Nominal Markers of Filipino, Part 1

In my quest to identify which case markers to use for the Filipino conlang, I will use some of the previous studies done on Philippine-type languages as source materials. In this first part of the post, I want to set out the different grammatical cases of the Filipino conlang nominal markers.


As mentioned in Metonymy of Nang, these particles have various names, some named after a subset of their functions (case markers, case marking particles, determiners, articles, topic markers, common noun markers) and some names encompass other particles with different functions as well (construction markers,  phrase-introducing particles, specifiers, and proclitics). I will use another term mentioned in the same article, nominal marker, which I think captures their function exactly.

Grammatical Case – Malcolm Ross

In 1999 Malcolm Ross wrote The reconstruction of Proto-Malayo-Polynesian construction markers. I can’t find a copy online, but Daniel Kaufman in The Nominalist Hypothesis in Austronesian created a table based on it and shown below:


From the above table, we can see which grammatical cases are the most distinguished among the languages: NOM (4 times), OBL (3 times) , GEN (2 times) and ACC (once). We want to have a conlang that distinguish all four cases.

By 2006,  Ross published Reconstructing the Case-Marking and Personal Pronoun Systems of Proto-Austronesian and standardized the grammatical functions of each “set of case-marked labels similar to those used by Huang, et al. (1999)” so as to describe the languages faithfully and for these labels to be applicable cross-linguistically.

Case Description
NEUT free form with functions including DISJ, TPC, FRFOC and one or more core grammatical functions (i.e. of SBJ, AGT and PAT)
NOM free or clitic form serving as SBJ or VSBJ
GEN free or clitic form serving as AGT (in NAF) and often as PSRA or occasionally PSRN
PSR free form serving as PSRA and/or PSRN
ACC free form serving as PAT only
OBL free form serving as PAT and LOC (and sometimes in other peripheral functions)
LOC free form serving as LOC (and sometimes in other peripheral functions)

The other acronyms above are defined below as:

Label Scope Description
DISJ Information Structure marking of disjunctive, i.e. one-word answer
TPC Information Structure marking of fronted topic, often followed by a topic marker
FRFOC Information Structure marking of focus-fronted argument
PREDN Non-Verb Clause Functions marking of predicate noun
NSBJ Non-Verb Clause Functions marking of subject of nonverbal clause
AVSBJ Special Subject Functions marking of subject of actor-voice verbal clause
SBJ Special Subject Functions marking of subject of verbal and nonverbal clauses
VSBJ Core Verb Args marking of Subject of verbal clause
AGT Core Verb Args marking of agent argument of undergoer voice clause
PAT Core Verb Args marking of patient argument of actor-voice clause, third core argument of undergoer voice clause (e.g. patient when, say, a location or instrument is subject)
LOC Peripheral Args marking of location argument (‘at my place’, ‘from me’ etc), goal, source
BEN Peripheral Args marking of beneficiary argument
PSRA Possessor Functions marking of possessor adnominal (‘my’ etc) without a ligature
PSRN Possessor Functions marking of possessor nominal (‘mine’ etc; in some languages it also occurs adnominally with a ligature)

Notes on the above case–marked labels:

  1. OBL includes LOC and PAT functions together, which I don’t want in the Filipino conlang. I want nominal markers which make a distinction between LOC and PAT so we will not have the OBL case markers and have LOC and ACC nominal markers instead separately. The  ACC marker can be used for both PAT or BEN.
  2. I will retain the GEN nominal markers which marks AGT (in NAF sentences), PSRA and PSRN.
  3. NOM marks either the VSBJ only or also the NSBJ as well in a SBJ. I want the VSBJ only in a Filipino conlang.
  4. NEUT marks DISJ, TPC and FRFOC plus one or more core grammatical function. We want to exclude any core grammatical functions from this marking and add instead the PREDN and/or NSBJ in the conlang.
  5. To wrap up, we want a Filipino constructed language with NEUT, NOM, GEN, ACC and LOC case markers.

Looking at the Table 1 of Common Case Markers in Ross 2006, only Amis language (Sakizaya dialect) fits well with all 5 nominal cases that we want with no extra markers. The Amis data is presented below although we don’t know if the NOM and NEUT forms function as we want them to be.


For personal pronouns, reading Ross’s paper and looking for a Formosan language that matches the above requirements in function usage, we have the following summary:

