Morphologie - Lexique
Ch. Bauer
Humboldt University, Berlin
Affixes, syncretism, and residual forms in Middle Mon
Attempts at reconstructing the morphology of proto-Austroasiatic (AA) and proto-Mon-Khmer (PMK) have been rare since Schmidt (1906), as have been in-depth studies for individual languages (Jenner and Pou (1980-81), Jacob, passim).
Zide and Anderson (2001) identified connections between Munda verb morphology and Mon-Khmer; for PMK Bauer (passim) proposed the following affixes to be reconstructed:
Verbs:
*s- irrealis
*p- causative
*-N- frequentative
*-m- attributive (?)
*-r- reciprocal
Nouns:
-n- instrumental
-m- agentive
-p- (?) > Mon -w- > mod. Khmer -ɓ- [pace Shorto]
A consolidated epigraphic corpus of Mon enables us now to trace the development of affixes within a single language.
Texts from the Middle Mon period (14th to 18th centuries) lend themselves to such a task as it is during this time that Mon underwent the most fundamental phonological changes which had a bearing on its morphological system (mediocluster reduction, initial cluster reduction, devoicing of initials).
These complex changes led, among others, to syncretistic phenomena.
A deeper historical understanding of Mon morphology providse diagnostic tools for comparative MK work.
Zide, Norman H., Anderson, Gregory D.S.,The Proto-Munda verb system and some connections with Mon-Khmer, in: The Yearbook of South Asian Languages and Linguistics 2001, New Delhi: Sage, 517-540.
Stephanie Farmer
UC Berkeley
On the Possibility of Metathesis as a Source for Infixation in Khmer
Modern Khmer is thought to have two types of infix: a syllabifying /-Vm(n)-/ infix and a set of single-consonant non-syllabifying infixes, the most common of which is /-ɓ-/. Haiman (1998) suggests that the /-Vm(n)-/ infix is the result of a “perceived gap in structure” created when Khmer’s iambic rhythm led to the gradual degradation of initial syllables. On the path from Old Khmer to Modern Khmer (and perhaps much earlier than this as well), many disyllabic words became “sesquisyllables” (Matisoff 1973) or “subdisyllables” (Jenner 1969)—that is, words with one and a half beats due to their unusual initial (i.e. stop plus stop) consonant clusters. These clusters provided a “slot” for the /-Vm(n)-/ infix. The infix /-ɓ-/ is just as ancient and perhaps more mysterious in that it is rarer (both in Khmer and in Mon-Khmer languages in general), is non-syllabifying, and does not seem to fill a perceived structural gap as does its /-Vm(n)-/ counterpart by Haiman’s account. I propose that the /-ɓ-/ infix arose via metathesis between prefix /p-/ and the initial consonant of the root (almost always a liquid /r/ or /l/). While this type of liquid metathesis is crosslinguistically unusual because it results in articulatorily complex clusters, it was desirable in Khmer because it preserved the iambic rhythm of the derived form.
The proposed sequence of events is below:
1. Liquid-initial roots were prefixed with /prV-/, the only CrV- prefix in Old Khmer according to Jacob (1976).
2. /prV-/ alternated with /pV-/ in Old Khmer (as happens in Modern Khmer), at first in careful versus rapid speech, but later as acceptable allomorphs.
3. /p-/ and /r/ metathesized to form /rp/, a cluster that ensured a sesquisyllabic pronunciation of the derived word. Perhaps because Khmer has another type of infix that is inserted after the first consonant of the root, /-p-/ was analyzed as an infix.
4. /-p-/ became /-ɓ-/.
Some examples of words infixed with /-ɓ-/ are below:
lɯən ‘quick’ à lɓɯən ‘speed’
rəŋ ‘hard’ à rɓəŋ ‘hard’
le:ŋ ‘play’ à lɓaeŋ ‘game’
Lewitz (1978: 7) notes in her data “a preponderance of forms” with initial liquids. I believe that this in not a coincidence. While Khmer has few restrictions on consonant clusters, /r/ and /l/ are the only consonants that may follow an initial stop without creating a “half-syllable” (i.e. without necessitating the epenthesis of a short vowel). Because of this, liquid-initial roots with single-stop prefixes become confusable with monosyllabic, monomorphemic words of identical phonetic structure. Metathesis preserves the sesquisyllabic structure that nearly all other derived Khmer words have. As David Thomas suggests in response to Jacob (1976):
The possibility presents itself that in these cases it is not so much the phonological shape of the affix or presyllable that is important. Perhaps, rather, it is simply the presence of some consonantal affix or presyllable, of whatever shape, that carries the signal to the hearer, contrasting merely with the absence of any prefix or presyllable.
Metathesis serves to preserve this signal in an environment where it is threatened.
In Saveros Pou and Philip Jenner’s 1982 Lexicon of Khmer Morphology, the authors cite one word as having arisen due to metathesis. This is rpuj ‘one after the other en masse’ which is identical in meaning to pruj. Both have the root ruj meaning ‘fly’(N). Pou and Jenner label this a “false derivative via metathesis” and therefore do not consider it a case of genuine infixation.
In addition to this proposed example in Saveros Pou and Philip Jenner’s Lexicon of Khmer Morphology, I have found several instances in Modern Khmer of words that seem to have metathesized:
sra:j ‘to untie’ à rsa:j ‘to get loose’
sra:k ‘diminished, lessened’ à rsa:k ‘to have faded’
sraəp ‘(to be) aroused momentarily à rsaəp ‘(to be) ticklish’
without apparent cause’
krɯŋ ~ rkɯŋ ‘pockmarked, pimpled’
krə:m ‘rough as a surface’ à rkə:m ‘in a scuttling manner’
cruəl ‘to panic’ à rcuəl ‘(to be) in a state of panic’
cre:c ‘chirping sound’ à rce:c ‘noise made by a bird or young chicken’
These words, however, do not have recognizable /r/-initial roots, and therefore cannot properly be called infixed. The fact that they have nevertheless undergone metathesis to become sesquisyllabic suggests that there is a general tendency in Khmer to “bulk up” words in either in anticipation of their being ground down or simply because more structure implies more meaning.
Khmer appears to be engaged in a tug-of-war between affixation and monosyllabification; this tug-of-war most optimally results in a sesquisyllable—the most nearly monosyllabic word that is perceptibly dimorphemic. Stop-liquid à liquid-stop metathesis is articulatorily undesirable, but it succeeds in producing a recognizably dimorphemic derivative.
References
Coedes, G. 1937-66. Les Inscriptions du Cambodge, 8 vols. Hanoi and Paris: EFEO.
Haiman, J. 1998. Possible origins of infixation in Khmer. Studies in Language 22:597-617.
Headley, Robert K., Kylin Chhor, Lam Kheng Lim, Lim Hak Kheang, Chen Chun. 1977.
Cambodian-English Dictionary. Washington, D.C.: Catholic University of America.
Jacob, Judith M. 1976. "Affixation in Middle Khmer with Old and Modern Comparisons." In
Philip N. Jenner, Laurence C. Thompson and S. Sarotsa (eds.), Austroasiatic studies, part 1: 591-624. Oceanic Linguistics special publication 13, Honolulu: University of Hawaii Press.
Jenner, Philip N. 1969. Affixation in Modern Khmer. University of Hawaii dissertation.
Jenner, Philip N., Laurence C. Thompson and S. Sarotsa (eds.). 1976. Austroasiatic studies.
Oceanic Linguistics Special Publication 13. Honolulu: University of Hawaii Press.
Jenner, Philip N. and Saveros Pou. 1982. “A Lexicon of Khmer Morphology.” Mon-Khmer
Studies IX-X.New York: University of Hawaii Press.
Matisoff, James. 1973. “Tonogenesis in Southeast Asia.” In Larry M. Hyman (ed.) Consonant
Types and Tone 71-95. Southern California Occasional Papers in Linguistics, No. 1. Los
Angeles: UCLA.
Pou, Saveros. 1976. “The Infix /b/ in Khmer.” In Philip N. Jenner, Laurence C. Thompson
and Stanley Starsta (eds.), Austroasiatic studies, part 2, p. 1111. Oceanic Linguistics Special Publication, 13. Honolulu: University of Hawaii Press.
Thomas, David D. 1976. "Comment re ‘Affixation in Middle Khmer,’ by Judith M. Jacob." In
Philip N. Jenner, Laurence C. Thompson and Stanley Starsta (eds.), Austroasiatic studies, part 1: 1111. Oceanic Linguistics Special Publication 13. Honolulu: University of Hawaii Press.
Yu, Alan C. L. 2007. A Natural History of Infixation. New York: Oxford University Press.
Theraphan Luangthongkum
Chulalongkorn University, Thailand
Proto-Karen Worldviews as Reflected in the Use of Nominal Classifiers in Six Modern Karenic Languages, Sgaw, Pwo, Pa-o, Kayah, Kayan and Kayaw*
In the modern Karenic languages, Sgaw, Pwo, Pa-o, Kayah, Kayan and Kayaw, there are two major types of noun classifier, i.e. the general or basic type and the specific one. The former occurs more frequently since its use is less restricted or, in other words, general nominal classifiers are used to classify many more things than specific classifiers.
Karen speaking people make a distinction between ‘human beings’ and ‘nonhuman beings’. There is only one classifier for nouns belonging to the human category, e.g. /Äa#/, /Äa$/ and /prā/ in Sgaw, Pwo and Kayan, respectively. The nonhuman category comprises animals (animate) and things (inanimate). Animal species (except mammals) and inanimate objects are classified with the same set of classifiers depending primarily upon their shape; for example, in Kayan, the classifier /bɔˊ/ which has the semantic feature [+length] is used to classify snakes, pangolins, crocodiles,
eels, millipedes, earthworms, chameleons, as well as needles, candles, pillars, cigarettes, incense sticks, pestles, bangles, roads and so forth.
To reconstruct Proto-Karen noun classifiers with a focus on the general type and Proto-Karen worldviews as reflected in the use of classifiers, a three-step data collecting method was devised: an impromptu interview, an in-depth interview after preparing working hypotheses and a final interview based on the devised thematic word list. The data used for the analysis is fresh and solely collected by the author. To compare and conclude the research findings, the comparative method was applied.
*This research report is part of the research project on “Karen Phonetics” sponsored by the Thailand
Research Fund (TRF) from 2009-2012.
| Fichier attaché | Taille |
|---|---|
| Abstracts_Morpho_lex.doc | 64 Ko |
»