As the Wikipedia article on reduplication states, "There is a tendency for prefixing reduplicants to copy left-to-right and for suffixing reduplicants to copy right-to-left". But there are counterexamples, e.g. Tillamook has reduplication of the form C₂-C₁VC₂-. Does the opposite pattern exist, where the onset of a root is reduplicated as a suffix?
-
Can you give an example of the kind of Tillamook reduplication you mean?– Omar and LorraineCommented Oct 16 at 13:56
-
@OmarandLorraine See the wiki article: things like [ɡaɬ] 'eye' → [ɬɡaɬ] 'eyes'.– TKRCommented Oct 16 at 22:12
-
So you're looking for something like [ɡaɬɡ], right?– Omar and LorraineCommented Oct 17 at 7:29
-
@OmarandLorraine Yes, that would be an example.– TKRCommented Oct 18 at 2:03
Add a comment
|
1 Answer
Kind of, in Sumerian.
Sumerian verb conjugations divide into two basic types, called ḫamṭu and marû. Nobody knows exactly what these represent, but they're usually taken to be some form of TAM - perhaps ḫamṭu past perfective vs. marû everything else.
The marû usually, but not always, uses the same verb stem as the ḫamṭu. When it doesn't, the marû stem is either a suppletive stem, or else derived from the ḫamṭu stem via a suffix... or reduplication, and sometimes that reduplication surfaces as C₁VC₂-C₁V (emphasis mine):
ii) Verbs belonging to the REDUPLICATING CLASS use a reduplicated stem in the present-future. If the stem ends with a consonant, then the consonant is either omitted in the reduplicated form, e.g.: ŋar PT/ŋar/ : ŋa₂-ŋa₂ PF/ŋa~ŋa/ “to put”, naŋ PT/naŋ/ : na₈-na₈ PF/na~na/ “to drink”, kur₉ PT/kur/ : ku₄-ku₄ PF/ku~ku/ “to enter”; or it is preserved only in the first syllable: ḫal PT/ḫal/ : ḫal-ḫa PF/ḫal~ḫa/ “to distribute”, te-en PT/ten/ : te-en-te PF/ten~te/ “to cool off”. If the monosyllabic stem ends with a vowel, then the whole stem is reduplicated, e.g.: gi₄ PT/gi/ : gi₄-gi₄ PF/gi~gi/ “to return”; mu2 PT/mu/ : mu₂-mu₂ PF/mu~mu/ “to grow”. Some bisyllabic verbs form their present-future stem by a reduplication of the first syllable with an accompanying voicing of the consonant: tuku PT/tuku/ : du₁₂-du₁₂ PF/du~du/ “to have”; taka₄ PT/taka/ : da₁₃-da₁₃ PF/da~da/ “to leave”.
- Gábor Zólyomi, An introduction to the grammar of Sumerian, 2016, Section 9.3
(note: PT here stands for "preterite", which is what Zólyomi is calling the ḫamṭu; PF stands for "present-future", which is what he's calling the marû.)
So we can say at least that C₁VC₂-C₁- reduplication is attested in Sumerian, but I don't know if you would call it a pattern - it's certainly not regular.