  1. NEUT personal pronouns of Paiwan and Puyuma seems a close match to what I want, functioning as FOCFR, TOP, DISJ and nothing else. Saaroa and Bunun adds to these forms other uses like PAT use (na OBL + NEU) and PAT/AGT use ([k]i OBL + NEUT) respectively. For Amis its used for PREDN and TOP while Rukai just TOP. Mayrinax Atayal uses it for TOP, DISJ and PREDN but also uses for SBJ/AGT/PAT by prepending with iʔ NOM/OBL case marker, as BEN with prepended niʔ GEN, and  LOC by prepended kiʔ. Pazeh also uses NEUT for TOP and NSBJ but also for PAT and VSBJ. I don’t like the following for including core roles:  Kanakanavu (DISJ, AVSBJ), Tsou (VSBJ, PAT), Kavalan (FOCFR, TOP, SBJ) and Siraya (TOP, SBJ). No NEUT personal pronouns in Saisiyat and Thao. I want a NEUT form that marks DISJ, TOP, FRFOC and no other in the conlang. I’m still undecided whether NSBJ and/or PREDN should be marked as NEUT or NOM. If it’s  NEUT, then equational sentences’ subject and predicate might have different markers, and subject markers for verbal and non-verbal sentences would be different which I don’t want.
  2. NOM personal pronouns in Saisiyat, Saaroa, Paiwan, Rukai, Puyuma, Amis and Mayrinax Atayal are used for SBJ, while other Atayal dialects, Thao, Siraya, Pazeh, Kanakanavu and Kavalan  use it for VSBJ. Bunun has two NOM forms, one for VSBJ and another for SBJ (VSBJ and NSBJ). Tsou have combined NOM-GEN. I want a NOM form that marks only the VSBJ of verbal clauses.
  3. LOC personal pronouns for Pazeh seems to be pure LOC and can be optionally preceded with di LOC case marker and means “at my place, etc”. Saisiyat, Kavalan and Bunun have LOC forms that are distinct from ACC or PAT. Atayal LOC is merged with PAT/PSRN in an OBL marker, Paiwan’s LOC marged with PAT in an OBL marker, and Rukai’s LOC includes with PAT other peripheral roles in an OBL. In Sakizaya Amis, OBL only has PAT functions but has LOC as well in Nataoran and Central dialects. No mention of LOC forms in Thao, Siraya, Tsou, Kanakanavu, Saaroa and Puyuma. I want a pure LOC function distinct from other functions.
  4. ACC personal pronouns. Saaroa PAT is either ACC or OBL+NEUT. Pazeh and Tsou use the NEUT for PAT. Thao has two ACC forms, one used for PAT and the other for BEN. Kanakanavu PAT is merged with BEN in an OBL marker. Siraya and Kavalan’s ACC is only PAT. Sakizaya Amis ACC is PAT only, with other dialects PAT+LOC. Atayal’s PAT is merged with LOC and PSRN in an OBL form. Paiwan’s’ PAT is merged with LOC in an OBL. Rukai PAT is merged with other roles in an OBL. Bunun’s PAT is combined with AGT in a ACC-GEN marker. Saisiyat has distinct ACC and BEN forms. I want a ACC form that can function both as PAT or BEN.
  5. GEN personal pronouns in Pazeh has AGT and PSRN roles, and uses a ligature with PSRN to use it as PSRA. Atayal’s GEN includes AGT and PSRA, and has a separate PSRN forms from NEUT forms. Thao’s GEN is a combined AGT/PSRA/PSRN form. Siraya, Kanakanavu and Saaroa’s GEN is AGT/PSRA. Tsou’s AGT/PSRA is combined with NOM-GEN. Paiwan has two GEN forms, one for PSRA/AGT and one for PSRN/AGT. Puyuma has 4 forms for PSRA, PSRN, GEN as AGT and a GEN-OBL for AGT/PAT. Amis has 2 GEN forms, one for GEN/PSRA, the other for GEN/PSRA/PSRN. Bunun has two forms, with  PSR forms separate from GEN-ACC forms. Kavalan has separate forms for GEN (as AGT) and PSR (za+GEN). I want a GEN form that functions as AGT, PSRA and PSRN.

Although Paiwan, Puyuma and Amis have the right NEUT use, their NOM is used for both NSBJ and VSBJ, which I don’t want. Pazeh, Saisiyat, Kavalan and Bunun have LOC functions, but only Kavalan and Saisiyat have distinct ACC forms.

Although Kavalan has the right NEUT, NOM, ACC, GEN+PSR and LOC functions, I don’t like the forms themselves (LOC ending in –an, ACC prefixed with tima-, PSR prefixed with za-). For Sakizaya Amis which we mentioned earlier with the same 5 functions, the OBL also ends in –an (Central) or –anan (Nataoran/Sakizaya). Amis also does not have a distinct LOC form for personal pronouns, with Central and Nataoran dialects’ OBL a combined PAT and LOC which we don’t want. Amis’ Central dialect does not distinguish NEUT from NOM but instead just use the NOM forms. Siraya has 4 of the functions right (NEUT, NOM, GEN, ACC) but the ACC or OBL is also ending in –an. Pazeh has got 3 functions right (NOM, GEN, LOC) but its ACC (as PAT) and NEUT are combined in one form and the LOC form ends in –an. Kanakanavu has 4 forms (NEUT, NOM, GEN, OBL/ACC) right but the OBL also ending in –an and the OBL is derived from NEUT.

Formosan Nominative Nominal Markers

In Ross’ Reconstructing the Case-Marking and Personal Pronoun Systems of Proto-Austronesian, the Formosan data show two groups of languages where s– or k– are the formatives for NOM forms. He gave the Formosan forms of these nominal markers and he reconstructed the NOM forms as starting with *k-.

We’ll go over these languages in groups and check if we can show that k– is weak for NOM but stronger for ACC or OBL.

