Printer Friendly

The morphological realization and production of a nonprototypical morpheme: the Tagalog derivational clitic *.

The distinction between lexemes, proclisis, and prefixation in some languages is not an easy one to make, as certain affixes may exhibit behavior from all areas of the morphological continuum. This paper will show how, when using natural data, we cannot always appeal to the definitional criteria of morphological classification when describing the behavior of morphemes in spoken language. From a corpus of spoken Tagalog, the national Austronesian language of the Philippines, certain prefixes will be introduced to show why such is the case. I will exemplify the possible functions and formal realizations of Philippine-type derivational prefixes and will show that, in at least one language family, preroot derivational morphemes may have quite diverse realizations in their form, and their production: from prefixes to derivational proclitics to unbounded phonological sequences.

The paper will be divided into seven parts. Section 1 provides a general overview of Tagalog typology to equip the reader with general knowledge about the affixes to be detailed in this study; section 2 illustrates the derivational morphology and outlines the argumentation for considering the affixes derivational; section 3 gives an overview of inflectional morphology in the language; sections 4 and 5 furnish evidence for a derivational proclitic analysis of the prefixes under study (section 5 with code-switching data); section 6 supplies production evidence of midword pauses corresponding to morpheme breaks, alluding to a possible lexemic interpretation of the prefixes; and section 7 concludes the analysis, culminating in the discussion of a new morpheme type: the derivational clitic.

1. Philippine prefixes: an outline

Philippine languages, although quite distinct from each other in various ways, all exhibit relatively similar morphological typologies. They are predicate-initial, highly prefixing, and agglutinative. Nominal and verbal morphology is mostly derivational with various idiosyncrasies and lexicalization patterns, while verbal-aspect morphology is inflectional. Sections 2 and 3 outline the derivational and inflectional properties of the prefixes to acquaint non-Philippinists with the system, as a crucial understanding of the general morphological framework is necessary to follow this study.

2. Derivation

There are four important (very productive) systems of derivational affixation in Tagalog, most of which are highly, if not exclusively, prefixing in nature. The morphological behavior that I will describe in this paper transcends all these major systems, so I will outline them briefly as follows: section 2.1, focus system; section 2.2, mode (aptative prefixes); section 2.3, nominalization patterns; section 2.4, causation and direction; and section 2.5, other prefixes.

2.1. The Tagalog focus system: a case of derivational morphology with inflectional aspect choices

A curious and well-studied feature of languages of a Philippine typology is the mandatory encoding of focus on all verbs. Focus refers to the privileged syntactic position of the trigger argument (argument in ang case) whose semantic relationship with the verb is cross-referenced with a focus affix.

The nomenclature for this particular argument has been a controversial topic in Philippine linguistics, the four competing terms being "subject" (Bloomfield 1917; Blake 1925; Lopez 1941; Ramos 1974; Bell 1976), "topic" (Schachter 1977; Foley and Van Valin 1984; Naylor 1995), "absolutive argument" (Gerdts 1988; Payne 1982; Mithun 1994; DeGuzman 1997; Rubino 1997), and "pivot" (Foley and Van Valin 1984: 134ff.; Foley 1998), plus the case-neutral term "trigger" (Cumming 1986; Schachter 1987). I will refer to this argument as the "trigger" argument in order to sidestep the problem of case assignment in Tagalog (subject vs. absolutive argument) and borrow discourse-base terms that do not connote structural interpretations (topic).

The trigger argument in Tagalog is realized by either a pronoun in the absolutive (1) ako (or ang) series case, or a full NP marked with the article ang, or a demonstrative in ang case. The special semantic and syntactic position that the trigger argument shares with the verb can be seen in the following focuses. (2) The Tagalog focus system is relatively impoverished compared to the systems of some sister languages; certain focus markers may represent up to three different focal categories as shown below by the prefix i-:

(1) Actor focus (AF): trigger argument is an ACTOR

a. L{um}i-lipad ang ibon. (3,4)

AF.IMPF {RE}-fly TRIG bird

'The bird is flying.'

b. Mag-bu-bukas ang bago = ng gusali bukas.

AF-FUT-open TRIG new = LNK building tomorrow

'The new building is opening tomorrow.'

(2) Patient focus (PF): trigger argument is a PATIENT

P{in}alo niya ang bata.

spank{RE.GF} 3sERG TRIG child

'He spanked the child.'

(3) Benefactive/instrumental/theme focus (BF/IF/TF): trigger argument is a BENEFACTOR or INSTRUMENT or CONVEYED

a. I-bili mo siya ng kendi.

BF-buy 2sERG 3sABS NTRIG candy

'Buy some candy for him/her' (benefactor trigger).

b. I-punas mo ito-ng basahan sa lamesa.

IF-wipe 2sERG this (TRIG)-LNK rag OBL table

'Wipe the table with this rag' (instrument trigger).

c. I-labas mo na ang pera!

TF-outside 2sERG now TRIG money

'Take the money out now!' (theme trigger).

