kělok curve, arc, tělok bay, jělok deep-curved (of a boat), kělun (suffix u) spirals (of smoke).[1]
§ 3. What has made it harder to solve the functions of prefixes even in the restricted Indonesian group is that a prefix and an infix frequently combine to form a compound prefix[2] or ‘prefix of the second degree’ giving rise to a new and crystallized formative. The functions of the obsolete simple separate prefixes and infixes, it is then supposed, are obliterated.
In Malay we get
(a) m + nasal infix, § 44, especially II.
p + nasal infix, § 54.
(b) m + r that is to be seen as a relic in certain nouns
and adjectives, especially plant-names, and for example in the rare variants mĕrawan and mĕrapi where bĕrawan and bĕrapi are the ordinary modern forms.
b + r,[3] § 49, which is said by Kern to be a fusion of adjectival měr and verbal b.
t + r, § 52.
p + r, §§ 54-56.
- ↑ In addition to the ‘simple’ affixes given above, there are a few others which occur in one and another of the Indonesian languages, but which in Malay, at any rate, need not trouble the grammarian. Two may be mentioned. In Fiji there is d-, Javanese j-, Malay j- or ch-, examples of which Kern detected in jěbul (Malay chabul) rape from bul a hole and jebur (Malay chěbur) plunge into water from the onomatopoeic bur. Relics of such a formative are clear in such doublets as tengok, jengok see; pijak, jijak tread; kangkang, jangkang a-straddle; abu, jabu (cp. děbu, lěbu) dust; lotong, jĕlotong monkey; alit, palit, chalit smear; bělah split, chělah crevice; těgang, chěkang laut; kebek, sebek, chebek awry. s which in a few Bugis words betokens reciprocity and reflexive action may be seen in the Malay kepit, sepit pinch (one thing by another); kilau, silau flash (flash after flash).
- ↑ Real ‘compound prefixes’ are those given in § 43 (II).
- ↑ From ajar are formed bělajar, pēlajaran; l taking the place of r. [With regard to the theory that the -ng and -r terminations of such prefixes as mĕng- and bĕr-, &c., are mere phonetic links, it is to be noted that (1) it is not supported by the results of the comparative study of the Indonesian languages; (2) it is in conflict with the data of the Kota Kapur inscription (probably of the seventh century A.D.), which contains such forms as mangujāri to speak with; marjjahāti to do harm to; parsumpahan curse, and the like. (See Part 67 of the Bijdragen tot de Taal-, Land- en Volkenkunde van Nederlandsch-Indië.) It is more probable that ng and r were originally separate formatives.—C.O.B.]