  1. The first group of languages (Saisiyat, Rukai, Bunun) have k– initial forms in (a) NOM and (b) either OBL or ACC,  in both personal and common nouns. (In Bunun the first of the pair is used after a vowel.)
    Common Nouns TOP NOM GEN ACC OBL LOC
    Saisiyat ka no [ka] ka no [BEN] ray
    Budai Rukai ku, ka ka, ku
    Takbanuad Bunun ka ki
    Personal Nouns TOP NOM GEN ACC OBL LOC
    Saisiyat (hi) ni hi ?ini kan, kala
    Budai Rukai ku, ka ki
    Takbanuad Bunun V=kat

    The most pertinent comment is that of Yeh (1991) stating that subject NPS occurs with ACC markers in Saisiyat: could it be the same reason why hi appears as ACC in Ross’ Table 2 although it is cognate with PMP NOM *si? In wikipedia, hi is NOM as well as ACC, so I put it in the NOM column as well above inside a parentheses.  Another noteworthy data about Saisiyat is that it retained the LOC d– initial in the form of ray

    Ross (2006) concluded when discussing about the forms that “there has been a merger of NOM and ACC forms in Saisiyat, followed by loss of NOM forms”, where he means the NOM k– forms were lost.  But the opposite scenario is also plausible, that the forms with s– initials were the ones lost in the NOM (The same can be said of Rukai and Bunun.) since the personal pronouns of Saisiyat below starts with kan- for OBL/LOC, a pattern found in a lot of Philippine languages and no NOM personal pronoun forms even starts with k– or s– which supports the alternative scenario. He himself said :”Since personal pronoun systems in Formosan languages often incorporate the case-markers more generally used in noun phrases, and it is impossible to interpret the histories of the pronouns without taking account of the case-markers….”

    Saisiyat Personal Pronouns OBL/LOC
    1st Singular kanman
    2nd Singular


    3rd Singular kansia
    1st Plural Inclusive


    1st Plural Exclusive kan-yami
    2nd Plural kan-moyo
    3rd Plural kanlasia

    Does it follow as well that the LOC (directional locatives?) kan/kala in the personal markers should instead be classified as ACC or OBL since they mark benefactives? That should make sense.

    But the presence of accusative, dative and possessive forms in the Wikipedia confuses things and we don’t know how these can be distinguished unless we get hold of some more Saisiyat sentences. In Budai Rukai, it’s the TOP markers that begin with ku- as well as its personal pronouns in NEUT or TOP (from Ross 2006):image

  2. Amis has a k– initial NOM marker in the common nouns but not in the personal noun markers. What is noteworthy is that the NOM personal nominal marker ci is cognate with PMP *si, thus supports s– initial. Common ACC t– initial markers are different from NOM markers, unlike personal markers where ACC seems to be derived from NOM through affixation. (See table below. Sakizaya Amis has a Definiteness distinction (D) in common nouns and number distinctions in personal nouns.)
    Common Nouns TOP NOM GEN ACC OBL LOC
    Sakizaya Amis u
    +D: kiya
    +D: iya
    +D: [nu]niya
    +D: tiya
    Personal Nouns TOP NOM GEN ACC OBL LOC
    Sakizaya Amis S: ci
    P: ca
    S: ni
    P: na

    Amis NOM personal pronouns though overwhelmingly show k– (from Ross 2006), with the ACC/OBL showing t– initial which I think is cognate with s– initial forms. image

    But Amis personal noun NOM marker form ci is cognate with that of Paiwan ti and  PMP *si, which conflicts with its  personal pronoun formative k-. More languages support s– formative for NOM personal markers and k– in the OBL/ACC case. Additionally as Ross himself said: “An examination of the data in Appendix B shows a number of cases in which a new NEUT or NOM set has been formed by preposing a case-marker to an existing pronoun set. A transparent instance appears in Amis (B.12), where the old NOM set has become the NEUT set, and a new NOM set has been created by preposing the NOM common case-marker to the new NEUT set.”

  3. Atayal seems to be a different case since although the NOM ku starts with k-, the whole syllable is retained in the GEN and OBL forms of the common nouns like a base where a prefix is added, so we can’t generalize that NOM has k– form or case marked and starts with k– as opposed to other grammatical cases starting with C-. All we can say is that the form ku is originally NOM. But this can’t be used to show that there was a movement from ACC to NOM of k– forms as well. In personal marker forms, k– initial forms function as LOC and not as NOM. Mayrinax Atayal has a Specificity distinction (+SP/-SP).
    Common Nouns TOP NOM GEN ACC OBL LOC
    Mayrinax Atayal +SP: ku?
    –SP: a?
    +SP: nku?
    –SP: na?
    +SP: cku?
    –SP: cu?
    Personal Nouns TOP NOM GEN ACC OBL LOC
    Mayrinax Atayal i? ni? i? ki?

    Wulai Atayal’s personal pronouns though have a clear s– formative in the NOM (from Wikipedia).  image

    Cross-referencing with Sakizaya/Nataoran Amis common markers or personal markers with Wulai Atayal’s NOM pronouns, Sakizaya Amis’ OBL pronoun seems to have an obvious origin, and that Sakizaya Amis’ OBL pronouns could be originally NOM pronouns.

    Sakizaya/Nataoran Amis LOC Common Marker Sakizaya Amis ACC  
    Singular Peronal Marker
    Wulai Atayal NOM Pronouns Sakizaya Amis OBL Pronouns
    1s. i / [i]an cian saku?, mu, ku? [i] takuan
    2s. i / [i]an cian su? [i] tisuan
    3s. i / [i]an cian [i] cira-an
    1p (incl.) i / [i]an cian ta? [i] titaan, [i] kitaan
    1p (excl.) i / [i]an cian sami [i] taman
    2p. i / [i]an cian simu [i] tamuan
    3p. i / [i]an cian [i] tuhn-an
  4. Saaroa’s ka according to Radetzky (Ross 2006) is from a demonstrative grammaticized as a definite marker in all contexts, which implies that it does not only serve NOM cases.
    Common Nouns TOP NOM GEN ACC OBL LOC
    Saaroa a, ka na
  5. Pazeh doesn’t have personal markers too, but its forms actually look like they were originally the personal noun forms. Although its common noun marker starts with k-, its personal pronouns don’t start with either k– or s– either and there is no definitive NOM or LOC/ACC/OBL consonant initial formative for the personal pronouns.

    Common Nouns TOP NOM GEN ACC OBL LOC
    Pazeh ki ni o di

From the above languages with k– in NOM, we can see that k– support is weak for it to be reconstructed as NOM marker onset. Now we come to languages where NOM is cognate with s-.

  1. Siraya has s– onset for NOM and k– onset ACC (see Saisiyat above as well) for common noun ACC markers, with the NOM ta form seemingly cognate with s– and k– form as OBL. But Siraya NOM ta is both used for for common and personal pronouns. Siraya’s GEN na does not mark agentive of genitive, so was excluded from here.
    Common Nouns TOP NOM GEN ACC OBL LOC
    Siraya ta ki tu
    Personal Nouns TOP NOM GEN ACC OBL LOC
    Siraya   ta     -ang  
  2. Thao and Kanakanavu share the same marker for both NOM and OBL, in both languages they begin with s-, a model similar to Saisiyat above.

    Common Nouns TOP NOM GEN ACC OBL LOC
    Thao sa sa tu
    Kanakanavu s[u]a, si s[u]a na

    Thao, Kanakanavu and Saaroa don’t have personal markers. Thao’s personal pronouns don’t start with either s– or k– and there is no definitive NOM or LOC/ACC/OBL consonant initial formative for the personal pronouns, but Kanakanavu’s Personal pronoun focused suffixed forms starts with k-. Are these NOM or OBL/ACC forms? image      Thao and Kanakanavu sharing the same marker for both NOM and OBL that begin with s- (hi) for common nouns is similar to Saisiyat for personal nouns.

  3. Paiwan’s common noun marker ta which is cognate to s– is OBL and not NOM just like in other Meso Philippine languages. But its OBL personal pronouns all starts with k– (from Ross 2006) as well as the personal markers. Notable is the loss of LOC and possible TOP markers, with OBL functioning both as ACC and LOC in personal pronouns.
    Common Nouns TOP NOM GEN ACC OBL LOC
    Paiwan a nua tua, ta, tu
    Personal Nouns TOP NOM GEN ACC OBL LOC
    Paiwan S: ti
    P: tia
    S: ni
    P: nia
    S: cay
    P: c(a)ya


  4. Puyuma does not have s– initial NOM markers but shows k– onsets in the OBL. Ross said that another dialect, Nanwang has lost the GEN forms, but it is clear from the data that even Tamalakaw has lost its GEN as well (-D-SP & +D) with OBL taking over their function, although its not complete, with kana –D+SP still not used as GEN. Tamalakaw distinctions Indefinite Specific and Indefinite nonspecific corresponds to Nanwang Definite and Indefinite, with Tamalakaw Definite ni looking like its derived from a GEN marker, like Pazeh GEN ni.
    Common Nouns TOP NOM GEN ACC OBL LOC
    Tamalakaw Puyuma +D: ni
    –D+SP: [i]na 
    –D-SP: a
    +D: nina
    –D+SP: —
    –D-SP: za
      +D: nina
    –D+SP: kana
    –D-SP: za
    Personal Nouns TOP NOM GEN ACC OBL LOC
    Tamalakaw Puyuma   S: i 
    P: na
    ni kani

    Puyuma OBL personal pronouns (Ross 2006) all starts with kan-:


To sum up, support for NOM s– onset in personal marker forms is from Saisiyat ACC hi, Paiwan NOM ti, Amis NOM ci, and PMP NOM si personal noun markers and from the NOM personal pronouns of  Wulai Atayal. On common marker forms, support comes from Siraya, Kanakanavu and possibly Thao.

From the data above, its possible that the changes that happened to the common noun markers are the following, based solely on their phonetic form.

  1. NEUT/NOM/GEN/LOC > ∅, ACC > NOM/ACC: Saisiyat, Rukai and Bunun lost their NEUT and NOM markers, with the ACC markers expanding to NOM. Rukai and Bunun lost their GEN and LOC, while Saisiyat GEN also used as OBL for benefactives.
  2. NEUT/GEN/ACC > ∅, NOM > NOM/ACC : Thao, Kanakanavu and Siraya lost their NEUT and GEN, with Thao and Kanakanavu further losing their ACC and being taken over by NOM.  Both Thao and Siraya has LOC with t- while Kanakanavu has n-.
  3. NEUT/NOM/LOC > ∅, ACC > NOM/ACC , NEUT > OBL/LOC : Amis and Pazeh  lost their NOM markers, with the ACC markers doing double duty as well as NOM function.  Furthermore, Pazeh lost the NEUT function of the NEUT form, being used instead for OBL. Amis LOC is from NEUT, with the ACC t- form seemingly cognate with Thao and Siraya LOC t-.
  4. NOM/ACC/LOC > ∅, GEN > NOM , NEUT > LOC : Puyuma lost its NOM, ACC and LOC,  the GEN becoming NOM and the NEUT becoming LOC. It created new forms for use in both GEN and OBL.
  5. NEUT > ∅, NEUT > LOC : Atayal lost its NEUT, reformulated its NOM, GEN and OBL, and derived the LOC from a NEUT.
  6. NEUT/GEN/ACC/LOC > ∅, NEUT > NOM , GEN > OBL : Saaroa lost its NEUT, ACC and LOC, with one of their NEUT moving to take on one of NOM functions, and GEN losing its function to take on OBL.
  7. ACC(OBL?)/LOC > ∅, NOM > OBL, NEUT > NOM : Paiwan lost its  ACC (OBL?) and LOC, and its NOM move to become one of the OBLs, NEUT lost its NEUT function and move to NOM. It has additional OBL of t- onset. It formed additional GEN and OBL.

Note these common direction of changes:

  1. ACC > NOM/ACC – happened in Saisiyat, Rukai, Bunun, Thao, Kanakanavu, Siraya, Amis and Pazeh.
  2. NEUT > OBL or LOC – happened in Amis, Pazeh, Puyuma and Atayal
  3. NEUT > NOM – happened in Saaroa and Paiwan
  4. NEUT > ∅ – happened in Saisiyat, Rukai, Bunun, Thao, Kanakanavu, Siraya, Amis, Pazeh, Atayal and Saaroa.
  5. LOC > ∅ – happened in Saisiyat, Rukai, Bunun, Amis, Pazeh, Puyuma and Saaroa and Paiwan

Why was there a movement in Amis, Pazeh,  Saisiyat, Rukai and Bunun  from ACC/OBL to NOM of k-, while Paiwan, Thao and Kanakanavu from NOM to ACC/OBL of s-?  Were Paiwan, Thao and Kanakanavu influenced by the loss of LOC d– forms? It didn’t affect Amis, Rukai and Bunun.

Grammatical Case – Laurence Reid

A much earlier study on these markers is Laurence Reid’s Problems in the Reconstruction of Proto-Philippine Construction Markers in 1978, where he used a 5-way distinction to label the different cases of construction markers (nominal markers in this post) in Philippine languages, descriptions of which are provided below:

Acronym Label Description
TOP Topic Mark the predicative NP in an equational sentence, as well as a topicalised NP
NOM Nominative Marks the subject of the sentence
GEN Genitive The case form which marks the Agent of a ‘passive’ verb or Experiencer, which throughout the Philippines also marks the Possessor in a possessive construction.
ACC Accusative The case form that typically marks an indefinite Object in an unembedded sentence. In such languages, a definite object is either subjectivalised or reinterpreted as a kind of a Locative and marked with a Locative case form.
LOC Locative Marks for location and direction, plus the definite object of unembedded sentence.

He used the above to identify 6 determiner systems in Philippine languages:

Type TOP NOM GEN ACC LOC Example Languages


B Blaan, Ilokano
2 A B C Agta, Bontok, Kalinga, Balangaw
3 A B C Tagalog, Hiligaynon, Tagbanwa, Batak, Mamanwa, Mansaka, Subanon
4 A B C D Kapampangan, Maranao, Cebuano
5 A B C D Inibaloi, Pangasinan
6 A B C D E Ivatan, Itbayaten, Amis

Notes on the above table:

  1. Type #2 and Type #5 are similar, in that the ACC and LOC share the same marker. The only difference is the introduction of a separate TOP marker in Type #5 in the common markers for the only 2 languages in this type. Example languages are those found in Northern Philippines, and Reid said its pretty widespread among Cordilleran languages.
  2. Type #3 and Type #4 are similar in that LOC and GEN or ACC have distinct markers, the  only difference is the presence of a separate marker for indefinite objects in Type #4. Example languages are those found in Central Philippines (Visayas, Central and Southern Luzon, Northern and Eastern Mindanao).
    • I have an objection with Hiligaynon being classed as Type #3. In the example provided, Cebuano is of type #4, with ug as the separate ACC marker for indefinite object. Yet Hiligaynon, classed as Type #3, has a similar indefinite object marker sing, and there could be a few more Meso-Philippine languages with such definite/indefinite distinction in the GEN, like Samarnon (hin), Romblomanon (ning), Bisakol (sin) and Bikol (nin).
    • Another objection is that these indefinite markers are used not just for indefinite ACC but also for indefinite GEN, so there is actually no separate ACC marker from GEN in Cebuano, Samarnon, Romblomanon, Bisakol and Bikol but instead separate definite vs. indefinite GEN-ACC markers. I haven’t studied Kapampangan (yang) or Maranao much so can’t conclude that there is no Type #4 separate from Type #3. As Reid said, this type occurs in only a few languages.
    • Maranao ACC seems to have the same form as that of Ivatan ACC. This is discussed in NOM.
  3. The difference between Types #2/#5 and #3/#4 is that according to Reid, the indefinite object or ACC is marked as GEN in Types #3/#4 but LOC in Types #2/#5. Both groups mark the definite ACC as LOC. It is unclear if in Type #1 the LOC also marks the definite objects or ACC but would assume it does.
  4. Type #6 has Ivatan as the sole representative, has possible rearranged its nominal markers. 
    • Its NOM ?u is from the NEUT form.
    • It has a separate indefinite ACC common marker su which I think came from an original NOM with Pangasinan, Kallahan and Maranao providing NOM common marker examples.
    • Its GEN common nu marker is also used to mark NEUT.
    • Its ACC also marks manner phrases.
    • Although Reid mentioned that the Amis system is similar to Type #6 of Ivatan, its only a superficial similarity. Sakizaya Amis’ NEUT and GEN forms are not identical, and Amis LOC form is derived from NEUT.
    • Ivatan might not be a single language but two with Itbayaten as a separate language and not a dialect. These languages border Formosan languages in the far north of the Philippines.

In “A Brief Syntactic Typology of Philippine Languages” (2003), Lawrence Reid with Hsiu-chuan Liao described the same 5 cases as below:

  1. Topic (TOP) – marks the definite, non-cased NP constituent which acts as the theme of the construction. definite noun (as predicate),  topic (fronted noun phrase, or theme),
  2. Nominative (NOM) – marks the least indispensable complement of a basic predication (the subject of the sentence, predicate NP of an equational sentence), and the one that is most likely to undergo deletion under conditions of coreference in a relative clause, whether transitive or intransitive.
  3. Genitive (GEN) – marks the noun phrases that are the Correspondents (or ‘possessors’) of possessed nouns and the Agents of passive transitive constructions.
  4. Locative (LCV) – location, direction, time NPs, purpose expressions, means (instruments) and correspondents (2nd complement of s dyadic intransitive clause).
  5. Oblique (OBL) – marks indefinite nouns that are the 2nd complement of dyadic intransitive constructions.

Reid concluded

“It is probable that Proto-Philippines did not distinguish between determiners which mark Topic and Nominative NPs. It is also probable that these were distinct from those which marked the Genitive and Locative NPs. It is also probable that the determiners which marked Genitive and Locative NPs were different from each other since no Philippine language uses the same case form for these two NPs. Whether Proto-Philippines had an Accusative form distinct from both the Genitive and Locative, or whether it was the Genitive form or the Locative form which marked indefinite object is unclear.”

Grammatical Case – Raleigh Ferrel

There is another study on nominal markers done by Raleigh Ferrel “Construction Markers and Subgrouping of Formosan Languages” which I have not really look into yet.

Nominal Marker Case Forms

I want the nominal markers to be monosyllables in the form CV, just like most of existing forms in the various languages. I want the onset consonant to mark the grammatical case, with uniform onset consonants cross-cutting usage in common vs. personal/proper names, definiteness and distance distinctions in the same grammatical case.

These  initial consonants were mentioned by Joseph Finney in The Yap Language: Whence and How?:

“These monosyllables all begin with members of a small set of initial consonants: zero, n, k, s, t and (not in Formosa) d. Though PCMs (prepositions and case markers) are notorious for having multiple meanings and for shifting meaning, each of these initials has kept an old core meaning to a remarkable degree. Those with zero initial serve as the unmarked Case 1, ofter called nominative, absolutive, subject or focus. There is also a locative with initial zero. Those with initial n– (Case 2) are typically genitive (“of”), and in many languages serve also as the marked agent, the Doer that is not the syntactic subject, in constructions called passive or ergative. The particles with initial k, t, d, and s may all have originated as allatives (“to”, “toward”), and we’ll call them Case 3. The Case 3 forms compete with with one another in various languages and survives in niches.”

Our forms will be the following:

A. Genitive Form

Reid said that there is general consensus that *ni is the case form for Genitive. And Malcolm Ross said: “The distribution of initial consonants is not chance. Most obvious is that GEN markers usually begin with n-, a long recognised fact.”.

n– initial forms is also prevalent not just in nominal markers but also in personal pronouns in the GEN case. Since it is generally agreed that GEN starts with n-, there is no need to discuss it further here.

B. Locative Form 

What is the form of the Locative case marker? Although Reid said that it is either di or sa, I think di has more evidentiary support. He also mentioned in The Early Switch Hypothesis: Linguistic Evidence for Contact between Negrito and Austronesians :

“Although di occurs in many languages with various other case making functions, it is as a locative common noun marker that it is reconstructible for Proto-Philippines. This is the function it has in Bilaan in the very south of the Philippines, and although it has been replaced in most other Philippine languages by some other locative markers, it is retained in languages throughout the Philippines as the initial formative of locative demonstratives.”

Some more reasons to suppose that the LOC marker should start with d-:

  1. Looking at the various case marker / determiner systems (Reid 1978), it seems that those where the LOC functions solely as Locative has the form d-. These include Type 1 (Blaan di) and Type 6 (Ivatan du/di). Ilokano’s  ɥiti seems to be like Amis in having ɥi. Types 2, 3, 4 and 5 have no pure locative case markers and their markers starts with either k-, s-/t- or ɥ-. Type 2 and Type 5 combine in one case form both the Accusative (definite and indefinite objects) and Locative functions. Although Type 3 and Type 4 has separate Locative, this is also used to mark definite objects of non-embedded transitive clauses. The nominal markers with a Locative function in these types (2,3,4,5) must have been originally from other case forms.

    Common Noun Markers

    Language Type TOP NOM GEN ACC LOC
    Ilokano 1 ti ɥiti
    Bilaan 1 ɥi di
    Ivatan 6 nu ɥu nu su du
    Itbayaten 6?4?          
    Amis 6 ɥu ku nu tu ɥi [….an]

    Personal Noun Markers

    Language Type TOP NOM GEN ACC LOC
    Ilokano 1 ni ken ni
    Bilaan 1 ku, kane
    Ivatan 6 si si ni di
    Itbayaten 6?4?   i ni si di
    Amis 6 ɥi ci ni   ɥi ci
  2. Demonstratives have prefixed di– or starts with d-
      here there(proximal) there
      Ilokano ditoy dita
      Tagalog dine diyan
      Tiruray dini diyaqan diyoq, diyoqo
  3. Di is a generic locative marker in Malay and other languages. Traditionally called a preposition in Malay, some examples are found in here, here and here.
Malay English Tagalog Bikol Conlang
di sini here dito, dine digdi dini
di sana/situ there, over there doon duman ditu
di dalam inside, in sa loob sa laog di leog
di luar outside sa labas sa luwas di lubas
di tepi at the side sa tabi sa gilid di tebi?
di antara between sa gitna sa tahaw di tahaw
di tengah in the middle sa gitna sa tanga? di tanga?
di bawah below, downstairs, beneath, under sa baba sa baba di baba?
di atas above, upstairs, on top sa taas sa taas di taas
di puncak on top of sa ibabaw sa ibabaw di babaw
di belakang behind sa likod sa likod di likod
di depan/hadapan in front, before sa harap sa atubangan di harap
di seberang across (the street) from sa ibayo sa balyo di ballo
di sebelah next (door) to, beside sa kabila sa kataid di kabila
di samping beside sa tabi sa kataid di tebi
di Indonesia in Indonesia sa Indonesia sa Indonesia di Indonesia
di rumah at home sa bahay sa harong di baley / barong
di meja on the table sa mesa sa lamesa di mesa
di gelas in the glass sa salamin sa salming di salming
di hospital at hospital sa hospital sa hospital di hospital
di dalam rumah inside the house sa loob ng bahay sa laog kang harong di leog nang baley
di luar mobil outside the car sa labas ng kotse sa luas kan kotse di labas nang koce

Paranan also has di marker (di balay “to the house”, di Manila “to Manila”). Some of the languages with di markers mentioned above are Ivatan, Itbayaten, Blaan, Malay, Casiguran Dumagat and Paranan.

For the LOC forms starting with s-/t- or k-, see the discussion under Accusative Form.

C. Accusative Form 

What should be the Accusative form? The ACC function comes out marked either as:

  1. Marked separately (Types #4 and #6)
  2. Combined with LOC (Types #2 & #5)
  3. Combined with GEN (Type #3)
  4. Combined with NOM and GEN (Type #1)

Since we have already settled the LOC and GEN forms that we will use in the conlang (d– and n– respectively), which might also be the form in the proto language, we can use the rest of the LOC or GEN markers for those markers with combined LOC & ACC (Types #2 & 5) or GEN and ACC (Type #3 & 4) to find out which ACC form to use.

For Type #3 and Type #4 determiner systems where the LOC forms do not resemble the LOC d- forms but still uses LOC forms for definite objects, we will postulate that they have lost their original LOC forms that starts with d-, with the ACC forms expanding to cover the normal LOC functions as well as the traditional ACC function (definite objects of unembedded sentences) but losing the indefinite objects to the GEN forms. The same can be said of Types #5 since the same nominal marker is used for both ACC (indefinite objects) and LOC (definite objects and locations) . In Types #3, #4 and #5, there are two competing consonant onsets, k– and s( or t-). The reasons I favour k rather than s– as onset consonant for ACC are:

  1. Central Philippine languages are the only ones with s– in the OBL, LOC/ACC, or ACC/GEN, with the only exception of Subanon, which is geographically beside Central Philippine languages (Cebuano, Hiligaynon). It should also be noted that sa might have come from saŋ, with Cebuano GEN sa from what is cognate to Hiligaynon, Samarnon, Bisakol GEN saŋ, and Tagalog, Hiligaynon, Cebuano, Samarnon, Bisakol LOC sa from what is cognate to Mansaka LOC saŋ.

    Sa also seems to function as a prefix, since it can be preceeded by ang/an and ng/nin~kan in Tagalog/Bikol respectively.

    Common Noun markers

    Mansaka 3 yaŋ n s
    Tagalog 3 ɥaŋ n sa
    Hiligaynon 3 ɥaŋ s sa
    Cebuano 4 ag sa ug sa
    Subanon 3 ɥog nog sog
  2. The languages with k– in LOC for common and personal markers cover all the example languages except Subanon and Inibaloi. This personal marker LOC is used for human definite objects.

    Common Noun markers

    Kapampangan 4 ɥiŋ n yaŋ k
    Tagbanwa 3 ɥiaing ɥit kat
    Batak 3 tu ɥit kat
    Mamanwa 3 ya na ka
    Maranao 4 su u sa ku
    Inibaloi 5 say ɥi ni (su) ni
    Pangasinan 5 say so na ed

    Personal Noun Markers

    Tagbanwa 3 si ni ki
    Maranao 4 si ɥi ki
    Tagalog 3 si ni kay
    Hiligaynon 3 si ni kay
    Batak 3 si ɥi kay
    Mansaka 3 si ni kay
    Mamanwa 3 si ni kan
    Kapampangan 4 ɥi n k
    Cebuano 4 si ni k
    Subanon 3 si ni ?
    Inibaloi 5 si nen (su) nen
    Pangasinan 5 si nen kinen
  3. For Bilaan which is a Type #1 language and has retained the older LOC marker for common nouns, the  LOC marker for personal nouns starts with k-. It is noteworthy what Reid said:

    Bilaan does not mark any personal noun except a (directional) Locative, which is marked with either /ku/ or /kane/.

    I interpret directional locatives as Accusatives when applied to ACC objects. Ilokano also has LOC personal marker that starts with k– (see above).

We will postpone discussion of Types #2 and 6 ACC forms (ɥ– and s-) until the NOM section.

D. Neutral Form

Reid concluded that there is probably no separate marker for Topics. I would assume that his reason for this are that there are no distinct forms for NEUT:

  1. Types #1 to #4 have forms that function both as NEUT and NOM
  2. Type #6 has forms that function both as NEUT and GEN.
  3. Type #5 has a NEUT form which is derived from the NOM.

However, the only Philippine-type language to have a uniform distinction between the NEUT and NOM is Amis ( ∅- initial for NEUT and k- for NOM) in both the nominal markers and personal pronouns. Reid has mentioned Amis in his work but he only noted its similarity to Ivatan’s determiner system type #6. Malcolm Ross (2006) has identified two reasons below for reconstructing the NEUT forms different from the NOM forms, and with the NEUT starting with ∅- initial (my ɥ– initial):

“The case labels of the reconstructions follow fairly obviously from the data, except for the assignment to NEUT of forms consisting of a vowel only. This assignment is based on two facts. First, many modern languages have a NEUT vs. NOM distinction in pronouns, and it is likely that this distinction occurred in Proto Austronesian. Second, although only one modern Formosan language, Amis, maintains this distinction in its case-marking paradigm, I have assumed that the Amis distinction between *∅- NEUT and *k- NOM reflects a PAn distinction: this helps explain why both forms occur in the NOM paradigms of modern languages. Two languages which no longer reflect the NEUT/ NOM distinction in their case-markers retain it in their personal pronouns, and the NEUT pronouns are marked by reflexes of PAn *i NEUT:PS:S: Pazeh i- and P-Puyuma *i- (see Appendices, §B.1 and§B.11.)”

So we will use the Amis model to have ɥ initial nominal markers in the Neutral form. Referring to the Amis table above, we will notice that all the common noun markers share the same vowel ( –u) and personal markers share the same vowel (-i).  In our conlang, we are going to have the same pattern. Thus the forms for NEUT common noun and personal noun is similar to current Maranao GEN ɥu/ɥi forms. Also note that in Bikol and Samarnon, there is definiteness and distance distinctions, so this will need to be carried here as well.

As a result, the current NOM markers in a lot of languages like “ang”( Tagalog, Sugbuhanon,  Hiligaynon) “an” (Bikol, Samarnon) and “ing”  (Kapampangan) will become NEUT marker here.

E. Nominative Form

What is the form of the Nominal Case marker? We have 3 alternative initial consonants for NOM.

  1. One is Ross’ proposed kwhich we have assigned as the ACC form onset consonant
  2. ɥ– or those that starts with a vowel, as Reid calls them, and we have assigned as the NEUT form onset consonant.
  3. sis supported by the data from personal case markers, where Reid said that *si is a fairly confident reconstruction.

So what are the things I can use as basis for positing a s– initial formative for NOM case”

  1. Data from Formosan languages which I have described above.
  2. Kalinga has si for ACC-LOC common marker which could come from a NOM common marker si. Additionally, Inibaloi and Pangasinan has say as NEUT/TOP marker, which as Reid himself said could be from sa + NOM i, boosting a possible s– initial for NOM case.
  3. In Reid, Binukid Manobo has common noun starting with s-, with only Bangon and Ilianen Manobo with k– initial forms in the NOM case.
  4. That a s– initial NOM common marker can migrate to an ACC is shown by Ivatan ACC su marker.



In Ross (2006), there is no Philippine language in the samples provided but his table indicated a reconstruction by Reid (1978) where the common noun NOM starts with either k– or ɥ– . I will now disregard that and suppose that s– in the initial formative for NOM case..

Filipino Conlang Nominal Markers

The nominal markers of the Filipino conlang will then be:

ɥ- s- n- k- d-

In the second installment of my post, I will outline the other semantic features to be incorporated in the Filipino conlang.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: