t   

tadpole

take, bring

take with fingers

take out

Dempwolff (1938) used this proposed etymology to posit ‘Uraustronesisch’ *luRut ‘to take out’ (herausnehmen), but his comparison appears fanciful.

take with hand

tall

Through a number of arbitrary morpheme cuts, proposed to isolate *anzaŋ as a doublet of *panzaŋ 'long, tall'. However, the forms cited here, together with Kelabit kadaŋ 'long', Lampung tijjaŋ 'long' suggest a collection of non-cognate forms related by a common monosyllabic 'root' (Blust 1988).

taro

The resemblance of Kowiai/Koiwai et to the Micronesian forms cited here is attributed to chance.

taro

taro sp.

taut:   tight, taut

tear, rip

tear:   split tear

The Oceanic members of this comparison may be related, but until better evidence is available the resemblance between all of these forms is treated as a product of convergence.

tear:   to tear

Dempwolff (1938) reconstructed ‘Uraustronesisch’ *tes ‘to tear’.

teeth:   show the teeth

tell:   say, tell

tell

temporary aid

Dempwolff (1938) reconstructed *simbat ‘temporary aid’, but there is little evidence for this form apart from the two languages that he used as support, and the first vowel of the Toba Batak form is irregular. Although borrowing from Javanese into Toba Batak without passing through the medium of Malay is virtually unknown, it seems best to treat this as a loan distribution pending the collection of a more convincing cognate set.

ten, in counting certain objects

tender, weak

Malay lemuk, cited by Dempwolff, does not appear in any source I have been able to consult. The remaining part of this comparison is best considered unconnected. Dempwolff (1938) reconstructed Uraustronesisch *lemuk ‘weak’.

tense:   past tense marker

Apparently reconstructible for PCEP. The Paiwan and Amis forms disagree in pointing to *Si and *i respectively. I take the resemblance of PCEP *i to all other forms cited here as a product of chance.

term of address for females

term of address for girls

textile pattern

that:   provided that, in spite of

that:   so, so that

that

The KALL and Manobo (Dibabawon) forms reflect earlier *suyaq. The similarity of Thao huya to these words is attributed to chance.

that

that, there

that, this

Chance. These forms may each contain a reflex of *ia '3sg.'.

theft

Chance. Dempwolff (1934-38) cites this comparison under *ikut 'theft', but I have been unable to find the Tagalog form, and for Javanese I find only ikut 'take away'.

that, there

thirsty

The Iban and Malay forms are cognate; the Balinese and Sasak forms probably are Malay loans, and the similarity of Puyuma Hau to all of these is a product of chance.

that, this

Chance. These forms may each contain a reflex of *ia '3sg.'.

this, proximal deictic

throat:   neck, throat

Chance. Although LAKA bolo- and Kwaio ono could reflect POc *ciono-, the Kwaio form appears to reflect *tonom 'to swallow' (cf. onom-anal 'throat, necḱ, onom-ia 'swallow').

throaty sound

Tae' iʔduk and Manggarai iruk/ may contain a variant of the root *-dek 'hiccough, sob'. Whether this turns out to be the case or not, the similarity of these forms is best attributed to convergence.

throng:   crowd, throng

throw

throw down

thump:   hit, thump

Chance (based on a common root *-bek 'sound of breaking, etc.').

tie a knot

tight, taut

till the soil

Based on this comparison Dempwolff (1938) positted *saka ‘to till the soil’. However, this reconstruction has multiple problems, and the similarity of the forms compared now seems best attributed to chance. English (1986) gives Tagalog sáka as a borrowing of Spanish saca ‘taking out, extraction; exportation’, which seems equally questionable, but none of the other meanings appear to be connected in any obvious way.

tin

tip:   point, tip

Dempwolff proposed *zuru ‘point, tip’ (dbl. *duRu ‘corner’), and included such semantically divergent forms as Malay juru ‘trained worker (at some occupation other than a handicraft)’ and Javanese juru ‘person who performs a certain [skilled] job’, under the general gloss ‘master’ (also cf. Old Javanese juru ‘head, leader, chief (of a division, military or administrative; of a group or trade); tradesman, trained worker’). The entire collection seems extremely forced, and is best abandoned as ill-conceived.

together:   close together

together:   dense, close together

topsy-turvy

torture

Chance. This comparison is an intriguing example of how a spurious etymology can be manufactured and perpetuated by competent comparative linguists despite a telling lack of credentials. Dempwolff (1938) glossed the Tagalog form 'tortured' and the Tongan form 'torture'. Although he rejected the phonologically irregular Samoan form, Dyen (1953) otherwise accepted Dempwolff's comparison unchanged, including his glosses for Tagalog inís and Tongan ino. Neither Dempwolff's own sources (Laktaw 1914 for Tagalog, Colomb 1890 for Tongan) nor reliable alternatives of more recent date (Panganiban 1973 for Tagalog, Churchward 1959 for Tongan), however, support the discrepant glosses cited in Dempwolff (1934-38).

toward:   harbor bad feelings toward

trap:   animal trap

trap

Dempwolff (1938) reconstructed *seru ‘trap’. However, Tagalog síloɁ reflects PPh *siluq ‘noose, snare; net, and is therefore phonologically incompatible with the other forms in this comparison, which are also phonologically incompatible with one another.

tray

Casiguran Dumagat bílao probably is a Tagalog loan. The resemblance of the Isneg word to these forms is attributed to chance.

tree:   palm tree sp.

tree sp.

Chance. For the etymology of the Batak forms cf. *qabaŋ qabaŋ 'float'.

tree sp.

tree sp.

tree sp.

Chance, Ngadha vala is assumed to be cognate with Manggarai balak 'tree sp.: Scindapsis, Rhaphidophora'.

tree sp.

tree sp.

tree sp.

tree sp.

tree sp.

The forms in Gedaged and Gitua are cognate. The similarity of these words to those in LAKA and Lau is attributed to chance.

tree sp.

tree sp.

tree sp.

The Sulawesian forms from Bare'e to Wolio are assumed to form one geographically and genetically restricted cognate set. The resemblance of the members of this set to the other forms cited here is attributed to chance.

tree sp.

tree sp.

tree sp.

It is tempting to consider Hanunóo buybúy a loan from one of the languages of northern Mindoro, and so recognize a PPh *buRbuR 'silk cotton tree: Ceiba pentandra'. However, this hypothesis fails to account for the appearance of a palatal glide in the Tagalog and Aklanon words without an implausible extension of the hypothesis to include these languages. The comparison thus appears to be a product of chance.

tree sp.

tree sp.

tree sp.

tree sp.

tree sp.

tree sp.

tree sp.

tree sp.

tree sp.: Barringtonia

tree sp.: Erythrina indica

Gedaged baz, Gitua baram 'k.o. tree: Erythrina indica' probably are cognate. The similarity of these to the other forms cited here is attributed to chance.

tree

trees:   bare, of trees

Chance, or borrowing from Malay.

tremble

Chance. Bimanese βiβi probably reflects *berber or *birbir.

trickle:   drizzle, trickle

Chance. I cannot find a Tagalog word of this shape with the stated meaning in any modern dictionary. Based on these two forms, and Javanese luh ‘tears’, which clearly reflects PAn *luSeq, Dempwolff (1938) nonetheless proposed Uraustronesisch *luRuq ‘to drizzle’.

turbid

KOM butéʔ is assumed to be a Javanese loan. The resemblance of Fijian butō to these forms is attributed to chance.

turn

turtle sp.

twin

twinkle

tying material