(4) Locative/directional focus (DF): trigger argument is a LOCATION or argument in whose general DIRECTION an action is performed

Ni-layas-an niya ang kanya-ng pamilya

RE.DF-go.away-DF 3sERG TRIG 3sOBL-LNK family

* 'S/he abandoned (ran away from) his/her family' (directional trigger).

2.2. Aptative mode

As in sister Philippine languages, Tagalog verbs also derive for mode: dynamic (indicative) and aptative (APT). Dynamic mode is unmarked and indicated by the focus affixes exemplified in section 2.1. Aptative mode is indicated by the prefixes ma- (transitive) and ma(ka)- (intransitive) with various combinations. Verbs in the aptative mode refer to actions that are nondeliberate, abilitative, or coincidental.
(5) Verb root Aptative form Dynamic form

 siko masiko sikuhin
 'elbow' 'elbow accidentally' 'to elbow purposely'
 langoy makalangoy lumangoy
 'swim' 'can swim' 'swim'
 tanggap matanggap tanggapin
 'get' 'receive; happen to get' 'to accept; take'

When used with verb stems where control or the lack thereof is not inherent, the aptative form of the verb will specify either lack of control or less energy exerted in the performance of the action.
(6) Verb root Aptative form Dynamic form

 pansin mapansin pansinin
 'notice; heed' 'to notice' 'to pay attention to'
 dinig marinig dinggin
 'hear; listen' 'to hear' 'to listen to'

Compare the use of the two modes with the root pansin 'notice; heed'

(7) Aptative mode:

Ma-buti at na-pansin ang usok na galing sa kisame.

ADJ-good and APT.PF-notice TRIG smoke LNK coming OBL ceiling

'It's good that (they) noticed the smoke coming from the ceiling.'

(8) Dynamic mode:

Huwag mo-ng pansin-in ang sinabi niya.


'Don't pay attention to what he said.'

2.3. Nominalization patterns

A cursory look at Tagalog grammar will reveal that the language has rather regular paradigms for nominalizing verbs. Unlike verbs, these nominalized forms do not inflect for imperfective aspect or focus. (5) Nominalizations (gerunds) are formed off the actor-focus (intransitive) form of the verb and contain an element of the prefixes pag- or paN-. The same idiosyncrasies and lexicalization patterns that exhibit themselves in the actor-focus system are present in the nominalization patterns. These prefixes are responsible for altering lexical class and are undoubtedly derivational prefixes.

The gerunds are formed as follows:
(9) Actor Gerund Examples

 -um- pag- humanap 'to search' [right arrow]
 paghanap 'search; hunt'
 mag- pag- CV maglakbay 'to travel' [right arrow]
 paglalakbay 'traveling; voyage'
 maN- paN- aN- mangaso 'to hunt with dogs' [right arrow]
 pangangaso 'hunting with dogs'
 maka- pagka- makapigil 'to impede; inhibit' [right arrow]
 pagkapigil 'repression'
 ma- pagka- mabuhay 'to live' [right arrow]
 pagkabuhay 'livelihood'
 maki- pakiki- makisama 'to unite, join' [right arrow]
 pakikisama 'companionship; getting
 along with others; society'

For example,

(10) a. Wala ako=ng karanasan sa pag-hawak ng

NEG.EXIST ls = LNK experience OBL NOM-hold NT



'I don't have any experience handling the machine.'

b. Lubus-in mo na ang pag-tulong!

do.completely-PF 2sERG now TRIG NOM-help

'Give all the help you can!'

2.4. Theprefix pa-

The prefix pa- in Tagalog combines with a variety of other affixes to form causative (indirect-action) verbs. Indirect-action verbs designate the permitting or causing of actions to be performed as exemplified by the following verbs:
(11) Direct action

ma-kita 'see'
linis-in 'clean'
ma-tulog 'sleep'

Indirect action

magpakita 'show, cause to see'
ipalinis 'permit to clean'
patulugin 'put to sleep'

(12) a. Bago siya {um}alis, i-p{in}a-kita sa

before 3s leave {RE.AF} THM-CAUS{RE}-see OBL

akin ang aklat.

1sOBL TRIG book

'Before she left, she showed me the book.'

b. Hindi namin kaya = ng mag-pa-aral ng

NEG we.EXCL.ERG able = LNK AF-CAUS-study NT



2.5. Other derivational prefixes (6)

Other common prefixes in Tagalog include ka- (prefix of companionship or concommitance; recent past prefix), paki- (social prefix, prefix of polite request), and paN- (instrumental prefix).
(13) a. ka-: laro 'play'
 ka-laro 'playmate'
 mag-ka-laro 'to be playmates, play with each other'
 alis 'leave'
 kaaalis 'has just left'
 b. paki-: from the verb magluto 'to cook'
 Paki-luto mo nga ito = ng karne.
 SOCIAL-cook please 2sERG this = LNK meat
 'Please cook this meat.'
 c. paN-: from the root linis 'clean'
 'for use in cleaning; to use to clean'
 from the root bahay 'house'
 'household; for use in the house'

2.6. Derivational qualities of the affixes (7)

The affixes to be examined in this study can be argued to be derivational for a number of reasons. The focus affixes and aptative prefixes are verbalizers; they change the word class of the root.
(14) asal 'conduct' mag-asal 'to behave'
 negosyo 'business' magnegosyo 'to deal in business'
 gitara 'guitar' maggitara 'to play the guitar'
 kastila 'Spanish' magkastila 'to speak Spanish'
 sanga 'branch' sumanga 'to branch off'
 tinggal 'store, supply' tinggal 'to store up'
 kilay 'eyebrow' kilayan 'to paint eyebrows
 on someone'
 init 'heat' mainit 'to be hot (hot)'

The affixes also exhibit selection restrictions, a common criterion used to differentiate derivational affixes from inflectional affixes, and affixes from clitics (Bybee 1985; Payne 1985). We can compare the actor-focus prefix mag- with another productive actor-focus verbalizer, the infix -um-. Certain roots are verbalized by mag-, some by -urn-, some by both. Mag- and -um- do not enjoy full generality of employment as verbalizing prefixes; their use is mandated by the lexicon, not syntax: (8)
(15) from the root
 pili 'choose':
 pumili 'to choose' *magpili
 lura 'spit':
 lumura 'to spit' *maglura
 laro 'play':
 *lumaro 'to play' maglaro
 hihip 'blow':
 humihip 'to blow' *maghihip
 kanser 'cancer':
 *kumanser 'to suffer from cancer' magkanser
 luto 'cook':
 * lumuto 'to cook' magluto
 bilis 'speed':
 bumilis 'to speed, go fast' *magbilis

The meanings associated with the mag- verbs and -um- verbs are not always completely predictable, another reason to treat them like derivational affixes. Mag- verbs display semantic idiosyncrasies in certain combinations. The following chart illustrates the nontransparent meanings of verbs that may take both affixes:
(16) from the root
 tae 'excreta':
 tumae 'to defecate' magtae 'to have diarrhea'
 bili 'sell; buy':
 bumili 'to buy' magbili 'to sell'
 laban 'fight':
 lumaban 'to fight' maglaban 'to fight one
 akyat 'climb; go up':
 umakyat 'to climb' mag-akyat 'to bring up'
 alis 'leave; remove';
 umalis 'to leave' mag-alis 'to remove'
 tawa 'laugh';
 tumawa 'to laugh' magtawa 'to laugh
 kamay 'hand':
 kumamay 'to shake magkamay 'to shake hands;
 hand' eat with
 taas 'height':
 tumaas 'to rise' magtaas 'to raise; set up'

Focus affixes also contribute to the valency frame of the verbs, which in turn provide Tagalog speakers with yet another source for deriving lexical items. Notice the idiosyncratic meaning differences between the actor vs. goal (nonactor) forms of the following verbs:
(17) Root Actor focus Goal focus
 (intransitive) (transitive)

 kanser 'cancer' magkanser 'to suffer ikanser 'to donate
 from cancer' something to the
 *kumanser cancer society'
 talo 'win; disagree' magtalo 'to contend; talunin 'to defeat;
 quarrel; argue overcome'
 tubig 'water' manubig 'to urinate' tubigan 'to water

The aptative prefixes also contribute to lexicalization. The change in meaning in verbs employing aptative prefixes is lexical in nature, not grammatical; some aptative frames may be transparent, but others are clearly lexicalized (note words below in italics):
(18) Bare root Aptative verb Nanaptative verb

 lason makalason lumason
 'poison' 'to be able to poison' 'to poison'
 isa makaisa mag-isa
 'one' 'to agree' 'to be alone, do
 makaisa alone'
 'to be able to get one
 of something'
 dumi marumihan dumuni
 'dirt' 'to become soiled, 'to become dirty;
 polluted' defecate'
 hanap mahanap humanap
 'look for' 'to find' 'to seek'
 taranta mataranata tarantahin
 'upset' 'to be confused, 'to confuse, perplex'
 dinig marinig dinggin
 'audible' 'to hear' 'to listen to'
 laman malaman maglaman
 'meat; contents' 'be meaty; fleshy; 'contain; hold; develop
 venereal' flesh'

In summary, it must be emphasized that Tagalog roots cannot take all focus affixes productively. Focus affixes generate lexical items that contribute much more to the verb valence than a syntactic focus shifting. This semantic endowment to the verbality cannot be reflected in the simple morpheme glosses that refer to focus and has thus been poorly understood by linguists, who overgeneralize their use in their theories. The choice of focus is governed by many factors and is indeed affected by idiosyncratic properties of certain verbs that manifest themselves in lexicalization patterns. To illustrate this, consider the root kanser 'cancer' and the various focuses available to it, all responsible for considerable meaning shifts.

(19) Actor focus

mag-ka-kanser 'to suffer from cancer'


Patient focus

k-in-anser (perfective) 'to have infected with cancerous genes'

Theme focus

i-kanser 'to donate something to the cancer society'

It should be clear that derivational affixes in Tagalog, although quite productive, do provide considerable meaning shifts that should be documented on a case-by-case basis.

3. Inflectional verbal morphology

All verbs in Tagalog that derive for focus and mode, as exemplified in section 2.1, inflect for aspect, based on their derived form. Aspectual distinctions are indicated morphologically by portmanteau morphemes that surface according to rather regular paradigms. As seen in Table 1, incomplete aspect is indicated by initial CV reduplication of the stem with the realis morpheme, and future (contemplated) aspect is indicated with CV reduplication without the realis morpheme. The realis morpheme (used with perfective and incomplete action verbs) has been historically realized with the infix -in- but emerges in modern Tagalog in different ways with the actor-focus verbs. It has been lost with -urn- verbs (historically -umin [right arrow] -umm [right arrow] -um-) and retains only n- with mag- verbs.

(20) Examples with the root dating 'come; arrive':


d{um} ating



Gabi na nang dumating ang aming magulang.

night already when arrive.RE.AF TRIG our.EXCL.LNK


'It was already late (night) when our parents arrived.'


Panay ang tanaw ko kapag may dumarating na kotse.

steady TRIG view my COND EXIST come.IMPF.RE LNK


'I kept a steady lookout if there were cars coming.'


Kailan ka ba darating?

when 2s QUES = LNK arrive.FUT.AF

'When will you come?'

4. Proclitic status

Preroot derivational morphemes in Tagalog exhibit qualities that should compel us to classify them as prefixes. However, they also share certain properties characteristic of proclitics. Clitics may be defined as material positioned relative to adjacent syntactic constituents, rather than relative to particular parts of speech (Carstairs 1981). Clitics also exhibit low selectional restrictions with regard to the lexical category of their host (Zwicky and Pullum 1983). Like a proclitic, the prefixes can attach to constructions that are longer than the word. They may also attach to words belonging to distinct word classes. The polylexemic lexical items and phrases to which they may attach are shown below in square brackets (exemplified with the prefix mag-, the most productive of the derivational morphemes, and the affix most widely used in this manner):

(21) a. mag-[boda de plata]

AF-[silver wedding anniversary]

'to celebrate one's silver wedding anniversary.'

b. mag-[tenedor de libro]


'to be an accountant.'

c. Nag-[hubo'-t hubad] siya.

RE.AF-[naked.from.waist.down-and shirtless] s/he

'She undressed completely.'

d. Nag- [inay ko po] ako

RE.AF mother my POLITE 1s

'I was amazed' (lit. 'said wow, my mother!').

e. Nag-[bunton-g-hininga] sila.

RE.AF [heap-LNK-breathed] they

'They sighed.'

It is also apparent in Tagalog compound constructions where the resulting form appears to be created from two separate roots. In Tagalog, objects of mag- verbs are preceded by a form of the noun marker ng [nang].

(22) Nag-kibit siya ng balikat.

RE.AF-jerk s/he NT shoulder

'S/he jerked his/her shoulder.'

However, certain compound constructions exist in which the object noun is actually incorporated into the mag- verb (preceded by the linker na/-ng). Many of these are idiomatic expressions, abundant in the language in this form.

(23) a. Nagkibit-balikat siya.

nag-kibit-balikat siya

RE.AF-jerk-shoulder s/he

'He ignored it' (lit. 'shrugged his shoulders').

(cf. magkibit ng balikat: 'to shrug the shoulders')

b. Nagaagaw-tulog siya.

nag + R-agaw-tulog siya

RE.AF + IMPF-snatch-sleep s/he

'He is falling asleep.'

c. Bago magbukang-liwayway ay dumating sila.

bago mag-buka-ng-liwaywayay d{um}ating sila

before AF-open-LNK-dawn PM arrive-{AF} they

'They arrived before daybreak.'

d. Nag-kulay-rosas.



'It was pink' (lit. 'the color of roses').

e. Magdadalang-tao siya.

mag + R-dala-ng-tao

AF + FUT-bring-LNK-person s/he

'She is pregnant' (lit. 'is bringing a person').

The following idiomatic expressions with mag- listed in Sebastian and Mariano (S and M) (1954) and English (E) (1986) also show this tendency.

(24) a. magbigay-sala



'to blame' (E)

b. mag-asawang kalapati

mag-[asawa-ng kalapati]

AF-spouse-LNK dove

'to be always together' (S and M)

c. magbibig-anghel



'to realize what has been said' (S and M)

d. magdalang-habag



'to sympathize' (S and M)

e. magsilbing-kanin



'to serve (in another's household) with food as payment'

(S and M)

f. magtatalo-sira



'to take back one's promise' (S and M)

g. magwalang-bahala



'to keep silent, be unmindful of' (S and M)

h. nagmumurang-kamatis



'said of a mature individual trying to look young' (S and M)

The preceding idiomatic expressions illustrate that the prefix mag- may combine with polymorphemic compounds. It is with nonidiomatic phrases, usually involving numbers, that the enclitic behavior of mag- is best exhibited. In the following examples, the prefix attaches to full noun phrases consisting of an adjective (number) linker and head noun to form temporal predicates. The following noun phrases are not lexicalized compounds.

(25) a. Mag[dadalawang araw] na silang hiwalay.

mag-R-dalawa-ng araw na sila-ng hiwalay

AF-IMPF-two-LNK day LNK they-LNK separated

'They have been apart for two days (It has been two days ...).'

b. Mag[tatatlong buwan] na siyang buntis.

mag-R-tatlo-ngbuwan na siya-ng buntis

AF-IMPF-three-LNK month LNK she-LNK pregnant

'She has been pregnant for three months.'

In (25c), two coordinated nouns take the prefix mag:-

c. Mag-[Marcus at Marcos] tayo.

AF-[Marcos and Marcos] we.INCL

'Let's support Marcos and Marcos' (Gil 1995: 16).

5. Code-switching evidence for a proclitic analysis

Further evidence of the proclitic behavior of mag- and other Tagalog prefixes can be shown with English borrowings. English code switching and borrowing is frequent in urbanized Tagalog, especially among the younger, educated generation. Words, phrases, and roots are systematically incorporated into Tagalog utterances:

(26) a. I-kanser mo na.

T-cancer 2s now

'Donate it (the shirt) to the Cancer Society.'

b. It's about time na magka-baby na tayo, kaya't

LNK AF-baby now 1pINCL so

hintu-an mo na ang pag-gamit ng pills.

stop-GF 2sE now TRIG NOM-use NT pill

'It's about time we had a baby so stop taking the pill.'

With English code switching we can get a sense of what sorts of constructions may appear in root position. Tagalog speakers have little trouble prefixing complex English constructions, where the polylexemic phrase substitutes for the root.

(27) a. Kung mag + a-[out of town] tayo minsan, mga

if AF + FUT-[] 1pINCL once, PL

dalawa-ng araw ...

two-LNK day

'If we could go out of town sometime, about two days ...'

b. Sa-sabay ka na-ng mag-[Love Boat]?

FUT-same.time 2s now-LNK AF-[Love Boat]

'Are you going to go on the Love Boat at the same time?'

c. Nag-[day off] si Nanay.

RE.AF[] PA mother

'Mother took the day off.'

Other common prefixes may just as easily attach to whole phrases or polylexemic constructions as shown in brackets below, giving the impression that perhaps most, if not all, Tagalog prefixes can be analyzed as proclitics:

(28) a. Nakaka-[pick up] sila ng AIDS!

IMPF.APT-[pick.up] they NT AIDS

'They are picking up AIDS' (referring to Philippine dancers in Japan).

b. Naka-[first base] ka na ba?

APT-[first.base] 2s already QUES

'Did you already make it to first base (with the girl)?'

c. Sa harap mismo ni Olivia, ni-[lips to lips] niya

OBL front right PA Olivia RE.GF-[] 3sERG



'Right in front of Olivia she kissed me on the lips.' (9)

d. Naka-[sexy shorts] pa!

RE.wearing-[sexy.shorts] even

'She's even wearing sexy shorts!'

e. Ma-dali-ng ma-touch, ma-[carried away]

ADJ-quick-LNK ADJ-touch ADJ-[carried.away]

'He is quick to touch, he gets carried away.'

f. Naka-[plastic bag] 'yung pasalubong ninyo.

in-[plastic.bag] that gift 2pERG

'Your gifts are in a plastic bag.'

g. Hiwalay na sa ka-[live in]

separate now OBL COMIT-[]

'He is now separated from his live-in mate.'

Some of the above examples deal with polylexemic roots that can be argued to be idiomatic collocations. Such is not the case with the following roots, examples of nonidiomatic English phrases created by speakers of Tagalog.

(29) a. Hindi na ma-appreciate ng lakai ang sobra-ng

NEG now INVOL-appreciate NT boy TRIG overly-LNK

pa-[Maria clara image]

ATTRIBUTIVE-[Maria Clara image]

'Boys no longer appreciate too much of a Maria Clara (virginal) image.'

b. Saan kaya galing ang mag-ama-ng

where WONDER come.from TRIG RECIP-father-LNK

ito, Mare, at pareho-ng naka-[dress to die]?

this Kumare RESULT same-LNK STATIVE-[dressed to kill]

'Where did this father and son come from, kumare (term of intimate address between co-godmothers), since they are both dressed to kill?' (from nakapamburol 'attired in dead man's clothes: dressed in one's best clothing').

6. Tagalog prefixes as words

To make matters more interesting, and to show how morphemes can share characteristics of all categories of the morphological continuum, I will now provide evidence that the Tagalog prefixes can be considered phonological words, the criteria being the ability to be bounded on both ends by a pause.

Although the Tagalog prefixes exhibit classic affixal behavior, including selectional restrictions and morphological and semantic idiosyncrasies when they attach to certain hosts, one can appeal to production phenomena to see if they can be possibly considered bona fide words. In Tagalog, and many other Philippine languages, affixes may often be uttered separately from their hosts. The data below show examples of this seemingly disfluent production of polymorphemic words where there is a pause between the prefix and root.

(30) a. Natutong mag-., reklamo,

na-tuto-ng AF-.. complain

'(He) learned how to complain.'

b. Iyon ang pinaka- ... intimate sa lahat ng bagay.

that TRIG SUPER- intimate OBL all NT thing

'That is the most ... intimate of all things.'

c. Kapag alas dose ng gabi ang mga tao ay

when at twelve GEN night TRIG PL people PM

nag-., nag-., papa- nag

RE.AF- RE.AF- causative prefix reduplicated RE.AF

papa: pa-putok ng labentador,

CAUS-CAUS CAUS-burnt NT fireworks

'At twelve o'clock midnight the people explode fireworks.'

As pauses are not supposed to occur word-internally, it may be argued that the productive prefixes are actually words that have not yet completed their cycle of development into morphological affixhood. One may likewise argue that productive prefixes may have more cognitive salience than other types of affixes, and their production in discourse may at times be disfluent, reflecting an intermediate status of prefixes in the lexicon between word and affix (Rubino 1996).

Native speakers of Tagalog may have trouble classifying prefixes as words or affixes in their writing. Some speakers choose to write the prefixes as separate words, or followed by a hyphen, (10) indicating the psychological reality of at least a morpheme break. This goes against the established Tagalog orthographic tradition.

7. Conclusion

This paper has shown that our understanding of morphological theory can be enhanced considerably by the study of natural data. As the Tagalog prefixes illustrate, there is no fine line that separates the various morpheme types; certain preroot morphemes may exhibit behavior from all points on the morphological spectrum.

In Tagalog and other Philippine-type languages, derivational prefixes may behave like bona-fide affixes, clitics, and even phonological words (segments bounded by pauses). The prefixes detailed in this study exhibit behavior reflective of derivational affixes through their use in lexicalizing new words and also because their overall use is far from general.

It is therefore necessary to provide a prototype analysis of the morphological continuum when exploring what can be considered an affix, a clitic, or an independent word. If speakers of languages are able to utilize their morphology to frame polylexemic constituents, perhaps we must not continue to view the "word" as our primitive level of analysis when defining morpheme types in terms of the "root." We should regard derivation, at least in Tagalog, as a means to elaborate concepts morphologically that can be expressed with various forms of complexity.

If we maintain that derivational affixes, especially those that make a considerable contribution to the semantics of the overall construction, have a tighter morphological and semantic bond than inflectional ones, it is fascinating when productive derivational affixes exhibit syntactic or phonological behavior reflective of clitics or independent words. With this in mind, we should accept that in the historical development of a derivational affix, combinatory rules and semantic idiosyncrasies in lexicalization do not entail morphophonological fusion or form. Clitics, at least in Philippine languages, can be produced separated from their hosts and can be derivational in nature, with the power to create new lexical items.

Australian National University

Received 19 August 1997

Revised version received

2 June 1998


Bell, Sarah J. (1976). Cebuano subjects in two frameworks. Unpublished Ph.D. dissertation, University of California, San Diego.

Blake, Frank R. (1925). A Grammar of the Tagalog Language. New Haven: American Oriental Society.

Bloomfield, Leonard (1917). Tagalog Texts with Grammatical Analysis. Studies in Language and Literature 3. Urbana: University of Illinois.

Bybee, J. (1985). Morphology: A Study of the Relation Between Meaning and Form. Typological Studies in Language 9. Amsterdam: Benjamins.

Carstairs, Andrew (1981). Notes on Affixes, Clitics, and Paradigms. Bloomington: Indiana University Linguistics Club.

Cumming, Susanna (1986). Word order change in Malay. In FOCAL: Papers from the Fourth International Conference on Austronesian Linguistics, Paul Geraghty, Lois Carrington, and Stephen A. Surm (eds.), 97-111. Canberra: Australian National University.

DeGuzman, Videa P. (1997). Verbal affixes in Tagalog: inflection or derivation? In Proceedings of the Seventh International Conference on Austronesian Linguistics, Cecilia Ode and Wim Stokhof (eds.). Amsterdam: Rodopi.

English, Leo James (1986). Tagalog-English Dictionary. Quezon City: Kalayaan.

Foley, William A. (1998). Symmetrical voice systems and precategoriality in Philippine languages. Unpublished manuscript, University of Sydney.

--; and van Valin, Robert D. (1984). Functional Syntax and Universal Grammar. Cambridge: Cambridge University Press.

Gerdts, Donna B. (1988). Antipassives and causatives in Ilokano: evidence for an ergative analysis. In Studies in Austronesian Linguistics, Richard McGinn (ed.). Athens: Ohio University Press.

Gil, David (1995). Parts of speech in Tagalog. In Papers from the Third Annual Meeting of the Southeast Asian Linguistics Society, M. Alves (ed.), 67-90. Tempe: Arizona State University.

Himmelmann, Nikolaus P. (1991). The Philippine Challenge to Universal Grammar. Institut fur sprachwissenschaft universitat zu Koln, Arbeitspapier Nr. 15 (Neue Folge). Cologne: University of Cologne.

Kroeger, Paul R. (1991). Phrase structure and grammatical relations in Tagalog. Unpublished Ph.D. dissertation, Stanford University.

Llamzon, Teodora A. (1976). Modern Tagalog: A Functional-Structural Description. The Hague: Mouton.

Lopez, Cecilio (1941). A Manual of the Philippine National Language. Manila: Bureau of Printing.

McFarland, Curtis D. (1976). A Provisional Classification of Tagalog Verbs. Tokyo: Institute for the Study of Languages and Cultures of Asia and Africa.

McGinn, Richard (ed.) (1988). Studies in Austronesian Linguistics. Athens: Ohio University Press.

Mithun, Marianne (1994). The implications of ergativity for a Philippine voice system. In Voice: Form and Function, Barbara Fox and Paul Hopper (eds.), 247-277. Amsterdam: Benjamins.

Naylor, Paz Buenaventura (1995). Subject, topic, and Tagalog syntax. In Subject, Voice, and Ergativity, Selected Essays, David C. Bennett, Theodora Bynon, and B. George Hewitt (eds.). London: School of Oriental and African Studies.

Payne, Doris L. (1985). Inflection and derivation: is there a difference? In Proceedings of the First Annual Meeting of the Pacific Linguistics Conference, Scott De Lancey and Russell S. Tomlin (eds.). Eugene: University of Oregon.

Payne, Thomas E. (1982). Role and reference related subject properties and ergativity in Yup'ik Eskimo and Tagalog. Studies in Language 6(1), 75-106.

Ramos, Teresita V. (1971). Tagalog Structures. Honolulu: University of Hawaii Press.

--(1974). The Case System of Tagalog Verbs. Canberra: Pacific Linguistics.

Rubino, Carl (1996). Morphological integrity in Ilocano. Studies in Language 20(3), 333-366.

--(1997). A reference grammar of Ilocano. Unpublished Ph.D. dissertation, University of California, Santa Barbara.

--(1998). Tagalog (Pilipino) Standard Dictionary. New York: Hippocrene.

Schachter, Paul (1977). Reference-related and role-related properties of subjects. In Grammatical Relations, P. Cole and J. Saddock (eds.), 279-306. Syntax and Semantics 8. New York: Academic Press.

--(1987). Tagalog: In The World's Major Languages, Bernard Comrie (ed.), 936-958. London: Croom Helm.

--; and Otanes, Fe T. (1972). Tagalog Reference Grammar. Berkeley: University of California Press.

Sebastian, Federico B.; and Mariano, Antonio D. G. (1954). Idiomatic Expressions in the Tagalog Language. Manila: Committee on the National Language Textbook Preparation.

Zwicky, Arnold M. (1985). Clitics and particles. Language 61 (2), 283-305.

--; and Pullum, Geoffrey K. (1983). Cliticization vs. inflection: English n't. Language 59, 502-513.


* I would like to thank my Tagalog students for allowing me to use some of their data and the following people who have offered valuable comments on a previous draft of this paper: Nikolaus Himmelmann, Sasha Aikhenvald, Randy La Polla, Paul Kroeger, David Gil, the McFarlands, and Emmanuel Esguerra. This paper was first presented at the 1998 Annual Meeting of the South East Asian Linguistics Society in Kuala Lumpur, Malaysia. It was also presented at ALS-98 (Brisbane) by Claire Bowern, to whom I am extremely grateful. Correspondence address: Department of Linguistics, Australian National University, Canberra ACT 0200, Australia. E-mail:

(1.) Pronouns and personal noun articles in Tagalog function on an ergative/absolutive basis. Common noun articles, however, do not. They encode trigger (ang) vs. nontrigger (ng) in the core arguments, vs. oblique (sa). It is for this reason that I gloss pronouns and personal noun markers ERG, ABS, OBL, and common nouns TRIG, NTRIG, OBL. This system should not be seen as prototypical for all Philippine languages.

(2.) It is beyond the scope of this paper to enumerate the number of focus distinctions in Tagalog. Traditional Tagalog grammar recognizes four (one "active" and three "passive") (Bloomfield 1917: 154). Ramos (1971: 21-23) claims five basic focus formations, McFarland (1976: 16-24) counts seven, and Schachter and Otanes (1972: 344) propose 11, to name a few. See Himmelmann (1991: 27) for further details.

(3.) Lexical and morphological glossing in this paper are greatly simplified for the sake of clarity in the interlinearizing. For those readers not familiar with Tagalog who would like to know more about the morphology and typology, I urge you to consult Schachter and Otanes (1972). A very comprehensive Tagalog dictionary is English (1986). The glosses used are as follows (in alphabetical order): ADJ adjective; AF actor focus; APT aptative (involuntary--abilitative--coincidental mode); CAUS causative; COMIT comitative; COND conditional; EXCL exclusive; EXIST existential; FUT future; GEN genitive; GF goal focus; IMPER imperative; IMPF imperfective; INCL inclusive; INVOL involuntary; LNK linker; NOM nominalizer; NT(RIG) nontrigger (core) noun marker; OBL oblique; PF perfective; PM predictive marker (ay 'inversion'); QUES interrogative; RE realis; RECIP reciprocal; SUPER superlative; TRIG trigger article.

(4.) Because of a long tradition of literacy in Tagalog, I choose to abide by the conventions of popular Tagalog orthography. Vowel-initial words are articulated with an initial glottal stop, which is not represented in the orthography. Glottal catch word-medially is represented by a hyphen. Word-finally, the contrastive glottal catch is usually not orthographically represented, but I adhere to the native system of final glottal-stop representation whereby word-final glottal stop is represented by the circumflex accent (^) on the last vowel if the word is stressed on the last syllable, or the grave accent (') on the last vowel of words with penultimate stress.

(5.) The formal stages of nominalization patterns are not uniform throughout the languages of the Philippines. Ilocano, for instance, has gerunds that do inflect for imperfective aspect (Rubino 1997).

(6.) This study only provides an overview of the common prefixes. For a more complete list of Tagalog morphemes, please see Schachter and Otanes (1972) or the Grammatical Introduction to Rubino (1998).

(7.) Some have asserted that focus affixes are inflectional (Llamzon 1976; McFarland 1976). Certainly, the choice between an actor-focus verb and a goal-focus verb can be governed by the syntax (relative clause restraints) and/or discourse factors (salience/identifiability of patient argument). However, I will argue that the actor-focus and goal-focus affixes are still derivational because the choice that exists between the different affixes of either the actor or the goal focus is never completely predictable, as illustrated in this section. Verbs in Tagalog are lexicalized in many ways with their focus affixes. Competent speakers of Tagalog must be aware of the lexicalization patterns and how they apply to each verb root.

(8.) The actor-focus affixes mag- and -um- may also appear together with certain roots, indicating layers of morphological development. They are not inflectional affixes in paradigmatic distribution: i.e. maghumiyaw 'to shout at the top of one's voice', mag-umiyak 'to cry at the top of one's voice', mag-umunat 'to stretch oneself to the limit.'

(9.) This is actually a case of the infix -in-, normally inserted after the first consonant of a root, metathesizing to ni- before /1/.

(10.) A hyphen in Tagalog orthography represents a word-medial glottal stop. When the prefix mag- attaches to vowel-initial roots, the glottal stop is actually pronounced, as Tagalog words must have an onset. However, certain Tagalogs continue to write the hyphen with consonant-initial roots. I have similar data of this orthographic tendency in other Philippine languages, i.e. Ilocano, Pangasinan, Chabacano, Hiligaynon, and Adasen, to name a few.
Table 1. The Tagalog aspectual system

Affix Example Complete Incomplete Future

ACTOR magtanong nag- nag- CV mag- CV
mag- 'to ask' nagtanong nagtatanong magtatanong
-um- tumanggi -um- C -um- V CV
 'to refuse' tumanggi tumatanggi tatanggi
GOAL tanawin -in- C -in- V CV -(h)in
-(h)in 'to look at' tinanaw tinatanaw tatanawin
-(h)an iyakan -in- -(h)an C -in- V -(h)an CV -(h) an
 'to cry about' iniyakan iniyiakan iiyakan
i- ituloy i- -in- i- C-in-V i- CV
 'to continue' itinuloy itinutuloy itutuloy
APTATIVE makatulog naka- naka- CV maka- CV
maka- 'can sleep' nakatulog nakakatulog makakatulog
ma- matulog na- na- CV ma- CV
 'sleep' natulog natutulog matutulog
COPYRIGHT 1998 Walter de Gruyter GmbH & Co. KG
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1998 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Short note
Author:Rubino, Carl
Publication:Linguistics: an interdisciplinary journal of the language sciences
Article Type:Report
Geographic Code:9PHIL
Date:Dec 1, 1998
Previous Article:The stability of UG principles in second-language acquisition: evidence from Japanese *.
Next Article:"Two's company, more is a crowd": the linguistic encoding of multiple-participant events *.

Related Articles
Conservation of grammatical knowledge: on the acquisition of possessive noun phrases by Turkish and Moroccan learners of Dutch(*).
Sign-oriented vs. form-oriented linguistics and word-formation. (1).
Inflectional and/versus derivational morphology: clear-cut types or continua?
No "double plurals" in Dominican Spanish: an optimality-theoretic account *.
Turkish suspended affixation *.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters