesian and Papuan languages@ -hi'h suggests multiple migrations* Whether the an'estors of todayFs Austronesian speakers or Papuan speakers arri/ed on the island first appears to ha/e been a matter of some debate: aa superfi'ial glan'e at a map might suggest that these Papuan en'la/es 2-ithin an area dominated by Austronesian languages@ $=3 are simply remnants of an earlier and larger distribution o/errun by Austronesian speakers@ but re'ent resear'h on the Papuan languages as a -hole may make su'h a /ie- rather simplisti'b 26ell-ood ! 8: 1!:3* Wurm 217933 suggests the Papuan languages of Timor deri/e from the third and main Papuan language migration@ -hi'h essentially amounts to the spread of his massi/e Trans(+eCuinea family and -as more or less 'ontemporary -ith the arri/al of Austronesian speakers in the area* Wurm posits this e/ent at around 3@ 6.@ -hile a''ording to 6ell-ood 2! 8: 1!33@ Austronesian settlers rea'hed Timor only by about !@ 6.* ;oss 2! "3@ on the other hand@ seems to 'ontend that Papuan speakers arri/ed in Timor earlier than Austronesian speakers* A possible migration route for Papuan speakers from +e- Cuinea@ starting some-here near the 6omberai peninsula and passing through the %ei and Tanimbar 4slands on-ards to Timor@ is depi'ted in =ull 2! :: #:3* .hinese and $apanese sour'es mention the island as early as the 1: th 'entury@ mainly as a sour'e of 5andal-ood 2>oI ! 8: #3* =o-e/er@ little is kno-n of the islandFs history before the arri/al of the first European eIplorers and traders by the early 1" s* The first European /isitors report that the island is fragmented into small kingdoms ruled by liurais 2kings3* The three ma1or kingdoms -ere@ a''ording to a 1#th 'entury sour'e@ Wehale@ -hi'h dominated the 'enter of the island@ 5onbaFi in the -est@ and Likusaen 2today: LiHuiNa3 in the east* >rom about 1"" @ European 'olonialism -as the ma1or for'e shaping the islandFs history* The earliest European a'ti/ity in the area -as by Portuguese traders and missionaries* 4nitially@ there -ere no permanent administration@ military presen'e or trade posts* =o-e/er@ these -ere established later@ to 'ounter the gro-ing influen'e of the ?ut'h in the area* 4n 1"9#@ large parts of the island of Timor -ere de'lared

4+T;O?&.T4O+

the Pro/in'e of Portuguese Timor* The follo-ing 'enturies -ere 'hara'terised by ?ut'h(Portuguese ri/alry on the island@ as -ell as se/eral rebellions of the Timorese against the foreign 'olonisers* The 'olonial ri/alry bet-een the t-o European po-ers persisted until 19"7@ -hen the t-o parties agreed on a di/ision of the island in the Treaty of Lisbon* =o-e/er@ territorial disputes 'ontinued@ and the boundaries as in eIisten'e today -ere not established until 171#* &ntil the late 199 s@ there -as little permanent Portuguese presen'e outside ?ili 2.hamberlain ! 9: 1!3* ;ather@ the 'oloniser ruled through the lo'al 'hiefs and kings@ lea/ing traditional stru'tures inta't* The Timorese@ ho-e/er@ -ere far from subdued@ and the Portuguese military had to deal -ith numerous rebellions* 4t -as only to-ards the end of the 17th 'entury that a 'ountry(-ide net-ork of military posts -as established* 4n response to the Timorese rebellion of 1711,171! in Manufahi@ the Portuguese tried to further redu'e the po-er of the traditional rulers and established an administrati/e system -here the land -as di/ided into a larger number of smaller administrati/e units@ or su'os 2/illages3* ?uring World War 44@ East Timor sa- Australian and $apanese military a'ti/ity* ?uring this time@ many Timorese fled the 'ountry@ and it -as in su'h refugee 'amps in Australia that .apell 217::@ 178!3 'ondu'ted his sur/ey of Timorese languages 2see ` 1*8@ p* !83* After the 'apitulation of $apan in 17:"@ Portugal resumed 'ontrol o/er Timor and retained it until the .arnation ;e/olution in 178:@ -hen the fas'ist di'tatorship in Portugal -as abolished* >ollo-ing this e/ent@ the then ruling party >retilin 2>rente ;e/olu'ionLria do Timor(Leste 4ndependente3 de'lared the independen'e of East Timor in 178"* 5hortly thereafter@ ho-e/er@ internal strife that -as probably supported by the 4ndonesian se'ret ser/i'e broke out* The 'ountry seemed on the /erge of 'i/il -ar* Presenting this as a threat to stability in the region@ 4ndonesia laun'hed an armed in/asion of Timor* >earing 'ommunist tenden'ies in >retilin@ the international 'ommunity -as largely silent* Thus East Timor be'ame the !8th pro/in'e of 4ndonesia* What follo-ed -as a period of 4ndonesian repression@ 'ountered by guerilla -arfare from >alintil 2>orNas Armadas de LiberaNOo +a'ional de Timor(Leste3@ the military -ing of >retilin* East Timor -as only brought ba'k to the international stage in 1771@ -hen 4ndonesian military killed at least !" Timorese as -ell as a number of foreign nationals during a pea'eful protest in the 'apital ?ili* This e/ent@ kno-n as the 5anta .ruA massa're@ -as re'orded by a 6ritish 'ameraman* &nder the ensuing international pressure@ 4ndonesia finally offered East Timor the 'hoi'e bet-een a -ide(rea'hing autonomy -ithin 4ndonesia or independen'e* This referendum took pla'e on August 3 @ 1777 and resulted in 89*"c of /otes for independen'e from 4ndonesia* ?ire'tly after this@ 4ndonesian military and pro(integration East Timorese militia 'arried out a 'ampaign of /iolen'e and terrorism in retaliation: a4n one month@ this massi/e military operation murdered some !@ people@ raped hundreds of -omen and girls@ displa'ed three( Huarters of the population@ and demolished 8" per'ent of the 'ountryFs infrastru'tureb 2.homsky ! 3: 8!3* A multinational for'e -as finally allo-ed into East Timor to stabilise the area* After an interim &+ administration@ East Timor be'ame formally independent on May ! @ ! !* The brief period of independen'e has been 'hara'terised by episodes of ma1or unrest in ! ! and ! #@ as -ell as a nearly fatal atta'k on the then(president $osQ ;amos(=orta in ! 9* To this day@ East

.=APTE; 1

Timor is among the poorest nations in Asia and is hea/ily dependent on foreign 'redits and +COs* There is little se'ure information on the history of the 4liomar subdistri't* .hamberlain 2! 9: 1:3 suggests the /illages of 4liomar 4 and 4liomar 44 in their present(day lo'ation -ere established in about 197:@ -hen t-o leaders@ +okameta 2-hose eHuestrian statue@ though damaged in the 1777 unrest@ stands in 4liomar today3 and ;apimeta mo/ed the Makalero people some kilometers up the 'oast in a northeasterly dire'tion* 4n 17 :@ a military post -as established in 4liomar@ and a Portuguese fort@ no- ruined@ -as built presumably at more or less the same time* Of the other siI /illages making up the 4liomar subdistri't 2see ` 1*! p* "3@ Ailebere -as established in 1718 and Tirilolo mu'h later@ in 1783 2.hamberlain ! 9: !1@ :"3* 1 ?uring the 4ndonesian in/asion@ -hi'h started in 178"@ the population of 4liomar didnFt eIperien'e the -ar until 1788 2.hamberlain ! 9: #:3* Late in that year@ the >retilin leadership ordered the e/a'uation of the people to Mt* Matebian* Mo/ing in a non(dire't -ay to a/oid 4ndonesian troups@ the population of 4liomar arri/ed there in 1789* 6y this time@ se/eral 1 @ other /illagers from the surrounding distri'ts -ere already in hiding there* La'k of pro/isions and atta'ks by the 4ndonesian armed for'es led to hea/y 'asualties among the 'i/ilian population* When the >retilin resistan'e 'rumbled in 1789@ the 'i/ilians -ere ordered ba'k to their /illages@ -hi'h -ere by then under 4ndonesian 'ontrol* &pon their return from Matebian@ all 4liomar /illagers -ere 'on'entrated in the area of 4liomar to-n* Their mo/ement -as restri'ted to -ithin " metres of the ne- /illage limits* These measures -ere intended to isolate the /illagers from >alintil elements in the 1ungle* Thus restri'ted@ the /illagers -ere unable to 'ulti/ate their traditional fieldsd the result -as a se/ere shortage of food@ and parti'ularly in 1791 and 179!@ malnourishment and the resulting mortality -as /ery high* >rom 1787 on@ 4liomar re'ei/ed emergen'y feeding from the 4nternational .ommittee of the ;ed .ross* 4n 1793@ the 4ndonesian armed for'es and >retilin negotiated a 'easefire* 4t -as only then that proper 'rops 'ould be planted and the supply situation impro/ed some-hat* 4n the period from 1793 to 1797@ there -as no organised resistan'e in the 4liomar area@ and the 4ndonesian administration -as 'onsolidated* +e/ertheless@ the remaining guerillas in the forests 'ontinued harassing the lo'al 4ndonesian for'es@ and there -ere numerous skirmishes bet-een >alintil fighters and 4ndonesian military as -ell as their lo'al military auIiliaries* The 177 s@ finally@ -ere a period of relati/e se'urity* After the referendum in 1777@ ho-e/er@ as the 4ndonesian for'es prepared to -ithdra- to Lospalos@ they burned publi' buildings@ in'luding the administration building@ the health post and the 1unior high s'hool in 4liomar@ and shot and killed numerous domesti' animals@ lea/ing 4liomar de/astated* 5in'e then@ re'onstru'tion has been going slo-ly@ -ith the subdistri't =ealth .enter and ele'tri'ity reestablished in 4liomar to-n in ! ! 2.hamberlain ! 9: 83* =o-e/er@ the po-er supply is intermittent due to fuel shortages and repairs to the generators* ` 1*! 2p* "3 gi/es more information on the infrastru'ture found today in the subdistri't*

+ote@ ho-e/er@ that the /illage 'hief of Ailebere gi/es the year 18#8 as the founding date for the /illage*

4+T;O?&.T4O+

"

1.2 Geography and infrastructure

Makalero is spoken in the 4liomar subdistri't of LautQm distri't@ -hi'h is depi'ted in Map 3*

Map 3: LautQm distri't! The subdistri't 'o/ers an area of approIimately 3 km ! 2.hamberlain ! 9@ AnneI .@ !3* 4t is bordered in the east and +ortheast by the Lospalos subdistri't@ in the north by Luro@ in the north-est by 6aguia@ in the east by &atu'arbau@ and in the 5outh by the Timor sea* 4t is home to a population of #@8!# 2a''ording to the ! : East Timor population 'ensus retrie/able at http:,,dne*mof*go/*tl,'ensus,indeI*htm3* 4liomar subdistri't is hea/ily -ooded and mountainous: ele/ations range from sea le/el to almost 7 m on Mount +aunili* The main 'onne'tion by road is from Lospalos* The road is :# kilometers long and in fairly bad 'ondition* Traffi' is freHuently interrupted due to mudslides* The road 'ontinues -est-ards to &atu'arbau* 4t is in dri/eable but /ery bad 'ondition@ although re'ently ne- bridges 'rossing the ri/ers to the east of 4liomar ha/e been 'ompleted* There is a regular tru'k and sometimes bus ser/i'e from 4liomar to Lospalos* >urthermore@ the &+ staff o''asionally use a 'entrally(lo'ated football field@ neIt to the primary s'hool in 4liomar to-n@ as a landing site* There is a 'atholi' 'hur'h -ith a resident priest in 4liomar@ as -ell as se/eral 'hapels in the surrounding /illages* A number of primary s'hools operate a'ross the subdistri't@ and there is one se'ondary s'hool in 4liomar to-n* Ele'tri'ity is pro/ided by t-o diesel(generators@ -hi'h are s-it'hed on daily from 8 pm to 1! pm* A lo'al health 'entre pro/ides basi' medi'al 'are* 4n ! 7@ there -as mobile re'eption on Mount +aunili@ the highest point of the subdistri't@ some " km out of 4liomar to-n on the road to Lospalos* 4liomar to-n is furthermore
!

< $*(P* >is'her* ;eprodu'ed -ith the permission of the author*

.=APTE; 1

proud o-ner of a library@ established by the Timor LorosaFe +ippon .ultural Pro1e't 2T+..3 in ! :* 4t pro/ides a /ariety of books and s'hool materials for all ages@ predominantly in Tetum@ English and $apanese* Politi'ally@ the 4liomar subdistri't is di/ided into siI /illages 2su'os3 -hi'h are@ in turn@ di/ided into a total of !" sub/illages* This organisation is sho-n in Table 1*13* /illage 4liomar 4 sub/illages 4liomar Ara(ara %aentau Ossuhira &atamatar Akara Lihina Madarira 6okila %aidabu Leilor Marafal =itali Lalumato ;umutau Akadirilo &ataomar Titirauen Maluhira Liufalin %aidalauarin Larimi Tirilolo Tatalalarin Eteuata Table 1*1: Millages and sub/illages in 4liomar Map : sho-s the approIimate lo'ation of these sub/illages* +ote that the spelling of the toponyms is not entirely standardised*

4liomar 44

Ailebere

>uat

%ainliu

Tirilolo

The spelling of the pla'e names is ad1usted to follo- that used in this thesis 2see ` 1*7@ p* 3!3*

4+T;O?&.T4O+

Map :: 4liomar subdistri't:

1.3 Culture and social life

Traditionally@ the Makalero so'iety is organised into 3# 'lans* " E/ery 'lan has its o-n origin story@ -hi'h is 'onsidered taboo to outsidersd a -rong telling of the story is belie/ed to result in early death or the like* Though offi'ially repla'ed by a modern demo'rati' administrati/e system during the Portuguese 'olonial rule and the 4ndonesian o''upation@ the traditional elite and 'lan stru'tures are still highly important* Lo'al kings are /ital to the resol/ing of so'ial issues and disputes@ as are traditional 'eremonies* >un'tions and positions asso'iated -ith the modern adminstrati/e stru'ture often go to members of the traditional ruling elites* .onsider
: "

< $*(P* >is'her* ;eprodu'ed -ith the permission of the author* :! a''ording to .hamberlain 2! 9: AnneI .@ "3* =e lists their names as 6usarulu@ 4liomar@ ?irimuni@ %omil@ Luruira@ Pukakesi@ TaFamatu@ Mumun@ &ardai@ Loorasa@ MaFaleuet@ Liusoru@ &ruhuFa@ Tafarira@ +onira@ +aunira 2note: these are likely /ariant spellings of the same name3@ &aritir@ 6uabere@ Abubul@ =ulalain@ Lafidebar@ 4nameli@ Luanira@ 6etunomar@ >ehira@ ?erekun@ Maluhira@ Ailebere@ ?eni@ %aparesin@ Muarai@ Lutupere@ %iltau@ &puira@ &pulira 2it is possible that one among the last t-o is a misspelling and the t-o refer to the same 'lan3@ &aitelu@ Manir@ Perut@ Paidur@ +utupupul@ Luruboitafa and Mauberu* The spelling has been ad1usted to follo- that used in this thesis 2see ` 1*7@ p* 3!3* The present 'orpus does not 'ontain a 'omplete set of 'lan names* =o-e/er@ it 'onfirms eight of the abo/e(mentioned names@ namely ?irimuni@ %omil@ Loorasa@ Liusoru@ Tafarira@ +aunira@ 4nameli and Maluhira@ and additionally gi/es the names ?olomara@ Loopupul@ Muenira@ Pusaulu@ Punakosi and ?arapuFu* 4t is at present un'lear ho- these relate to the names gi/en by .hamberlain* With the eI'eption of 4liomar 2a nominal 'ompound 'onsisting of ili XstoneF and omar Xstilt houseF3@ none of the 'lan names@ from either sour'e@ is transparent* +ote the o''urren'e of the /oi'ed bilabial stop in se/eral of the names listed by .hamberlaind this suggests 'lan names follo- the same phonologi'al subsystem as pla'e names 2see ` 3*1*3*1*!@ p* 1113*

.=APTE; 1

the follo-ing episode from .hamberlain 2! 9: "83@ relating the reason for a disagreement bet-een a >alintil group and the >retilin 4liomar Bone in 178#: aAn important element in the dispute -as =ornayFs 2the leader of the >alintil group@ $=3 antipathy to-ards the 4liomar >retilin 5e'retary@ Temas Pinto* =ornay reportedly belie/ed that the position should be held by a respe'ted liurai 2traditional leader3* A''ording to Lere Anan Timor 2the >retilin Mi'e 5e'retary of the 4liomar Bone at the time@ $=3: aThey didnFt -ant Temas Pinto to be the Bona 5e'retary@ be'ause they -anted a Xblue(bloodF 2i*e* a member of a po-erful 'lan@ $=3* =e 2>ran'is'o =ornay3 -anted me to be the Bona 5e'retary@ be'ause 4 -as a Xblue(bloodFb* The so'iety is patriar'hi'ally organised* Lo'als 'onsider marriage 'eremonies the 'entral pillar of the 'ultural frame-ork* 4n the 'ourse of this@ the 'lan of the groom pays a substantial bride pri'e to the 'lan of the bride* An important so'ial di/ision in this 'onteIt is into -ife(gi/er 2 omaraha3 and -ife(taker 2tuumata3 'lans* Another important part of so'io('ultural life are funerals and the 'eremonies asso'iated -ith them* These areas of 'ultural life still a-ait anthropologi'al in/estigation* The residents of the 4liomar subdistri't are mainly subsisten'e farmers* The main 'rops 'ulti/ated are ri'e@ maiAe and /egetables* >urthermore@ there are fairly eItensi/e 'o'onut plantations* Other 'rops 'ulti/ated are 'andlenut 2aleurites molu''ana3@ are'a nut and lontar palms 2borassus sundai'us3* Li/esto'k in'ludes -ater buffalo@ 'attle@ pigs@ goats@ 'hi'kens and horses* ?espite the subdistri'tFs long 'oast line@ little fishing is undertaken due to rough seas* The large ma1ority of residents are ;oman .atholi' 27#*"c of the population nation( -ide@ a''ording to http:,,---*state*go/,r,pa,ei,bgn,3"989*htm3* +e/ertheless@ traditional animist belie/es remain strong and 'oeIist -ith the .atholi' pra'ti'es* 5ome degree of formal s'hooling is /ery 'ommon* Many young people 'ontinue edu'ation in other 'ities of East Timor@ either temporarily or permanently* =o-e/er@ the eIa't proportions are unkno-n* There are siAeable Makalero 'ommunities in the distri't 'apital Lospalos@ -here they are 'entred in the sub/illages of +atura and 5a-ari'a@ and in ?ili@ parti'ularly in the 6e'ora area*

4+T;O?&.T4O+

1.4 Language and language situation

Makalero@ sometimes also gi/en as Maklere #@ is the main language of the subdistri't of 4liomar* 4t borders the language area of +aueti in the -est@ Makasae in the north and north-est@ and >ataluku in the east* 6oth the >ataluku and the Makasae languages are 'learly related to Makalerod +aueti@ on the other hand@ is not* The linguisti' boundary -ith +aueti appears 'lear('ut@ marked by the ma1or 4rabere ri/er -hi'h also forms the -estern boundary of the 4liomar subdistri't* The boundaries to the other languages are less easy to define and appear to be more gradual* The inhabitants of the Luro subdistri't@ -hi'h borders 4liomar on the north@ appear to think of their idiom@ referred to as 5aFani@ as distin't from either Makasae and Makalero* The spee'h of this subdistri't has seen no linguisti' in/estigation yetd a first impression by the present author 'onfirms that the mutual intellegibility -ith Makalero is high* 5aFani may thus perhaps be seen as a transitional /ariant bet-een Makasae and Makalero* The present in/estigation -as 'arried out in a relati/ely restri'ted radius 'entred on 4liomar to-n* As su'h@ no notable diale'tal differen'es -ere re'orded* =o-e/er@ in a number of instan'es 'onsultants 'hara'terised the use of spe'ifi' leIemes as typi'al for the spee'h of one or another /illage* This suggests the presen'e of regional linguisti' /ariation* 4t is the authorFs impression@ ho-e/er@ that these are subtle and mainly leIi'al* At the time of -riting@ Makalero is relati/ely /ital* 4t is spoken by an estimated #@" speakers8 and is a ma1or identifi'ation fa'tor for the people of 4liomar* The language is used in both e/eryday life as -ell as in ritual 'onteIts* 4t is being learnt as a first language by all 'hildren of Makalero parents in the distri't* O/erall@ the population of the 4liomar subdistri't is fairly homogeneous@ -ith /ery fe- residents from other parts of the island* ?espite the relati/e isolation of the Makalero language area@ there is a high le/el of multilingualism* A large part of the population has some 'ommand of Tetum and ,
The origin of the name Makalero is un'lear@ and the speakers appear not to be able to eIplain it satisfa'torily* The first element@ maka@ is also found in the name of its nearest linguisti' relati/e@ Makasae* One speaker eIplained to me that maka" translates as XfirmF@ and lero is supposedly taken from the name of a founding an'estor named 4lilero Laualero@ -ho -as a headstrong and independent leader and kept separate from the others* The 'ombination@ a''ording to the said speaker@ signifies a people that is headstrong and independent* .hamberlain 2! 9: !3 Huotes another speaker -ho renders Makalero as Xthe people -ho ne/er tireF* Almeida 2177:: "9:3 notes that Makalero@ or aMa'alQreb@ is a aetnonimo formado por M#ka@ 'ansado + L$re% a denominaNOo do po/o daHui* Os MakalQre des'onhe'em a sua terra de origem e 'om Huem sOo aparentados antropologi'amente@ afirmando os /elhos Hue alguns deles se 'hamam a si proprios &nu L$fu L$re@ ou se1a: &nu gente + L$fu + L$re@ abelhab 2an ethnonym formed from maka XtiredF and lere@ the designation of the lo'al people* The Makalero do not kno- their origin and -ith -hom they are related anthropologi'allyd the elders 'laim that some of them refer to themself as anu lefu lere@ -hi'h is made up from anu 'peopleF + lefu + lere XbeeFd translation mine3* T-o of these a''ounts seem to agree on the translation of maka as XtiredF* 4ndeed@ it is attested in this meaning in the 'orpus* 4t is likely of Austronesian origin 2i*e* borro-ed from Tetum maka Xto limp@ be tiredF3* AlmeidaFs suggestion of anu lefu lere Xbee peopleF is unsubstantiated by the present 'orpus@ -hi'h re'ords uani for XbeeF* 8 A''ording to the East Timor population 'ensus of ! : 2http:,,dne*mof*go/*tl,'ensus,indeI*htm3 "@791 indi/iduals aged # and abo/e ha/e Makalero as their mother tongue* "@""7 of these li/e in the LautQm distri't and !7! in ?ilid the rest is s'attered a'ross the other distri'ts of East Timor* The estimate of a total number of #@" speakers adds a rather generous " more speakers lo'ated abroad*
#

.=APTE; 1

or 4ndonesian* Pra'ti'al kno-ledge of Portuguese is relati/ely restri'ted at the time of -riting* =o-e/er@ this language is the offi'ial language of instru'tion* ?espite this fa't@ Tetum or 4ndonesian -ere more 'ommonly used during my first /isit in ! 8* 5in'e then@ tea'hers ha/e re'ei/ed intensi/e training in Portuguese and -ere mostly using that language in the 'lassroom by the time of my se'ond /isit in ! 7* 4t is un'lear to -hat degree the pupils are able to follo- 'lasses in this language* Tetum@ -ith a large proportion of Portuguese(based /o'abulary@ is used in matters pertaining to politi's and national issues* 4t is also the language used in 'hur'h* Although 4ndonesian has lost ground sin'e 1777@ it is still present as the language of entertainment in that most tele/ision programmes follo-ed by the lo'al population are 4ndonesian* >urthermore@ it is /ery 'ommon for young people to lea/e the area temporarily or permanently in order to pursue edu'ation or 1ob opportunities* A language -ith relati/ely fe- speakers in a hea/ily multilingual nation -here mobility is on the in'rease@ Makalero must thus be 'onsidered endangered in the longer term* Although it is re'ognised as a national language in the 'onstitution of East Timor@ many pre/ious sour'es on the languages of the nation do not re'ognise Makalero as a language in its o-n right@ but as a diale't of the larger Makasae 2see e*g* Le-is ! 73* On the other hand@ =ull 2! :: !#,313 dis'usses it a separate language that is atransitional bet-een Makasai and >ataluku* 2***3 Of all the non(Austronesian languages of Timor@ Makasai and Makalero are 'learly the most 'losely related@ although mutual intelligibility is minimal* LeIi'al 2as -ell as morphologi'al3 disagreements bet-een the t-o /erna'ulars are ne/ertheless freHuent enough to 1ustify their 'lassifi'ation as separate languages*b The present study 'onfirms a /ery high degree of leIi'al 'orresponden'e bet-een Makasae and Makalero* The 'omparati/e -ord list in Table 1*! gi/es an impression of the degree of similarity bet-een the t-o languages* .ontrary to =ullFs assessment of minimal mutual intellegibility@ it appears dire't 'ommuni'ation bet-een the t-o groups is possible to a 'onsiderable degree* 4n fa't@ Makalero speakers 'hara'terise their language as a diale't of Makasae* 4t is noti'eable@ ho-e/er@ that the similarity bet-een the t-o idioms is 'entred on 'ontent /o'abulary@ -here regular sound 'orresponden'es 'an be established* >or instan'e@ /oi'ed plosi/es in Makasae 'orrespond to /oi'eless ones in Makalero@ -hi'h seems to ha/e lost this distin'tion* This 'onstitutes the main differen'e bet-een the phonologi'al systems of the t-o languages* >urthermore@ Makasae seems to ha/e lost a final ,r, in se/eral 'ases -here it is still present in Makalero* Also noti'eable are differen'es in /o-el Hualities in diphthongs or -here a glottal phoneme is in/ol/ed* 4tems -ith more grammati'al fun'tion@ on the other hand@ are in many 'ases -ildly different* EIamples are pro/ided by the last t-o items in Table 1*!@ namely the modal /erb Zbe ableF and the 'lause(internal negator*

4+T;O?&.T4O+

11

English day , sun night ri/er earth@ ground house meat bone name 'hild older sibling younger sibling grandparent, grand'hild one t-o three siI good -et long sharp hea/y say hear sleep sear'h beat be able +EC

Makalero uatu kamu ueir mu"a omar ( lopu seur afa nei mata nana noko dada u ( un loloi ( lolei lolitu douh rau hifal asan mera" ti"ir lolo uali" ti"a haka pase me"e nomo

Makasae )atu *amu )air ma"a oma seu sapa nai mata kaka noko dada u lola"e lolitu daho rau ipal asan mera" ti"ir lolo )ali" ta"e ha*a +ase +e"u to% noto% tonai% notonai% noko

Table 1*!: LeIi'al 'omparison bet-een Makalero and Makasae As al-ays@ the distin'tion bet-een languages and diale'ts is not an easy one to dra-* While there is a high degree of leIi'al similarity bet-een Makasae and Makalero@ there is a range of grammati'al differen'es* An instan'e is the parti'le , 'liti' ini@ -hi'h is found in both languages* 4n Makasae@ its main fun'tion is that of a sub1e't marker 2=uber ! 9a: 8"f3* 4n Makalero@ ho-e/er@ it is the most freHuently used 'liti' 'lause linker@ -hi'h signals a 'lose semanti' relationship bet-een the 'lauses it links 2` 3*"*!*"@ p* !:8 and ` 9*1*1@ p* :"83* 4t is@ ho-e/er@ also used as a marker of 'ontrasti/eness -ith +Ps and appears to be de/eloping into a sub1e't marker* >urthermore@ -hile Makasae is largely isolating@ Makalero retains some infle'tional morphology* Also@ it appears the distin'tion bet-een ob1e'ts and 'omplements -ithin the /erb phrase@ arguably the most 'ompleI part of Makalero grammar 2` "*!@ p* 3173@ is not paralleled in Makasae 2although further resear'h is

1!

.=APTE; 1

needed on this issue3* Crammati'ally@ Makalero is more 'onser/ati/e than Makasae@ retaining many features that appear to ha/e been lost in Makasae* 4n many parts of the -orld@ su'h a situation -arrants the use of the term language@ rather than diale't*

1.

Classification

Makalero is a non(Austronesian , Papuan language* 9 4t is /ery 'losely related to the neighbouring language Makasae 2see Table 1*! in ` 1*:@ p* 73@ as -ell as@ some-hat more distantly@ to >ataluku@ spoken to the east of it@ and Oirata on %isar* The relations bet-een these languages is detailed in >igure 1*1* The terms XEastern TimorF and XTimor tipF follo- ?onohue 2! 83* Eastern Timor Makasae(Makalero Makasae Makalero Timor tip >ataluku Oirata

>igure 1*1: The Eastern Timor languages The eIa't position of 6unak@ spoken in the 'entre of the island 25'happer ! 73@ in a family tree representing all four of the Papuan languages of Timor is not Huite 'lear* With respe't to higher(le/el 'lassifi'ation@ the non(Austronesian languages of Timor ha/e had a 'heHuered history* The non(Austronesian nature of 6unak and Makasae -as first re'ognised by .apell 217:3a@ 17:3b@ 17::3@ -ho regarded them as Papuan on a''ount of their stru'ture* .o-an 217#": 3#13 supported this /ie- and suggested a relation of the non(Austronesian languages of Timor to those of the 6irdFs =ead in 4ndonesian +e- Cuinea* .apell 2178"3 finally argued for the grouping of all non( Austronesian languages of Timor and the neighbouring islands of Alor and Pantar -ith the languages of the 6irdFs =ead and the 6omberai peninsulad these in turn are argued to be a member of the eItensi/e Trans(+e- Cuinea phylum in Moorhoe/e 2178"3* This 'lassifi'ation has sin'e largely been upheld 2see e*g* Wurm 179!@ 1793 and ;oss ! "3*7 The Trans(+e- Cuinea family@ -hi'h 'o/ers a large part of the +e- Cuinea mainland as -ell as some nearby islands@ may 'ontain more than " languages* 5e/eral resear'hers@ among them Pa-ley 2! ": 833@ emphasise that the Timor(Alor( Pantar languages are highly aberrant among Trans(+e- Cuinea languages* ?espite that familyFs enormous siAe and the fa't that its members eIhibit a -ide range of stru'tural /ariation@ it follo-s from su'h remarks that it is ne/ertheless possible to identify a typi'al Papuan@ or Trans(+e- Cuinea@ language type* A summary of
9 7

The t-o terms are used inter'hangeably in the present thesis* +ote@ ho-e/er@ that it is 'ontested in ?onohue and 5'happer 2forth'oming3d the final -ord on the 'lassifi'ation on the Timor(Alor(Pantar group has thus not yet been spoken*

4+T;O?&.T4O+

13

general phonologi'al@ morphologi'al and synta'ti' features that are 'hara'teristi' for this language type is gi/en in >oley 21779: " #f*3* Tables 1*3 through 1*" summarise these features* The right 'olumn in ea'h table gi/es their Makalero /alues@ sho-ing ho- it relates to this general Papuan language type* feature /o-el system stops: pla'es of arti'ulation stops: /oi'ing distin'tion fri'ati/es liHuids syllables stress tone Papuan di/ersed 2,i e a o u, -idespread3 Makalero ,i e a o u,

3(:: labial@ dental( ": labial@ al/eo(dental@ al/eolar@ /elar@ 2palatal3 post(al/eolar@ /elar@ glottal di/erse patterns 2absen'e of /oi'ing 'ommon3 restri'ted 2 (1 members3 absent

,f s, 2UhV3

UrV and UlV not phonemi' ,r, and ,l, phonemi' 'ompleI 'onsonant 'lusters 'ommon .M fa/ouredd some 'onsonant 'lusters

di/erse rules 2phonemi' predi'table stress 'ommon3 'ommon absent

Table 1*3: Makalero and Papuan languages 'ompared: phonology T-o of the patterns found in Makalero phonology@ namely the fi/e(/o-el system and the absen'e of /oi'ing in stops@ are also -idespread in Papuan languages* Other than these@ Makalero is Huite distin't from the typi'al Papuan language -ith respe't to its phonologi'al 'hara'teristi's*

1:

.=APTE; 1

feature

Papuan

Makalero 'lose to isolating /ery -eakly de/eloped little 2'omplement(/erb 'ompleIes3

morphologi'al stru'ture -eakly or highly agglutinati/e infle'tional morphology strongly de/eloped deri/ational morphology nouns +Ps /erbs /ariedd freHuent use of 5M.s

normally uninfle'ted for uninfle'ted 2may be number and gender plural(marked3 typi'ally infle'ted for 'ase /aried 2the presen'e of at least an agreement affiI for the sub1e't is 'ommon3 /ariedd often suffiIald 5M.s 'ommon sharp di/ision into nouns and /erbs 'ommon uninfle'ted mostly uninfle'ted

TAM leIi'al 'ategories

pre/erbal parti'led separate predi'ations 'ategorial indetermina'yd some deri/ational pro'esses

Table 1*:: Makalero and Papuan languages 'ompared: morphology 4n the domain of morphology@ Makalero resembles the Papuan language type -ith respe't to nominal infle'tion@ -hi'h is largely non(eIistent 2eI'ept for the fa't that there is in Makalero optional plural marking3* Another possible analogy is TAM marking@ for -hi'h serial /erb 'onstru'tions are -idespread in Papuan languagesd although the present thesis does not use the term 5M.@ the ad/erbial /erbs used to eIpress the ma1ority of the TAM 'ategories in the form of distin't predi'ations may Hualify as su'h* 4n all other respe'ts@ Makalero de/iates from the Papuan language type*

4+T;O?&.T4O+

1"

feature phrase

Papuan right(headed f 5OM 2O5M3 f postpositions f + ?ET f MO? + 2+ MO? possible3 f 'lause 'haining@ no 'on1un'tions

Makalero right(headed 5OM no adpositions + ?ET + MO? 'lause linking -ith 'liti' 'on1un'tionsd 'omplement(/erb 'ompleIes

Table 1*": Makalero and Papuan languages 'ompared: syntaI 4n the domain of syntaI@ Makalero sho-s the largest degree of agreement -ith the Papuan language type@ 'onforming to it -ith respe't to general phrase headedness@ -ord order@ as -ell as the pla'e of the determiner and the modifier -ith a noun* Makalero agrees -ith >oleyFs !! 'hara'teristi' Papuan features 1 as reprodu'ed in Tables 1*3* through 1*"@ in only 9 instan'es@ most of -hi'h are found in syntaI* Apart from the /o'abulary@ syntaI thus pro/ides the strongest argument for its 'lassifi'ation as Papuan*11 >oley 21779: " #f*3 'ontrasts this Papuan language type -ith an XAustronesian language typeF* The 'hara'teristi's he lists for this language type may perhaps adeHuately represent the Austronesian languages in the +e- Cuinea area but 'annot be said to be 'hara'teristi' of the Austronesian family as a -hole and are thus not reprodu'ed here* +otably@ ho-e/er@ Makalero agrees -ith >oleyFs Austronesian language type to more or less the same degree 2i*e* number of features3 as it does -ith the Papuan language type* With respe't to other features@ it does not group 'learly -ith either of the t-o 'ontrasting language types* O/er /ery similar results@ 5'happer 2! 7: 313 'on'ludes that Papuan 6unak displays aa parti'ular Xinsular eastern 4ndonesianF linguisti' type -hi'h 'ross('uts the distin'tion bet-een Papuan and Austronesian*b This 'on'lusion 'learly also holds for Makalero* 4nterestingly@ %lamer@ ;eesink and 5taden 2! 93 propose an East +usantara linguisti' area@ in -hi'h Austronesian and Papuan languages 'oeIist and share a number of features -hi'h are partly of Papuan@ partly of Austronesian origin* 4t must be noted@ ho-e/er@ that this linguisti' area is based on only fi/e relati/ely broad features@ of -hi'h only t-o are 'learly manifested in Makalero 2i*e* possessor( possessum order in adnominal possession and the presen'e of an in'lusi/e ,
1

The total of !! 'ounts the synta'ti' features in Table 1*" ea'h separately@ despite the fa't that most of them are 'on'omitants of 5OM* 11 +ote@ ho-e/er@ that -ord order is not a sound basis for language 'lassifi'ation@ sin'e languages may 'hange their basi' -ord order 2see =eine ! 9: 3:f* for some 'onta't(indu'ed eIamples of this pro'ess3*

1#

.=APTE; 1

eI'lusi/e distin'tion in the personal pronoun paradigm3* +ote also that ?onohue 2! :31! denies the /alidity of an area -ith linguisti'ally definable borders in .entral and Eastern 4ndonesia* 4nstead@ he 'laims the region is part of a typologi'al 'ontinuum bet-een +orthern@ Western and Eastern Austronesian languages*

1.! A grammatical s"etch of #a"alero

The follo-ing se'tions gi/e an o/er/ie- of Makalero grammar* They largely follothe o/erall organisation of the thesis 2` 1*1 @ p* 3"3* ` 1*#*1 2p* 1#3 remarks briefly on the make(up of the /o'abulary@ -hile ` 1*#*! 2p* 173 dis'usses the phonology* ` 1*#*3 2p* !13 sket'hes the leIi'al 'ategories found in Makalero* ` 1*#*: 2p* !!3 and ` 1*#*" 2p* !33@ respe'ti/ely@ treat the noun phrase and the /erb phrase* ` 1*#*# 2p* !:3 gi/es a short eIamination of the 'lause* ` 1*#*8 2p* !:3 and ` 1*#*9 2p* !"3 dis'uss units larger than the 'lause@ namely the senten'e@ the utteran'e and the teIt* >inally@ ` 1*#*7 2p* !#3 gi/es some remarks on dis'ourse stru'ture*

1.6.1 The vocabulary

The bulk of the Makalero /o'abulary is non(Austronesiand indeed@ together -ith the languages 5OM profile 2` 1*"@ p* 1!3@ this is the main reason for its 'lassifi'ation as a Papuan language* =o-e/er@ it is also /ery ri'h in loan-ords@ most of -hi'h 'ome from Austronesian sour'es* 5e/eral sour'es note that 2Proto(3Austronesian loans permeate e/en the basi' /o'abulary of many Papuan languages 2see e*g* Wurm 179!: 9"@ >oley 179#: !113* >oley 2ibid*3 lists 13 leIi'al items 'ommonly found in Papuan languages -hi'h 'an be tra'ed ba'k to Austronesian sour'es* Table 1*# sho-s those items from the list that appear to deri/e from the same Austronesian roots13 in Makalero*

1! 13

;eferring to %lamer 2! !3@ -hi'h gi/es an earlier /ersion of the hypothesis* Most of the re'onstru'ted forms >oley 2179#: !113 lists appear to be Proto(Malayo(Polynesian@ rather than Proto(Austronesian@ as he indi'ates* 4n some 'ases@ his spelling is adapted in the Table a''ording to the forms in the Austronesian 6asi' Mo'abulary ?atabase retrie/able at http:,,language*psy*au'kland*a'*nA,austronesian,*

4+T;O?&.T4O+

18

Proto(Malayo(Polynesian re'onstru'tion gkulit XskinF ,t-ina XmotherF ,)ahi- X-aterF

Makalero ulit Xskin 2human or animal3F@ uli Xbark@ leatherF ina Xmother 2as a term of address only3F ueir Xri/erF

Table 1*#: .ommon Austronesian loans in Papuan languages and their refleIes in Makalero1: Tables 1*8 and 1*9 eIemplify other Austronesian loan-ords in Makalero -hi'h are presumably of 'onsiderable antiHuity* The leIemes in Table 1*8 must be 'onsidered part of the 'ore /o'abulary* Also@ related forms are found not only in Makalero@ but also in 2some of3 the surrounding non(Austronesian languages* This suggests they -ere borro-ed in the 'ommon an'estor language* The leIemes in Table 1*9 parti'ipate in su'h morphologi'al pro'esses as the distin'tion bet-een free and bound or redu'ed /erb forms in Makalero 2` 3*!*!*!*!@ p* 1313* These pro'esses are not produ'ti/e in the present(day language and are not used in ne- /erbs* The fa't that the morphemes in Huestion parti'ipate in this morphologi'al alternation thus points to their full integration into Makalero and suggests a long(term presen'e in the language@ too* Makalero 'ognates in other Papuan languages of Timor 6nH kuda@ >at ku.a@ Mks kuda XhorseF 6nH pip@ >at pipi@ Mks pipi XgoatF >at le XhouseF Mks oma XhouseF Mks seluk XotherF 'ognates in Austronesian languages 64 kuda@ Ttm XhorseF Ttm +i+i XgoatF kuda

kuda XhorseF pipi XgoatF lopu XhouseF omar1" Xstilt houseF selu XotherF

64 lepau Xfood stallF 2PA+ glepa)\3 64 rumah@ Ttm uma XhouseF Ttm seluk XotherF

Table 1*8: Austronesian 'ore /o'abulary in Makalero

1:

Another item listed as a 'ommon Austronesian root in Papuan languages in >oley 2179#: !113 is gt"ut"u XbreastF@ -hi'h appears to be refle'ted in Makalero dudu XbreastF* =o-e/er@ a''ording to the Austronesian 6asi' Mo'abulary ?atabase@ this leIeme re'onstru'ts as ,susu@ rather than gt"ut"u@ for both Proto(Austronesian and Proto(Malayo(Polynesian* 6orro-ing of this Austronesian root into the an'estor of Makalero -ould thus in/ol/e a more un'ommon 'hange from g,s, to ,d,* +ote@ ho-e/er@ that the ?atabase lists se/eral Austronesian languages 2mostly from the Philippines3 as ha/ing a form dudu or dudu/ for XbreastF* 1" The origin of the final ,r, is not Huite 'lear* 4t may possibly be identi'al to the nominal formati/e -r dis'ussed in ` 3*1*!*1*1 2p* 1 !3*

19

.=APTE; 1

Makalero hein h (sein X-aitF resin h resi- XremainF taru h -daru Xput@ pla'eF teri h (deri X'utF

'ognates in Austronesian languages Ttm hein X-aitF Ttm resin XeI'eed@ be left o/erF 64 taruh Xput@ pla'eF Ttm teri X'ut -ith s'issors or shearsF

Table 1*9: Austronesian loan -ords in Makalero that parti'ipate in morphologi'al pro'esses A semanti'ally delineable group of presumably old Austronesian loans are numerals* Table 1*7 sho-s a sele'tion of Makalero numerals of Austronesian origin* >or more details@ see ` 3*!*3*9 2p* 1813* Makalero fat lima fitu si)a rihun : " 8 7 1@ Proto(Austronesian g0epat glima gpitu gsi)a g1i+u

Table 1*7: Austronesian numerals in Makalero Makalero also abounds in ne-er Austronesian loans@ -hi'h 'an be tra'ed to spe'ifi' languages* Tables 1*1 and 1*11 gi/e a sele'tion of loan-ords from t-o of the regionFs lingua fran'as@ Tetum and 4ndonesian@ respe'ti/ely* The -ords gi/en in the tables o''ur /ery freHuently and are as su'h distinguished from non'e(borro-ings from these languages* +ote that the Tetum loans 'ome from a -ide range of semanti' domains and leIi'al 'ategories* Those from 4ndonesian are some-hat more restri'ted and often pertain to formal s'hooling@ a domain -hi'h -as firmly asso'iated -ith 4ndonesian for more than ! years during the 4ndonesian o''upation* Makalero +eik% peik XstupidF funu X-arF kole XtiredF kulat XmushroomF liurai XkingF matenek X'le/erF mate+ian XsoulF susuuk XmosHuitoF tam+a Xbe'auseF ukun XruleF Tetum sour'e +eik XstupidF funu X-arF kole XtiredF kulat XmushroomF liurai XkingF matenek X'le/er@ intelligentF mate+ian XsoulF susuk XmosHuitoF tan +a Xbe'auseF ukun XruleF Table 1*1 : Loan-ords from Tetum

4+T;O?&.T4O+

17

Makalero kasihan XpoorF karna Xbe'auseF lansun Xstraight a-ay@ dire'tlyF tamat XgraduateF

4ndonesian sour'e kasihan Xmer'y@ pity@ poorF karena Xbe'auseF lan*sun* Xdire't@ straight@ immediatelyF tamat Xfinish@ 'omplete@ endF

Table 1*11: Loan-ords from 4ndonesian As in all Timorese languages@ there is in Makalero a multitude of leIi'al items from Portuguese origin* 4t is un'lear -hether these -ere borro-ed dire'tly from Portuguese@ or through the medium of Tetum*1# Portuguese loans are generally 'ultural and in'lude many administrati/e terms* Makalero deue Xdebt@ borro-F findesemana X-eekendF *o2ernu Xgo/ernmentF istuda XstudyF kalsa% kaalsa XtrousersF komesa XstartF leh XreadF lo3a XshopF nosa siniora Xour ladyF orde Xannoun'ement@ notifi'ationF paun XbreadF po2u XpeopleF re3a XprayF
18

Portuguese sour'e de2er Xo-eF fim-de-semana X-eekendF *o2erno Xgo/ernmentF estudar XstudyF .al4as XtrousersF .ome4ar Xbegin@ startF ler XreadF lo3a XshopF nossa sinhora Xour ladyF ordem Xorder@ dire'ti/eF p5o XbreadF po2o Xpeople@ popula'eF re6ar XprayF

Table 1*1!: Loan-ords from Portuguese

1.6.2

honology

Makalero phonology is fairly straightfor-ard* There are 11 nati/e 'onsonant phonemes@ -hi'h are sho-n in Table 1*13* There is no phonemi' 'ontrast in /oi'ing* Although the presen'e of the phoneme ,d, seems to be an eI'eption to this rule@ Table 1*13 sho-s that it is distinguished from ,t, not only in /oi'ing@ but also in the pla'e of arti'ulation* The glottal phoneme@ represented as ,T,@ has a stop
1#

=ull 2! :: !83 'laims >ataluku and Makalero generally borro-ed dire'tly from Portuguese@ -hile Portuguese(based leIemes in Makasae and 6unak -ere transmitted through Tetum and other lo'al languages* =o-e/er@ he does not pro/ide e/iden'e to substantiate this 'laim* 18 +ote that Portuguese /erbs are not generally borro-ed in the infiniti/e form@ but rather in the third person singular present tense* An eI'eption is fiar Xtrust@ honourF@ -hi'h ultimately deri/es from Portuguese fiar Xtrust@ entrustF* 4t is found in an identi'al form in Tetum as Xbelie/eF*

.=APTE; 1

allophone UTV and a fri'ati/e allophone UhV* 6oth are freHuently pronoun'ed so faintly that they are pra'ti'ally inaudible* labial al/eo( dental t n s post( al/eolar d l r /elar glottal

Plosi/e ?i/erted airstream .ontinuant

p m f

Table 1*13: .onsonant phonemes 4n the domain of /o-els@ Makalero eIhibits a 'lassi'al fi/e(/o-el system as depi'ted in Table 1*1:* i 2i:3 e 2e:3 a 2a:3 Table 1*1:: Mo-el phonemes The mid(/o-els ha/e open(mid and 'lose(mid allophones* Mo-els are lengthened in a /ariety of 'onteIts@ -hi'h are largely predi'table: >irstly@ a seHuen'e of t-o identi'al /o-els interrupted by the glottal phoneme may alternati/ely be pronoun'ed as a long /o-el* 5e'ondly@ simple /o-el nu'lei of monosyllabi' morphemes are lengthened for the morphemes to 'onform to the bimorai'ity reHuirement@ and thirdly@ /o-el lengthening is a 'onseHuen'e of the resyllabifi'ation pro'esses asso'iated -ith e'ho /o-els* There are also some morphemes -ith unpredi'table long /o-elsd ` !*!*3*: 2p* #"3 argues that these 'an most 'ommonly be assumed to be 'ompounds@ in -hi'h one of the 'onstituent parts is monosyllabi'* The glides U-V and U1V are analysed as allophones of the high /o-els -hen they are used either as a syllable onset@ as part of a 'ompleI nu'leus@ or in 'oda position* The most 'ommon syllable stru'ture is .M@ although .M. may be found* .onsonant 'lusters are disfa/oured and epentheti' /o-els are freHuently inserted to a/oid them* At the end of a phonologi'al phrase@ an e'ho /o-el is 'ommonly added to a 'onsonant(final morpheme* ` !*"*:*1 2p* 933 sho-s ho- in that 'ase@ the nu'leus of the erst-hile final syllable is lengthened* The minimal leIi'al morpheme is bimorai'd the 'oda does not 'ontribute to syllable -eight* >or this reason@ a monosyllabi' leIi'al morpheme must ha/e either a long /o-el nu'leus or a diphthong nu'leus* >urthermore@ there is a restri'tion on the u 2u:3 o 2o:3

4+T;O?&.T4O+

!1

number of hea/y syllablesd a simple morpheme 'an 'ontain one su'h syllable at most* 5tress on -ord(le/el is generally on the se'ond(to(last mora of a morpheme* 5tress assignment in the utteran'e is roughly tro'hai'd ho-e/er@ more resear'h is needed on this*

1.6.! Le"ical categories

All 'ontent -ords in Makalero are analysed as either nouns or /erbs* 6asi'ally any leIeme 'an be used for referen'ing and for predi'ating@ hen'e the 'lassifi'ation of morphemes into these t-o 'lasses is not straightfor-ard* +e/ertheless@ 4 argue for the reality of a noun,/erb distin'tion in Makalero based on the fa't that there are some G largely unprodu'ti/e G 'ategory('hanging deri/ational pro'esses@ as -ell as a number of leIemes -hi'h re'ei/e -idely differing@ unrelated meanings if used either as prototypi'al /erbs or nouns* ` 3*1*1 2p* 1 3 and ` 3*!*1 2p* 1!"3 suggest a semanti' feature as a general identifi'ation 'riterion for nouns and /erbs: prototypi'al nouns@ if used predi'ati/ely@ are generally translated as ais an Ib* Prototypi'al /erbs in referring fun'tion@ on the other hand@ may denote athe a't of I( ingb@ athe produ't of I(ingb or aan instrument or ob1e't asso'iated -ith I(ingb* Makalero is largely isolating* The most produ'ti/e morphologi'al pro'ess is redupli'ation* 4n some 'ases@ it appears to deri/e nouns from /erbs* Other-ise@ -ith both nouns and a'ti/ity /erbs@ redupli'ation has a distributi/ity reading@ denoting a /ariety of referents* With stati/e /erbs@ it is generally read as denoting a high degree of the property eIpressed by the /erb* Other than that@ an unprodu'ti/e formati/e -r'an be identified@ -hi'h appears to be broadly nominalising in fun'tion* The suffiI -ini is another -ay of deri/ing nouns from /erbs* Also@ there is an agent noun suffiI -door@ borro-ed ultimately from Portuguese* 5ome more morphologi'al /ariation is found in the domain of /erbs* There is an unprodu'ti/e /erbaliser in the form of the glottal phoneme as -ell as some infle'tional morphology* A sub'lass of semanti'ally intransiti/e /erbs infle't for sub1e't number* A more siAeable group of /erbs eIhibits bound forms used in a 'omplement(/erb 'ompleI* 4n the domain of nouns@ there is e/iden'e for the distin'tion of nouns -ith the feature U+=&MV as opposed to those -ith the feature U(=&MV* The former 'an be marked -ith a plural suffiI@ -hile the latter are not 'ommonly found -ith this suffiI* Among U+=&MV nouns@ there is a subgroup of kinship terms@ -hi'h take a spe'ial plural marker distin't from the one found -ith other human nouns* Another -ay of 'lassifying nouns is -hether or not they take nominal 'lassifiers* There is furthermore a group of nouns -hi'h is obligatorily used in a possessi/e 'onstru'tiond these in'lude kinship terms and nouns that 'an be 'hara'terised as parts of a -hole* ?espite the absen'e of spe'ial marking@ these nouns 'an be 'lassified as inalienably possessed* 4n the /erbal domain@ there is a ma1or distin'tion bet-een stati/e /erbs and a'ti/ity /erbs* Most notably@ the MP(internal ad/erb hau XallF is read as a marker of 'ompletion -ith a'ti/ity /erbs@ but denotes high degree -ith stati/e /erbs* With respe't to synta'ti' /alen'y@ the ma1ority of /erbs is di/alent* While the first of the

!!

.=APTE; 1

t-o arguments asso'iated -ith these /erbs is a sub1e't@ the se'ond 'an take the form of either an ob1e't or a 'omplement* The t-o instantiations of the se'ond argument differ in position* +ominal undergoers are 'ommonly asso'iated -ith the ob1e't position@ -hile lo'ations and goals are usually found in the 'omplement position* Pronominal undergoers 'an stand in either position* A small minority of /erbs is analysed as being a/alent@ i*e* ha/e a /alen'y of Aero and do not take any argument* These generally eIpress temporal or modal 'on'epts su'h as soohe" XyesterdayF@ tone" XperhapsF@ or atanana XfirstF* Other /erbal subgroups are 'onstituted by negati/e /erbs@ interrogati/e /erbs@ numerals@ dei'ti' /erbs@ Huantifiers@ /erbs of physi'al and emotional states and /erbs -ith 'omplement 'lause eItensions* >urthermore@ there is a siAeable group of /erbs -hi'h are marked -ith the third person undergoer prefiI k- in the 'ir'umstan'es dis'ussed in ` "*!*!*#*3 2p* 3:73* 4t is argued that these /erb forms are undergoing a reanalysis into free and bound /erb forms* .losed leIi'al 'ategories in'lude free personal pronouns and their adnominal forms@ i*e* possessi/e pronounsd refleIi/e and re'ipro'al pronouns@ the former also -ith an adnominal formd t-o interrogati/e pronouns@ mu"a- X-ho 26?3F and sa"a- X-hat 26?3F@ as -ell as their adnominal formsd and an indefinite pronoun@ riparipa XanyF* A 'lass of determiners is made up of a series of demonstrati/es as -ell as ho"o XsomeF* They stand in the rightmost slot -ithin the +P* There is also a /ariety of bound morphemes* Apart from the deri/ational morphemes mentioned abo/e@ these in'lude plural suffiIes@ a honorifi' suffiI@ an agent noun suffiI@ a nominaliser suffiI@ an en'liti' relati/e marker and a pro'liti' attributi/e marker* All of the latter are asso'iated -ith the noun phrase* Within the /erb phrase@ there is the abo/e(mentioned third(person ob1e't prefiI k( -hi'h is found -ith a subgroup of /erbs@ the aspe't marker hai and the negati/e marker nomo* >inally@ there is a /ariety of phrase(le/el 'liti's* The ma1ority of these are 'lause linkers@ 'liti'ised to the first 'lause in a seHuen'e@ that spe'ify the semanti' relation of that 'lause to the one that follo-s it*

1.6.# The noun phrase

The noun phrase in Makalero is relati/ely straightfor-ard* The head noun is pre'eded by a possessi/e phrase@ -hile all other dependents 2i*e* modifiers and determiners3 follo- it* Most /ariation is found in the domain of the modifier@ -hi'h may be nominal@ /erbal@ or fully 'lausal* As -ell as unmarked modifiers@ there are modifiers introdu'ed by the 'liti' attributi/e marker ki=* These are generally 'ontrasting the referent in Huestion -ith other possible referents -ith respe't to the Huality they eIpress 2` :*3*:@ p* !9:3* There are@ furthermore@ genuine relati/e senten'es19 in Makalero* The marking of these is 'entred on the sub1e't of the relati/e 'lause@ be it the head of the modifier or -ithin the modifier* 4n the latter 'ase@ the head noun may be marked additionally* This results in a 'ombination of

19

They are 'alled relati/e senten'es rather than relati/e 'lauses be'ause they may be multi'lausal*

4+T;O?&.T4O+

!3

head(marking and dependent(marking patterns* The head of a relati/e senten'e may be either left(dislo'ated or in situ*

1.6.$ The verb phrase

Most of the 'ompleIity of Makalero grammar is found in the domain of the /erb phrase@ -hi'h is stri'tly head(final* The main 'hara'teristi' feature of the /erb phrase is that there are t-o distin't positions for the /erb phrase(internal argument 2see ` "*!@ p* 3173* The first of these positions@ termed ob1e't position@ is the leftmost element of the MP* 4t is follo-ed by the aspe't and negation markers as -ell as /erb( phrase(internal ad/erbials* The se'ond position@ labelled 'omplement position@ immediately pre'edes the /erb and in fa't forms a morphosynta'ti' unit -ith it* This unit may be o/ertly marked by the use of a redu'ed form of the 'omplement or a bound form of the /erb@ or both* The t-o argument positions are mutually eI'lusi/e* Either of them fills the /erbFs single non(sub1e't argument position* While the ob1e't position holds only undergoer +Ps@ the 'omplement position 'an 'ontain either nominal or /erbal 'onstituents* The nominal 'onstituents eIpress either undergoers or lo'ations and goals* Merbal 'omplements@ -hi'h are fully(fledged MPs eI'ept for the fa't that the /erb may be in a dependent form@ generally 'on/ey lo'ati/e or manner information* This 'onstru'tion type is also used in forming 'ausati/es@ 'omparati/es@ modal modifi'ations@ dire'tionality@ and the like 2` "*!*!*"@ p* 3373* There is some degree of fleIibility bet-een the ob1e't position and the 'omplement position for nominal arguments* This holds parti'ularly for pronominal undergoers@ the position of -hi'h /aries 'onsiderably* 4n many instan'es@ it is a leIi'al property of a gi/en /erb -hether a nominal argument is realised as an ob1e't or as a 'omplement* Any ditransiti/e /erb@ e/en those -hi'h are semanti'ally intransiti/e and thus not asso'iated -ith an undergoer parti'ipant@ is 'ompatible -ith a /erbal 'omplement eIpressing lo'ation or manner* 4f su'h ad/erbial information is eIpressed in the 'omplement position@ the undergoer 2if one is asso'iated -ith the /erb in Huestion3 'an no longer stand in the same predi'ation@ but must be eIpressed in a separate predi'ation -ith the help of the light /erb mei XtakeF* The 'omplement thus fills the /erbFs argument position* The fa't that e/en semanti'ally intransiti/e /erbs are 'ompatible -ith a 'omplement sho-s that any /erb is synta'ti'ally di/alent* Merbal modifiers in'lude an aspe't marker hai@ eIpressing the beginning of a nesituation@ a negator nomo as -ell as a /ariety of MP(internal ad/erbials* ` 3*!*3*3*1 2p* 1"73 analyses these as a sub'lass of full /erbs -hi'h 'an alternati/ely be used as modifiers -ithin the MP*

!:

.=APTE; 1

1.6.6 The clause

The unmarked -ord order in the 'lause is 5OM* The sub1e't argument generally has little reper'ussion on the rest of the 'lause* With a number of /erbs -hi'h infle't for sub1e't number@ the number of the sub1e't 2-hether o/ertly marked or not3 determines the form of the /erb* Apart from this@ the use of the refleIi/e pronoun ni marks 'oreferentiality -ith the sub1e't* >urthermore@ the +P 'liti' Y ini@ analysed in ` 7*!*3*1 2p* " 13 as a 'ontrasti/e marker@ appears to be de/eloping into a sub1e't marker@ -hi'h is used -hene/er a speaker 'onsiders -ord order alone not to be enough to 'larify the sub1e't referent* This may be the 'ase in senten'es -ith non( 'anoni'al -ord order@ or if the sub1e't argument is 'onsiderably lo-er on the topi'( -orthiness hierar'hy than the se'ond argument* As mentioned in ` "*! 2p* 3173@ the non(sub1e't argument 'an synta'ti'ally be realised in t-o -ays@ namely either as an ob1e't or as a 'omplement* Either of these is a possible filler of the /erbFs synta'ti' argument position* Any additional arguments need to be eIpressed in a separate 'lause* 5u'h multi('lause units are termed Xsenten'esF and dis'ussed in ` 8 2p* : 13* Operators 2as defined by Man Malin and LaPolla 1778: : 3 that -ork on 'lause(le/el in Makalero are aspe't@ negation and status*

1.6.% The sentence

The unit Xsenten'eF applies to eIpressions that pertain to one e/ent semanti'ally@ but may 'onsist of one or more 'lauses synta'ti'ally* Many senten'es are bi'lausal due to the mismat'h bet-een the rigid stru'ture of the Makalero 'lause@ -hi'h allo-s for the eIpression of t-o synta'ti' arguments -ith a /erb at most@ and the semanti' parti'ipant frames of /erbs@ -hi'h may 'ontain more than t-o parti'ipants* This leads to a situation -here all semanti'ally ditransiti/e states of affairs are ne'essarily eIpressed in t-o 'lauses@ -ith the light /erb mei XtakeF pro/iding an empty argument slot for the supernumerary argument* Another important group of bi'lausal senten'es in/ol/e /erbal 'omplements* These eIpress 'ir'umstantial information of manner or lo'ation and are synta'ti'ally treated like /erbal arguments in that they fill a /erbFs non(sub1e't argument position* 4n fa't@ ob1e'ts and 'omplements are /ariant eIpressions of the same synta'ti' fun'tion* A gi/en 'lause 'an 'ontain only one or the other@ but not both at the same time* .onseHuently@ if a 'lause is 'onstru'ted -ith a /erbal 'omplement@ an undergoer in the /erbFs semanti' parti'ipant frame must be eIpressed in a separate 'lause* Again@ the light /erb mei XtakeF is used for this purpose* The last group of multi('lause senten'es in/ol/e a/alent ad/erbial predi'ates@ -hi'h are 'lausal by themsel/es@ but 'annot form utteran'es on their o-n* 5u'h ad/erbial predi'ates eIpress time@ aspe'tual notions@ as -ell as epistemi' modality* ?eonti' modality as eIpressed by a /ariety of modal /erbs also generally takes the form of a senten'e@ as do /erbs of saying -ith their 'omplement 'lause eItensions and a 'onati/e 'onstru'tion -ith fera XtryF* The

4+T;O?&.T4O+

!"

'onne'tion bet-een the 'lauses in a senten'e may be unmarked@ or marked o/ertly -ith a 'lause(linking 'liti' 2` 3*"*3*1@ p* !" 3* The main formal 'hara'teristi' of the senten'e is the presen'e of a left(deta'hed position 2L?P3@ to the left of the first 'lause* 4t 'an hold either left(dislo'ated arguments@ non(argument topi's or ad/erbial 'lauses* The position is parti'ularly 'onspi'uous in senten'es in/ol/ing ad/erbial predi'ates@ -hi'h may inter/ene bet-een a /erb and its semanti' parti'ipant* An eIample is gi/en in 213* 4n this senten'e@ the ad/erbial 'lause mu"a hai kamu 2ground +54T night3 Xit is already nightF inter/enes bet-een the /erb ma"u X'omeF and its sub1e't argument@ kiloo 23s3* .onseHuently@ this 'lause is dis'ontinuous* The sHuare bra'kets in this senten'e /isualise its stru'ture* A senten'e su'h as 213 sho-s 'learly that the L?P fun'tions abo/e the 'lause le/el* 213 7Kiloo85&6$ in L?P 7mu"a hai
3s

kamu8.L5Yte"e 7ma"u8 MP*

'ome

ground +54T nightYafter

X=e 'ame after night2fall3*F

2eli'#3:3

E/iden'e for an eHui/alent right(deta'hed position is slim 2` 8*#*!@ p* :3:3* 5enten'e types are de'larati/es@ interrogati/es@ imperati/es@ eI'lamations and optati/es* These are in many instan'es not marked as su'h by an o/ert morpheme@ but by intonation patterns*

1.6.& The utterance

The utteran'e is defined as beginning after a silen'e or a final 2falling3 intonation 'ontour and ending on the neIt final intonation 'ontour* As su'h@ it may in'lude /ery short units@ 'onsisting only of a -ord or an eI'lamation@ and relati/ely long stret'hes of spee'h@ 'onsisting of long 'hains of 'lauses* The 'lauses making up these 'hains are linked through 'lause linking 'liti's -hi'h atta'h to the right end of the first 'lause in a seHuen'e* The t-o most 'ommon linkers@ =ini and =isi@ are /ery broad in meaning@ eIpressing only a /ery 'lose(knit 'onne'tion bet-een the 'lauses in the 'ase of =ini@ and a more loose 'onne'tion in the 'ase of =isi* The t-o linkers are 'ommonly asso'iated -ith topi' 'ontinuity and topi' 'hange@ respe'ti/ely* Other linkers eIpress ad/ersati/ity@ 'on'essi/ity@ 'ondition@ 'ompletion@ alternati/es@ 'onseHuen'e@ purposi/eness or reported spee'h* The latter t-o are analysed as subordinating linkers@ on the basis of their possible positioning either on the first 'lause in the seHuen'e@ or on the sub1e't of the se'ond 'lause* All others are found only 'liti'ised on the first 'lause* These linkers sho- no e/iden'e of a synta'ti' hierar'hy bet-een the t-o 'lauses they link* They 'an be 'ombined -ith subordinating 'lause linkersd as su'h@ ` 9*3*# 2p* :993 argues that they are indeterminate as to a possible hiera'hi'al relation bet-een the t-o 'lauses they link* ` 9*3 2p* :913 argues that se/eral 'onstru'tions in Makalero 'an be termed subordinatingd the 'learest of these are relati/e senten'es* Like the subordinating 'lause linkers@ these eIhibit a marking pattern -here the relati/e 'liti' is found on either the head or the sub1e't argument of the relati/e senten'e* On the basis of this

!#

.=APTE; 1

parallel@ the purposi/e and reported spee'h linkers are analysed as subordinating* A rather different@ yet also 'learly subordinated stru'ture are MP 'omplements* These are full predi'ations and are able to stand -ith the same modifiers as an independent 'lause* =o-e/er@ the /erb stands in a dependent 2possibly less /erbal3 form* A third type of 'learly subordinated stru'ture are 'lausal arguments and ad/erbial 'lauses in the left(deta'hed position of a senten'e* A some-hat tri'ky 'ase is presented by /erbs of saying and modal /erbs and their 'omplement 'lause eItensions* ` 3*#*! 2p* 3*#*!3 argues that 'onstru'tions -here the 'omplementiser atta'hes to the /erb of saying@ introdu'ing its 'omplement 'lause eItension@ originate from a 'onstru'tion -here the 'omplement fun'tions as an argument of the /erb of saying* The 'omplementiser looks in those 'ases like a determiner or definite marker@ sho-ing that the 'lause fun'tions as a nominal 'onstituent -ithin that 'lause* As su'h@ 'omplement 'lause eItensions may Hualify as a type of subordinating 'onstru'tion*

1.6.' (iscourse structure

5in'e Makalero has no 'ase marking and no /erbal agreement@ and allo-s for the left(dislo'ation of parti'ipants@ the 'orre't interpretation of grammati'al roles is potentially problemati'* =o-e/er@ there are se/eral de/i'es aiding the hearer in this respe't* An important indi'ator helping referen'e tra'king a'ross the teIt are the 'lause linkers =ini and =isi* As noted in ` 9*1*1 2p* :"83@ the former is 'ommonly asso'iated -ith topi' 2or sub1e't3 'ontinuity@ -hile the use of the latter normally 'oin'ides -ith a topi' 'hange* Another grammati'al item that 'ontributes to referen'e tra'king is the refleIi/e pronoun ni@ -hi'h@ both as an +P on its o-n as -ell as in adnominal use@ denotes 'oreferen'e -ith the sub1e't parti'ipant* As su'h@ any +P 'ontaining this pronoun is ne'essarily a non(sub1e't* >urthermore@ the 'ontrasti/e marker Yini appears to be de/eloping into a sub1e't marker that is used to 'larify sub1e't referen'e in senten'es -ith non('anoni'al -ord order or 'ases -here the sub1e't argument is mu'h lo-er on the topi'(-orthiness hierar'hy than the se'ond argument* Also@ the +P 'liti' Yhaka@ -hi'h ser/es to mark the 'ontradi'tion to a presupposition@ o''urs on sub1e't arguments only* As su'h@ it has a se'ondary sub1e't(marking fun'tion* The ma1ority of senten'es in Makalero follo-s the unmarked 5OM -ord order* =o-e/er@ either argument may be pla'ed in the left(deta'hed position* These left( dislo'ated arguments may be both topi's and fo'i* Most fo'i in the L?P are relati/ely long and may be taken up in situ by a pronoun* As su'h@ ` 7*!*1*! 2p* :793 'on'ludes that the issue in those 'ases is the length of the 'onstituent in Huestion rather than its fo'ality* .onseHuently@ the main fun'tion of the L?P is to hold topi's* ` 7*!*! 2p* " 3 dis'usses an eIpli'it fo'us 'onstru'tion in -hi'h a 'lause -ith an empty position is follo-ed by the same or a /ery similar 'lause -ith that position filled* This fo'us 'onstru'tion Huite ni'ely implements the 'ommon definition of fo'us as the part of a proposition by -hi'h the assertion differs from the presupposition* Makalero disposes of a number of 'liti' or ad/erbial markers -hi'h ha/e to do -ith information stru'ture* The most important of these is the 'ontrasti/e 'liti' =ini* This

4+T;O?&.T4O+

!8

is found on both topi'al and fo'al parti'ipants* 4t is likely an eItension of the 'lause linker Yini and appears to separate the 'ontrasti/e argument from the rest of the 'lause* As su'h@ these 'onstru'tions display a high degree of similarity to 'left 'onstru'tions* The 'liti' =haka marks the 'ontradi'tion of a presupposition* 9a"amuni% -hi'h is predi'ati/e in status@ eHui/alent to a/alent ad/erbial predi'ates@ appears to mark the beginning of a ne- s'ene or e/ent* The 'liti' =oo XtooF is a freHuently used 'on'essi/e linker* =o-e/er@ it also o''urs in a /ariety of 'ases@ often in pairs@ -here this translation does not make sense* As su'h@ it appears to ha/e some information stru'turing use@ -hi'h reHuires further resear'h* On a higher le/el of teIt organisation@ it is noti'eable that narrati/es are normally introdu'ed by a senten'e stating the 'ontent of the teIt* These senten'es are /ery freHuently asso'iated -ith a parti'ular -ord order* TeIts are normally 'on'luded by a 'losing senten'e@ -hi'h states that the speaker is finished* A 'onspi'uous feature of Makalero narrati/es is the use of tail(head linkage* A feature 'hara'teristi' of ritual spee'h@ leIi'al parallelism@ has in some eIpressions also found its -ay into e/eryday spee'h*

1.$ %re&ious 'or"

As a 'onseHuen'e of its debatable linguisti' status as a diale't of Makasae@ Makalero has pre/iously re'ei/ed little attention from linguists* +o mention of it is made in the founding -orks on the languages of the island 2e*g* .apell 17:3a@ 17:3b@ 17::3* The probably earliest Makalero data 'ome from the notes of prof* Antenio de Almeida@ head of the Portuguese MissOo Antropolegi'a de Timor from 17"3* These data 'onsist of hand-ritten -ord lists and eli'ited senten'es as -ell as a tape re'ording of an eli'itation session -hi'h partly 'oin'ides -ith the -ritten materials and in'ludes an eli'ited dialogue bet-een a do'tor and a patient* 4t is not 'lear -hen eIa'tly the data -ere 'olle'ted G Almeida /isited Timor se/eral times in the period bet-een 17"3 and 178"* +either is there a re'ord of -here they -ere 'olle'ted* A number of leIi'al items used seem to be 'loser to Makasae than to the Makalero spoken around 4liomar as des'ribed in the present thesis@ suggesting perhaps that the data -ere 'olle'ted to the north of the Makalero(speaking area* 17 The materials are kept at the 4nstituto de 4n/estigaNOo .ientRfi'a Tropi'al 244.T3 in Lisbon and are partly published in Almeida 2177:3* ! : sa- the publi'ation of =ullFs 'omprehensi/e aThe Papuan Languages of East Timorb@ -hi'h gi/es a -ealth of information* =o-e/er@ the Makalero data sour'es on -hi'h this arti'le is based are 'learly s'ar'e and in'lude se/eral fa'tual errors and in'onsisten'ies* The $apanese(founded Timor Loro 5aFe +ippon .ulture .enter 2T+..3@ founder of the 4liomar library 2see ` 1*!@ p* "3@ brought forth se/eral -orks in and about Makalero by nati/e speakers* Among them are a monolingual Makalero di'tionary by Tomas Mariano Pinto@ a Makalero language 'ourse -ith translations in Tetum and English by the same author@ and a treatise of 4liomar 'ustoms and traditions by
17

An eIample is the -ord XbodyF@ -hi'h de Almeida re'ords as am+ere* 4n the Makalero 'orpus underlying the present thesis@ only afur is used* The 'orresponding Makasae is amu or amu +ere*

!9

.=APTE; 1

Caspar 5eiIas 2the latter@ as -ell as a Tetum translation@ 'an be a''essed online at http:,,---*geo'ities*1p,hkbtls,hakerek,=akerek13,iliomarimakalero*htm3* Lastly@ .hamberlainFs ! 9 monograph XThe struggle in 4liomarF gi/es in/aluable non(linguisti' 2mostly histori'al3 information*

1.( #ethodology and corpus

The 'urrent grammati'al analysis of Makalero is mostly based on a 'orpus 'olle'ted during t-o field trips to East Timor in ! 8 and in ! 7@ respe'ti/ely* The first of these trips -as fi/e months long@ -hile the se'ond -as only three months* The time in East Timor -as partly spent in 4liomar@ the 'entre of the Makalero(speaking area@ and Lospalos@ the main to-n of LautQm distri't* The latter has a siAeable Makalero population and offers better infrastru'ture* .on'retely@ the MQdi'os do Mondo team based in Lospalos kindly let me take ad/antage of their generator* The -ork -as di/ided a''ording to the opportunities offered by these t-o settings* While in 4liomar@ 4 spent my time making re'ordings@ eli'itating@ taking notes and so'ialisingd in Lospalos@ 4 put the materials 'olle'ted in 4liomar into order@ trans'ribed and translated my re'ordings* This di/ision of -ork allo-ed me to periodi'ally organise and update my analysis and form theories@ -hi'h 4 then tested on subseHuent trips to 4liomar* 4n the Makalero(speaking area@ 4 operated in a relati/ely restri'ted 'ir'le* 4 stayed -ith 5r* .arolino AssunNOo and his family@ -ho li/e neIt to the primary s'hool in 4liomar to-n* With the s'hool nearby@ there -ere al-ays numerous people around to talk to* 5r* .arolino and his -ife $oana also introdu'ed me to their friends and family@ -ho are mostly based in the /illages of Ailebere@ Osuhira and Tirilolo 2see Map :3* 4 /isited these /illages repeatedly and 'olle'ted a large range of data* The 'orpus of re'orded teIts is made up of personal memoires@ folk tales@ pro'edural teIts@ and some 'lan and an'estor legends and a small amount of 'hat* Also re'orded are se/eral retellings of MayerFs 217#7(178"3 frog stories* 4n total@ it adds up to o/er 7 hours* 4n trans'ribing@ preferen'e -as gi/en to re'orded teIts that -ere fluently deli/ered and seemed 'oherent* More than 9 hours -orth of re'ordings are trans'ribed and form the basis for the present thesis* Less fluent teIts -here left for a later date and a number of them@ eHualling approIimately 1 hour of re'orded spee'h@ is as yet untrans'ribed* Table 1*1" gi/es a list of trans'ribed and translated re'ordings* TeIt no* 1# 18 ! Title My father The e/a'uation to Mount Matebian =o- to get to Tirilolo 5peaker $oana da 5il/a Olinda Pinto $oana da 5il/a Age Origin !9 ": !9 5eI Length 2mins3 1:3: 1:"7 :33

Tirilolo > 4liomar 44 > Tirilolo >

4+T;O?&.T4O+

!7

!1 !3 !: !" !# 3 31 3! 33 3" 39 :: " "1 "! ": "" ""a "# "8 "9 "7 #1 #! #3 #:

The 'ustoms of 4liomar My an'estorFs life The story of the Loorasa 'lan Traditional healing

Mateus 5eiIas :7 Miranda Eugenio #" ]a/ier >ran'is'o ;osa #" "" #9 9 #9 #9 :9 3! !9 !9 !9 !9 !9 33 !: 3 !9 !9 !1 !1 !9 !9 !9 h:

Osuhira Tirilolo Osuhira Tirilolo Tirilolo Tirilolo Tirilolo Tirilolo Tirilolo Ailebere Tirilolo Tirilolo Tirilolo 4liomar 4 4liomar 4 Tirilolo Osuhira Osuhira Tirilolo Tirilolo 4liomar 4 4liomar 4 Tirilolo Tirilolo Tirilolo \

M M M M M > M M M M > > > M M > > M > > > > > > > M

1::1# #:"" !:!# :3 1:!" !: 9 1::# :31 #:!1 3:!9 1#:!" 1::8 1: 8 !::: !:17 1:1: 1!: ! 1 : : !:33 3:11 1:17 1:: !:1! ":1" 1::7 :::3

Theodoro Araj1o Life during the hunger Alfredo de season $esus My life and religion Maria da Cloria My 'hildhood as an orphan Alfredo de $esus My 'hildhood as an orphan Alfredo de 44 $esus My fatherFs life 5al/ador da 5il/a The history of Ailebere Euri'o $erenimo >rog story 1 $oana da 5il/a A restless night $oana da 5il/a =o- to 'ook mung bean $oana da 5il/a porridge The tale of Laapo and the 4gildo ]imenes king The tale of a naughty 'hild 4gildo ]imenes =o- 4 mistakenly used up Maria da 5il/a my un'leFs massage oil >rog story 1 Mafalda ;osa Wedding 'ustoms in AleIandre 4liomar 5erba ;osa My an'estorFs life $oana da 5il/a =o- to make 'o'onut oil $oana da 5il/a =o- 4 suffered a Marmelinda mis'ariage Este/es =o- 4 got together -ith Marmelinda my husband Este/es =o- -e 'aught a huge sea $oana da 5il/a turtle =o- to make yarn $oana da 5il/a =o- 4 nearly died as a $oana da 5il/a baby The tale of Lerenene and Mar'os da the snake .osta

.=APTE; 1

#" ## #8 #7 8 81 8! 83 8: 8" 8# 89 91 9! 93 9: 9" 9# 97 79 77 1 1 1 1 1 ! 3 "

=o- 4 -orked -ith the international polling staff in 1777 My life

$osQ 5erba dos !8 5antos 3! !9 ! "" "" "" !9 !9 :1 :1 !1 ! 3 3! h# " " " !3 13 33 !3 33 33

Ailebere Akara Tirilolo 4liomar 4 Ailebere Ailebere Ailebere Tirilolo 4liomar 4 Tirilolo Tirilolo

M M > > M M M > M M M

1 : : 1 : ! #:"7 13:3: 3:17 :3 ::33 ": 3 19:1: 1:33 1:1! !:13 1:"3 9:"" 1:! 3:3" !:!8 !:"# #:38 1 :1# "::1 "7:"9 ! : 9 !::! 1 :1!

.arlito $oOo Akara =o- my husband -as /ery $oana da 5il/a ill The tale of +onomata Mirandolina dos ;eis The story of the %omil Mateus da 'lan .osta My life Mateus da .osta My youth Mateus da .osta =o- 4 earned money to go $oana da 5il/a to s'hool A di/or'e Eligio Pereira My life .arlos da .osta The /illage of =iitali .arlos da .osta The prin'iple foods in 5abina Pereira 4liomar A disagreement in the $o/ita family $erenimo My life Mi'ente ;eis My life >ilomena 5eiIas The 4ndonesian o''upation Marta 5eiIas 4ndependen'e Olimpia $erenimo Edu'ation in Timor Olimpia $erenimo The tale of the king of Olimpia 4liomar and the snake $erenimo The importan'e of the Martinho Man Makalero language da .osta 6elo A dangerous monkey , my $osQ da .osta an'estor $erenimo My life Agusto 6arreto >rog stories 1 k ! Agusto Pinto An episode from my life Agusto 6arreto A message to the authorFs Agusto 6arreto father

4liomar 44 > Ailebere Marafal 4liomar 4 Ailebere Marafal Marafal Marafal Ailebere Marafal 4liomar 44 Osuhira 4liomar 44 4liomar 44 > M > > > > > M M M M M M

4+T;O?&.T4O+

31

1 # 1 8 111 113 11: 11" 119 1! 1!! 1!3 1!: 1!" 1!#

A message to the authorFs father ! A message to the authorFs super/isor Traditional beliefs and magi' A 'ar a''ident A tale about a mother >rog story 3 The tale of Laapo and the 'hildrenFs garden A ghost story The story of the Liusoru 'lan My grandfatherFs life My life My life My life

Agusto 6arreto 33 Agusto 6arreto 33 .rispin da .osta .rispin da .osta .arolino AssunNOo Agusto Pinto AdOo Pinto Agusto Pinto .rispin da .osta .rispin da .osta $oni da .osta Martinho Man da .osta 6elo Mafalda ;osa : : :!

4liomar 44 M 4liomar 44 M Ara(ara Ara(ara Tutuala M M M M M M M M M M >

3:1! 3:"8 1:"# ":: 3:"3 1#::7 "::3 !::! ::!8 8:3# 11::: ::!3 3 ::#

!3 Osuhira 19\ \ !3 : : 19 !3 !# Osuhira Ara(ara Ara(ara 4liomar Ailebere Osuhira

Table 1*1": Trans'ribed re'ordings EIamples from re'orded teIts are referen'ed in the grammar part by the number of the re'ording and the number of the senten'e in my trans'ription* An eIample is ! ( !@ -hi'h refers to the se'ond senten'e of re'ording ! * Eli'itation sessions -ere not re'orded* 4ndonesian -as used as the meta(language in these 'ases* The referen'e Xeli'F@ along -ith the number of a gi/en senten'e in my notes@ marks eli'ited senten'es@ senten'es 4 'onstru'ted myself to eli'it grammati'ality 1udgements from nati/e speakers@ as -ell as senten'es 4 o/erheard in dis'ourse* A number of senten'es -ere eli'ited -ith pi'tograms based on 5kopeteas et al* 2! #3* These are designed to lo'ate markers of information stru'ture and are marked as XfisHuestF* 4n ! 9@ 'ourtesy of the >ataluku Language Pro1e't 2http:,,fataluku*'om,3 and 5r* MRtor ;osado MarHues of the 4nstituto de 4n/estigaNOo .ientRfi'a Tropi'al 244.T3@ 4 tra/elled to Lisbon to e/aluate the re'ords on Timorese languages kept in the 44.T* ?uring this trip@ 4 -as thus able to /ie- the earliest Makalero data 'olle'ted some : to # years ago 2see ` 1*8@ p* !83* These data are Huite different in some respe'ts from those gathered by myself in Timor* 4t is not 'lear to -hat eItent these differen'esare due to lo'al /ariants of the language and to -hat eItent language 'hange is in/ol/ed* As a 'onseHuen'e@ -here/er su'h eIamples are used in the follo-ing@ they are marked as su'h -ith the referen'e XalmeidaF* 4 'ombined this trip to Lisbon -ith a short stay in Porto@ -here 4 had made 'onta't -ith OlL/io .abral@ -ho -as at the time a student at Porto uni/ersity and originates from Tirilolo in the

3!

.=APTE; 1

4liomar subdistri't 2see Map :3* OlL/io helped me immensly in trans'ribing re'ordings and ans-ering numerous Huestions 4 had prepared* ?uring the last t-o years of the pro1e't@ OlL/io -as /ery often a/ailable online@ so that 4 'ould ask him more Huestions to test hypotheses through email and 'hat* 5enten'es 'olle'ted during my /isit to Porto are marked as Xpeli'F* Also@ OlL/io did another run of the information stru'ture pi'tograms 2marked as XisHuestF3* 5enten'es gained through email or from an o''asional online 'hat are referen'ed to as X'hatF*

1.) *rthography
While Almeida0s data are in Portuguese orthography@ all other earlier sour'es listed in ` 1*8 2p* !83 use a phonemi' orthography* 4n fa't@ MakaleroFs phonologi'al system is fairly straightfor-ard and in/ol/es little arti'ulatory /ariation* As a 'onseHuen'e@ the orthography used in the present thesis manages -ithout the use of non(A5.44 symbols* All of the earlier -orks@ ho-e/er@ are more or less in'onsistent and idiosyn'ra'ti' in their treatment of /o-el length@ e'ho /o-els and the glottal stop* The orthography used in the present thesis is based on the analysis of these phenomena as brought for-ard in ` ! 2p* 383 and aims to gi/e e/ery morpheme a 'onsistent standardised graphi' representation* 4t is solely based on my o-n understanding and has not been dis'ussed -ith members of the Makalero 'ommunity* As su'h@ it is not meant to be a proposal for a formal Makalero orthography* The fi/e members of the Makalero /o-el in/entory are adeHuately represented by la e i o um* ` !*!*3 2p* #!3 identifies long /o-els as a marginal phenomenon in Makalero* The large ma1ority of long /o-els are predi'table* One su'h reason is the bimorai'ity reHuirement for leIi'al morphemesd monosyllabi' leIi'al morphemes must ha/e a 'ompleI nu'leus 'onsisting of either a diphthong or a long /o-el* 5e'ondly@ /o-el length o''urs in the 'onteIt of the resyllabifi'ation asso'iated -ith e'ho /o-els* This pro'ess is dis'ussed at length in ` !*"*:*1 2p* 933* >inally@ long /o-els are a possible realisation of a seHuen'e of t-o identi'al /o-els interrupted by the glottal morpheme* 4n su'h predi'table 'onteIts@ /o-el length is not represented in the orthography* 2!3 illustrates the t-o former pro'esses 2for the long /o-el realisation of the glottal morpheme@ see belo-3* +ote also that the regular pro'ess of e'ho /o-el addition is not represented in -riting* 2!3 ni fat la"it masan XmotherF XfourF XoldF Xs'aredF Uni:V Ufa:tV U0laTitV@ Ula0Ti:tiV U0masanV@ Uma0sa:naV

=o-e/er@ some leIemes in'lude long /o-els -hi'h 'annot be predi'ted by the abo/e rules* 4n these 'ases@ /o-el length is indi'ated by a double /o-el grapheme@ as sho-n in 233*

4+T;O?&.T4O+

33

233

puulata teersu

XheadF XrosaryF

Unpu:0lataV U0te:osuV

Mo-el length is not represented in the trans'ription used in the ma1ority of the sour'es dis'ussed in ` 1*8 2p* !83* An eI'eption are the Timor Loro 5aFe +ippon .ulture .enter /erna'ular publi'ations@ -hi'h G in'onsistently G appear to use the apostrophe lFm follo-ing a /o-el graph for the purpose* The glides are in the present thesis analysed as allophones of the respe'ti/e high /o-elsd as su'h@ lim and lum@ respe'ti/ely@ are used to represent U1V and U-V* EIamples 2:3 through 293 illustrate the 'orre't reading of the high /o-el graphemes as either /o-els or glides in different 'onteIts* 4n syllable(initial position@ ,i, and ,u, are normally onglides* T-o su'h eIamples are illustrated in 2:3* 5ee ` !*!*! 2p* # 3@ ho-e/er@ for the flu'tuation of the onglides bet-een onset and syllabi' status* +ote that the onset position 'an hold one element at most* 4n the eIamples in 2"3@ this position is o''upied by a 'onsonant@ and as a 'onseHuen'e the follo-ing high /o-el 'annot be read as an onglide* The su''ession of the t-o /o-els in these morphemes is thus disyllabi'@ as sho-n in the phoneti' trans'ription in sHuare bra'kets* Offglides form a 'ompleI nu'leus together -ith the pre'eding /o-el* As su'h@ high /o-el graphemes are read as glides in all 'onteIts -here they follo- another /o-el@ -hether or not they are follo-ed by a 'onsonant@ as in 2#3 and 283@ respe'ti/ely* >inally@ 293 sho-s that morpheme(internally bet-een /o-els@ high /o-els are assigned as onsets to the se'ond syllable* 2:3 2"3 2#3 283 293 uaro" iar nua tia faut heil leu rei ouar saia X-ashF X'ryF XeatF XsleepF XenterF XpullF X'allF Xout-ardsF XmasterF XskirtF U0-arohV U1arV U0nu*-aV U0ti*1aV Ufa-tV Uhe1lV Ule-V Uoe1V U0o*-arV U0sa*1aV

Pre/ious publi'ations generally use l-m and lim for these sounds* The ma1or diffi'ulty in the domain of 'onsonants is the graphi' representation of the glottal phoneme* This phoneme is alternati/ely realised as UhV@ UTV@ or as /o-el length if it o''urs bet-een t-o identi'al /o-els* The distribution of these allophones 'an be predi'ted for some instan'esd in others@ there appears to be free /ariation* 4n some morphemes@ only one or the other realisation is used@ apparently as a leIi'al property* ` !*1*!*1 2p* "13 argues that preglottalisation of phrase(initial /o-els is not phonemi'* +either does morpheme(initial UhV 'ontrast -ith its absen'e*

3:

.=APTE; 1

=o-e/er@ the o''urren'e of the fri'ati/e allophone morpheme(initially does not seem regularly predi'table as that of the preglottalisation* As su'h@ a morpheme( initial glottal fri'ati/e is taken to be a leIi'al property and is represented by the grapheme lhm@ as in the eIamples in 273* >ree /ariation bet-een all possible realisations of the glottal phoneme is found in the 'ase of the glottal /erbaliser 2see ` 3*!*!*1*1@ p* 1!93* The addition of an e'ho /o-el to su'h a /erb results in a seHuen'e of t-o identi'al /o-els interrupted by a glottal phoneme@ -hi'h eIplains the long(/o-el realisation* This /ariation is not refle'ted in the orthographyd the grapheme lFm is used to represent this /erbal ending@ as sho-n in 21 3* 4n all other 'ases@ either lhm or lFm are used depending on the unpredi'table realisation of the glottal phoneme in a gi/en morpheme* +ote@ ho-e/er@ that -ord(internal UhV is relati/ely rare in simple 2non('ompounded3 morphemes* 4n fa't@ of the eIamples gi/en in 2113@ rihun XthousandF is an Austronesian loan@ -hile elehaa Xold manF is likely a 'ompound* 4n 'ases -here a glottal phoneme o''urs bet-een t-o identi'al /o-els -ord(internally@ generally all three possible realisations are found@ as sho-n in 21!3* 273 21 3 hofar hifa" lafu" mutu" fani" 2113 mu"a to"i rihun elehaa da"al na"an ne"et Xne-F X'at'hF Xli/eF Xbe insideF Xbe likeF XgroundF XdigF XthousandF Xold manF XbreakF 2+EC*E]3 XthinkF U0hpfarV@ U0pfarV U0hifaTV U0lafuTV@ U0lafuhV@ Ula0fu:V 2from lafu Xlife3 U0mutuTV@ U0mutuhV@ Umu0tu:V 2from mutu XinsideF3 U0faniTV@ U0fanihV@ Ufa0ni:V U0muTaV U0toTiV U0rihunV Unqlq0ha:V U0daTalV@ Uda:lV@ U0dahalV U0naTanV@ Una:nV@ U0nahanV U0nqTqtV@ Unq:tV

21!3

All earlier trans'riptions are more or less in'onsistent in their representation of this phoneme@ refle'ting its phoneti'ally diffi'ult status 2` !*1*!*1@ p* "13* A morpheme( initial glottal phoneme is either unrepresented or indi'ated by lhm in all sour'es* 4n all other 'onteIts@ the graphemes lhm or lFm are used@ neIt to 'ases -here the glottal phoneme is unrepresented* 4n the domain of the loan 'onsonants@ l'm and l1m are used to represent the affri'ate phonemes@ follo-ing 4ndonesian orthography* lAm represents the /oi'ed al/eo(dental fri'ati/e@ l/m the bilabial 'ontinuant@ and lngm the /elar nasal* As dis'ussed in ` !*# 2p* 983@ stress is largely predi'table* As su'h@ it is not represented in -riting* Aberrant stress as found in some loan -ords is marked -ith an a''ent@ as e*g* in portu*$s XPortugueseF*

4+T;O?&.T4O+

3"

1.1+ *rganisation of the grammar

The present des'ripti/e grammar is organised into eight 'hapters* The first of these@ ` ! 2p* 383@ dis'usses the phonology* LeIi'al 'ategories and the morphology asso'iated -ith them are defined in ` 3 2p* 783* ` : 2p* !833 and ` " 2p* 3183 elaborate on the noun phrase and the /erb phrase@ respe'ti/ely* The remaining 'hapters dis'uss the 'lause and units larger than the 'laused ` # 2p* 3933 is on the 'lause and its 'onstituent parts@ ` 8 2p* : 13 on senten'es and ` 9 2p* :"33 on utteran'es* >inally@ ` 7 2p* :713 treats some issues on the dis'ourse le/el*