1    Body-part Prefixes (and Noun Incorporation) in Panoan and Takanan David W. Fleck, Adjunct Research Associate, University of Oregon, dfleck@amnh.org Amazónicas 4: Pano-Takanan Symposium, 27 April 2012, Lima Verbatim excerpt from Loos 2005, pp. 43-44: 5. PREFIJACIÓN QUE REPRESENTA UN CASO OBLICUO La mayoría de las lenguas panos poseen un juego productivo de prefijos. Algunos idiomas, como el cashibo, han perdido el empleo de los prefijos de manera que solamente se encuentran vestigios de ellos. EI prefijo sirve para incorporar una forma abreviada de un sustantivo a un verbo (11). (11) Capanahua [poríškikį] de poyan·riški·kin brazo-golpear-pres ‘golpear a alguien en el brazo' Puede añadirse un prefijo tanto a un verbo transitivo como intransitivo o a un adjetivo, y no cambia la transitividad. La prefijaci6n produce formas largas que luego son acortadas por las reglas que reducen los sustantivos trisilábicos, que mencionamos en la sección anterior. En las lenguas de la familia tacana también se encuentran vestigios del mismo sistema de prefijación. Por ejemplo, compárense los prefijos panos bi- ‘cara’ y mi- ‘mano’ en (12) (bu-), (13) (me-) y (14) (bo-/ba-). (12) Tacana (de Ottaviano y Ottaviano 1989) (13) Cavineña (de Camp y Liccardi 1989) bu-beta de dos caras emetucu mano bu-dhi besar en la cara emeiyuca dedo grande de la mano bu-dhi grano en la cara? meduya meter la mano en un hueco bupad’a golpear en la cara mepeya recoger bu-tse su al frente metara anillo del dedo ebu cara emeshuru cortarse las unas ja-bu-ba-ti mirarse la cara (14) Araona (de Pitman 1981) e-bo cara bo-tsabi golpear en la cara pa-ba-tatso-ti refl-cara-golpear-refl ‘golpearse en la cara’ Notes: 1) According to Girard (1971:159) proto-Panoan i = proto-Takanan e. So me- is potentially proto-Pano-Takanan, but bu-/bo-/ba- would be borrowings if they are not mere coincidence. 2) e- is a segmentable morpheme, so bu-/bo- would not be a reduced from of a body-part noun. 3) Incorporation and compounding of body-part nouns (and few other nouns) are common in Takanan (see page 7) so the forms in (12) and (14) appears to simply be productive or lexicalized compounds or cases of nouns incorporation. 4) In Cavineña, me- ‘hand’ is the only form that that has a corresponding poly-syllabic root, but metuku is not the proto-Takanan form, while in most if not all Takanan languages the noun for hand is me (see Girard 1971:101). 5) According to Guillaume (2004:143), the form mëtuku instead of me can sometimes be incorporated before verbs. 6) Conclusion: there is no productive or historical system of prefixation in Takanan languages. Only me in Cavineña could possibly be called a vestigial prefix, but more likely, the complex roots containing me are old incorporated verbs or compounded nouns. 7) Relevance to the purported Panoan-Takanan relation: Panoan prefixation may had evolved from noun incorporation and compounding. 2    Panoan languages do not have noun incorporation: Table 1: Comparison of the prefixes (h)an- and bë- in 3 Panoan languages. Table 1a: Shipibo (data collected by me, often guided by consultation of published materials). ————————————————————————————————————– Prefix Gloss of prefix Corresponding root ————————————————————————————————————– han- oral cavity, tongue hana bë- face bëmanan forehead bëtunku eye bëru eyebrow bëşhni, bëru karani tear bëun surface bëbi ————————————————————————————————————– Table 1b: Kashibo (data from Fleck and Zariquiey in press, mostly collected by R. Zariquiey) ————————————————————————————————————– Prefix Gloss of prefix Corresponding root ————————————————————————————————————– an- oral cavity, tongue ana palate manşhanta cavity, concave surface namë elongated hole kini bë- face, forehead, front bëmanan eye bëru eyebrow bëşhku tear bëun rheum (sleep of eye) bëşha in front of bëbun (postpositon) ————————————————————————————————————– Table 1c: Matses (data from Fleck 2006 or my database). ————————————————————————————————————– Prefix Gloss of prefix Corresponding root ————————————————————————————————————– an- mouth, tongue, palm (of hand), sole (of foot), (arm)pit ana gill slits (of fish) anmaëşh swampy depression in the ground anşhantuk cavity, concave surface, interior, underside — center (of path or stream) — bë- face, forehead bëtantete, bësbed eye ëşhë eyebrow bëşhni tear bëun rheum bëku front, surface — ————————————————————————————————————– Non-IPA symbols in my practical Panoan orthograpy: ë = ɨ, sh = ʃ, şh = ʂ, ch = tʃ, çh = tʂ, y = j, f = b, β, or ɸ (depending on the language), ‘ = ʔ. Stress will be ignored here. Marubo does not have noun incorporation or prefixation, but it has vestiges of prefixes (see Appendix B). 3    Grammatical properties of prefixation Prefixation of nouns: (1) Shipibo: ma-tunku head-bump ‘bump on the head’ Prefixation of adjectives: (3) Kashibo: bë-tunan uni eye-black man ‘man with black eyes’/‘black-eyed man’ (4) Matses: tsisembo tsi-ise-mbo butt-smooth-ABSOLUTIVE ‘smooth-butted (could refer to gun stock or machete handle)’ Prefixation of verbs: (5) Matses (example of lack of valence reduction) a. debi-n daskute-Ø pan-e-k Davy-ERGATIVE clothing-ABSOLUTIVE wash-NONPAST-INDICATIVE ‘Davy is washing clothes.’ b. debi-n Ø më-pan-e-k Davy-ERGATIVE 3ABSOLUTIVE hand-wash-NONPAST-INDICATIVE ‘Davyi is washing hisj hands.’ c. debi-Ø (aton mëdante) më-pan-ad-e-k Davy-ABSOLUTIVE (3GENITIVE hand) hand-wash-REFLEXIVE-NONPAST- INDICATIVE ‘Davyi is washing hisi hands.’ (6) Kashibo (example of “extra participant”) David-tan ka Dunú-ø Roberto të-tsoon-a-şh-a David-ERGATIVE INDICATIVE.3 Dunú-ABSOLUTIVE Roberto neck-seat-PAST-3- NONPROXIMATE ‘David sat Dunu on Roberto’s neck’ (lit., ‘David neck-sat Dunu on Roberto.’) (2) Matses (exmple of productive prefixation) [şhukkate pinchuk] şha-şhubu-wa-aşh fan palm.species crown-nest-VERBALIZER:make-after:S/A>S(SAME SUBJECT) [kapa piu] uşh-kid squirrel red sleep-HABITUAL ‘After making a nest in the crown of a şhukkate pinchuk palm, the Amazon red squirrel sleeps.’ (lit. ‘After making a “crown-nest” in the şhukkate pinchuk palm...’) 4    Having shown that this is not noun incorporation, let us next consider the origin of Panoan prefixes. Table 2. Inventory of Matses, Kashibo, and Shipibo Prefixes. ————————————————————————————————————– Matses Kashibo Shipibo (Basic) Gloss a ————————————————————————————————————– an- an- han- mouth ban- elbow ba- armpit bë- bë- bë- eye, face bu- bu- head da- ra- ra- torso dan- ran- ran- knee dë- rë- rë- nose ëk- kwë- kë- lips, jaw (cf. Matses kwi-) in- in- hin- tail, penis in- temple (cf. Matses pan-) ka- ka- ka- back kui- ku- jaw kuis- ki- ki- thigh (upper leg) ma- ma- ma- head më- më- më- hand mi- small ball nak- na- na- abdômen, center në- në- në- liquid nu- nu- belly, navel nik- navel pa- pa- pa- ear pan- temple pë- pë- shoulder, wing pë- pën- pun- arm, wing pi- pi- rib po- abdomen şha- şha- şhan- crotch, unopened frond şhan- unopened frond şhëk- şhë- şhë- tooth shik- shi- chest şhu- şhu- breast ta- ta- ta- foot tak- underside tan- tan- tan- cheek të- të- të- neck tsi-/chi- tsi-/chi- tsi-/chi- butt u- hun- testicle wi- i- wi- shank (lower leg) ————————————————————————————————————– a Note: the “Basic gloss” is not always completely accurate for all three languages, since the semantics of the prefixes sometimes varies among languages; the particularly problematic cases are the prefixes kwë-, kui-, ku-; pë-, pën-, pun-; and şha-, şhan. 5    Table 3. Comparison of corresponding roots Table 3a. Ten (of about 28) Mayoruna body-part prefixes, including formally similar full nouns. ————————————————————————————————————– Full nounsa Prefix Gloss of prefix Chankueshbo Kulina Matses Matis ————————————————————————————————————– an- mouth, inside, underside, concave surface ana ana ana ana dë- nose, front, tip, prow, beak, bill dëşhan dëşhan dëbiate dëşhan ëk- lips, labia, edge of round thing or opening ëkbid ëkşhan ëkbid ëkşhak in- tail, penis inkuente ina inkuente ina ma- head, fruit, top, etc. maşho maşho mapi maşho më- hand, forearm, branch mëdante mëçhate mëdante mëkën pa- ear, antler, edge of swidden, pot handle pabëşhan pabëşhan pabiate pabëşhan şhë- tooth, arrowhead, crab pincer şhëta şhëta şhëta şhëta të- neck, foreshaft, leaf sheath, peduncle tëşho tëtun tënidte tëtun tsi- butt, arrow notch, end, stern, vulva tsiben tsiben tsitsu tsitsu ————————————————————————————————————– a Nouns correspond to the first word in the prefix definition and usually not to the rest. Chankueshbo and Kulina of the Curuçá River data were collected by the Fleck. Matis data were collected by Fleck with the help of R. Ferreira. Table 3b. Ten Mainline body-part prefixes, including formally similar full nouns. ————————————————————————————————————– Full nouns a Prefix Gloss Kashibo Chakobo Shipibo Kapanawa Kashinawa Yaminawa Amawaka ————————————————————————————————————– (h)a(n)- mouth ana hana hana hana hana aşhban hana ‘tongue’ rë-/dë- nose rëkin rëkini rëkin rëkin dëkin rëchuku rëkin k(w)ë- lips kwëbi këbichi këşha këşha këbichi këşhan këşhaa (h)i(n)- tail ina hina hina hina (hina) tsispa (hina) ma- head mapë mapu mapu mapu mapu mapu mapu më- hand mëkën mëkënë mëkën mëkën mëkën mëkën mëkën pa- ear pabi pa’uki pabiki pabinki pabinki pachu (pabinki) şhë- tooth şhëta şhëta şhëta şhëta (şhëta) (piti) (şhëta) të- neck tëxa tëtu tëşhu tëşhu tëşhu tëşhu tëşhu tsi-/chi- butt chişhu chikaşha chishu chishki (chişhu) chipa chinkan ————————————————————————————————————– a Entries in parentheses in the Kashinawa, Yaminawa, and Amawaka columns represent a lack of data in lists of prefixes, i.e., the nouns can be looked up in dictionaries, but the published material does not specify that these prefixes exist in these languages (though the probably do). Sources of data: Chakobo: Zingg (1998); Kapanawa: Hall de Loos and Loos (1973), Loos and Loos (1998); Kashinawa: Eakin (1991); Yaminawa: Faust and Loos (2002), Amawaka: Hyde 1980). Some preliminary conclusions: Most prefixes can be reconstructed to Proto-Panoan. Only a few of the corresponding body-part nouns can be reconstructed to Proto-Panoan, but, interestingly, the first syllables of the nouns hardly vary. 6    Table 4. A selection Matses prefixes, illustrating analyzable corresponding nouns (check out the ones that are footnoted). ————————————————————————————————————– Prefix Gloss of prefix Corresponding noun ————————————————————————————————————– dan- knee dannëşh back of leg opposite the knee dantuante kneecap dantuku,a piuşhëmpi dë- nose, tip, prow, beak, bill dëbiate (nose) septum dëpun nose cartilage dëşhu (house) gable dëchin ëk- lips, labia (of vulva), edge of round thing or opening ëkbid b mustache ëkchish c më- hand, (tobacco pulverizing) mortar mëdante forearm mëpu wrist mëtete d projecting carpal bones mëşh e elbow mëntsimpis finger dëşhbi ‘finger, toe’ knuckles mëbuduşh fingernail mëntsis branch kuidi pan- temple pampun part of cheek near sideburns pampada f shik-/sik- chest shikdiadkid ribs, house rafters shiktodo keel (of bird or some fish) shiktun sternum (central front part of rib cage) shikçhoşhkon breast şhuma cleavage shikşhodo g wi- shank (lower leg) — shin wipu shin bone wispo calf wishuku h buttress root wibën base of tree trunk — ————————————————————————————————————– a tuku is a synchronic but obligatorily prefixed root meaning ‘round lump.’ b bid is a synchronic but obligatorily prefixed root meaning ‘skin’ or ‘skin + adjacent flesh.’ c chish is a synchronic but obligatorily prefixed root meaning ‘hair, fin or other projection.’ d tete is a synchronic but obligatorily prefixed root meaning ‘joint (of a limb).’ e ëşh is a synchronic but obligatorily prefixed root meaning ‘lump, tumor or swelling.’ f pada is free noun root meaning ‘flatness.’ g şhodo is a free noun root meaning ‘deepness.’ h shuku is a synchronic but obligatorily prefixed noun root meaning ‘round muscle.’ Further conclusions: Modern body-part terms appear to be composite forms, most of which appear to have arisen after the proto-Panoans dispersed. This suggests that there existed monosyllabic body-part morphemes in proto Panoan (or before). But what is still unknown is if the reconstructable monosyllabic body-part morphemes were prefixes or roots, and whether “basic” body-part terms created by prefixation or compounding. 7    Back to the purported genetic relation between the Panoan and Takanan families. Table 5. Takanan monosyllabic body-part and other incorporable nouns. ————————————————————————————————————– Cavineña noun root Proto-Takanan form Corresponding Panoan prefix ————————————————————————————————————– aa ‘branch’ pë ‘upper arm, branch’ bawa ‘face’ bu ‘face’ bë- ‘eye, face, forehead’ bi ‘arm’ pë ‘upper arm, branch’ metuku ‘hand’ më ‘hand’ më- ‘hand’ na ‘water’ na ‘water’ në- ‘water/fire’ ta ‘leg’(only Takana) ta- ‘foot’; wi- ‘shank’; kuis ‘thigh’ tse ‘tooth’ tse ‘tooth’ şhë- ‘tooth’ wi ‘beak’ dë-/rë- ‘nose, beak, tip’ ————————————————————————————————————– Payne (1990:218) wrote “In at least Arawakan, Harakmbet, Panoan, Cayuga, and Tupí-Guaraní languages, noun roots may be compounded, or incorporated, with verb roots.” Table 6. Comparison of morpho-syntactic characteristics of Panoan prefixation of verbs and Takanan noun incorporation. ————————————————————————————————————– Panoan Takanan ————————————————————————————————————– mostly body-parts mostly body-parts. prefix directly before the verb root noun directly before the verb transitivity class unchanged (extra participant added) valency unchanged transitive or intransitive verb roots mostly transitive verb roots adjectives can be prefixed a parallel process occurs with adjectives ————————————————————————————————————– Panoan data based on Matses, Kashibo, and Shipibo; Takanan data based on Cavineña (Guillaume 2004:143-46, 389-393) and Maropo/Reyesaño (Guillaume in press), where these languages are explicitly said to have productive noun incorporation and possess these properties. Entries in Araona and Takana dictionaries (Ottaviano y Ottaviano 1989 ; Pitman 1981) appear to be products of noun incorporation (see Loos’ examples on page 1 of this handout), but it is not certain if this is a productive process in these languages. I have not looked at all the published grammatical descriptions of Takanan languages yet. Aikhenvald (ms.) provided the following insights on noun incorporation based on new data from South American Languages: body parts and/or all inalienably (obligatorily) possessed nouns are the ones most likely to be incorporated. This is the case in many Tacana and Arawak languages (also Nadëb). A kind of incorporation in Yine (Piro) does not affect valency. Conclusions: 1) Takanan languages evidently did not have prefixes, but they have noun incorporation. 2) The Panoan languages that have been studied carefully do not have noun incorporation. 3) Modern body-part terms most likely arise from compounding; “non-basic” body-part terms certainly seem to have arisen from compounding (footnoted elements of Table 5 are mostly semantically nominals, but there is no evidence yet that Panoan prefixation of verbs was preceded by noun incorporation. 8    References: Aikhenvald, Alexandra. ms. “Some ideas about new insights into noun incorporation: A view from South America.” Eakin, Lucille. 1991. Lecciones Para el Aprendizaje del Idioma Yaminahua. Documento de Trabajo No. 22. Yarinacocha, Peru: Instituto Lingüístico de Verano. Faust, Norma and Eugene E. Loos. 2002. Gramática del Idioma Yaminahua. Serie Lingüística Peruana No. 51. Lima, Peru: Instituto Lingüístico de Verano. Fleck, David W. 2006. “Body-part prefixes in Matses: Derivation or noun incorporation?” International Journal of American Linguistics, 72: 59-96. Fleck, David W. 2007. “Did the Kulinas become the Marubos?: A linguistic and ethnohistorical investigation.” Tipití, Journal of The Society for the Anthropology of Lowland South America, 5:137-207. Fleck, David W. in progress “Panoan languages and linguistics.” Girard, Victor. 1971. Proto-Takanan Phonology. Berkeley: University of California Press. Guillaume, Antoine. 2004. A Grammar of Cavineña, an Amazonian Language of Northern Bolivia. Ph.D. dissertation, La Trobe University, Bundoora, Australia. Guillaume, Antoine. in press. “Maropa (reyesano).” In Lenguas de Bolivia. Hall de Loos, Betty and Eugene Loos. 1973. La estructura semántica y fonológica de los prefijos verbales en capanahua. Estudios Panos I, ed. E. Loos, pp. 63-132. Serie Lingüística Peruana No. 10. Yarinacocha, Peru: Instituto Lingüístico de Verano. Hyde, Sylvia. 1980. Diccionario Amahuaca (Edición Preliminar). Serie Lingüística Peruana No. 7. Yarinacocha, Peru: Instituto Lingüístico de Verano. Loos, Eugene E. and Betty Loos. 1998. Diccionario Capanahua-Castellano. Yarinacocha, Peru: Instituto Lingüístico de Verano. Ottaviano, Ida and John Ottaviano. 1989. Diccionario Tacana-Castellano, Castellano-Takana. Summer Institute of Linguistics, Dallas. Payne, Doris L. 1990. “Morphological characteristics of lowland South American languages.” In Amazonian Linguistics: Studies in Lowland South American Languages, ed. Doris L. Payne, pp. 213-241. Austin: University of Texas Press. Pitman, Mary de. 1981. Diccionario Araona y Castellano. Instituto Lingüístico de Verano, Riberalta, Bolivia. Zariquiey Biondi, Roberto & David W. Fleck. in press. “Prefixation in Kashibo-Kakataibo: Synchronic or diachronic derivation.” International Journal of American Linguistics, 78(3). Zingg, Philipp. 1998. Diccionario Chácobo-Castellano Castellano-Chácobo con Bosquejo de la Gramática Chacobo y con Apuntes Culturales. La Paz, Bolivia: Ministerio de Desarrollo Sostenible y Planificación Viceministro de Asuntos Indígenas y Pueblos Originarios.   9    Appendix A. Panoan classification by Fleck (2007, in progress) based on lexical and other similarities. Tally: about 19 extant languages, and about 13 documented extinct languages = 32. Languages in bold; dialects in italics; † = extinct; * = obsolescent (i.e., no longer spoken as an every-day language, but a few speakers remember it). Dialects with minor differences are listed on the same line. I.  Mayoruna branch (4 extant and 4 documented extinct languages)    A.  Mayo group      i.  Matses subgroup        a.  Matses (3 dialects):            Peruvian Matses; Brazilian Matses            †Paud Usunkid        b.  Korubo (2 dialects) (tentatively placed; may compose its own subgroup)            Korubo            *Chankueshbo        c.  *Kulina of the Curuçá River (3 dialects):            *Kapishtana; *Mawi            *Chema        d.  †Demushbo (last speaker died in 2010)      ii.  Matis subgroup (most similar to Mainline branch)        a.  Matis (most divergent from other extant Mayoruna languages)        b.  †Mayoruna of the Jandiatuba River        c.  †Mayoruna of the Amazon River (2 dialects):            †Settled Mayoruna of the Amazon River            †Wild Mayoruna of the Amazon River    B.  †Mayoruna of Tabatinga (phonologically most divergent Mayoruna unit)    II.  Mainline branch (about 14 extant and about 10 documented extinct languages)    A.  Kasharari (most divergent Mainline language)    B.  Kashibo (4 dialects; similar to Nawa group due to contact with Shipibo)        Kashibo (Tessmann’s “Kaschinõ”); Rubo        Kakataibo        Nokaman (formerly thought to be extinct)    C.  Nawa group (subgroups ordered from most to least divergent)      i. Bolivian subgroup        a. Chakobo/Pakawara (2 dialects of 1 language)        b. †Karipuna (may be a dialect of Chakobo/Pakawara)      ii. Madre de Dios subgroup         a. †Atsawaka/†Yamiaka (2 dialects of 1 language)        b. †Arazaire      iii. †Remo of the Blanco River      iv. †Kashinawa of the Tarauacá River      v. Marubo subgroup        a. Marubo (of the Javari Basin)        b. Katukina  Katukina of Olinda; Katukina of Sete Estrelas           †Kanamari        c. †Kulina of São Paulo de Olivença      10    Appendix A continued.   “Central Panoan Assemblage” (subgroups vi‐viii): evidently areal influence among neighbors has  blurred the relations among these languages.      vi. Poyanawa subgroup          a. *Poyanawa        b. *Iskonawa (very close to Poyanawa, but also resembles Shipibo‐Konibo‐ Kapanawa and Amawaka)        c. *Nukini        d. *Nawa (of the Môa River)        e. †Remo of the Jaquirana River      vii. Chama subgroup        a. Shipibo‐Konibo‐Kapanawa (3 dialects of 1 language)          Shipibo; Konibo (currently fused)          *Kapanawa of the Tapiche River        b. *Pano            †Pano            *Shetebo; *Piskino        c. †Sensi.      viii. Headwaters subgroup        a. Kashinawa of the Ibuaçu River          Brazilian Kashinawa              Peruvain Kashinawa           †Kapanawa of the Juruá River        †Paranawa      b. Yaminawa (large dialect complex)        Brazilian Yaminawa        Peruvian Yaminawa        Chaninawa        Chitonawa        Mastanawa        Parkenawa        Shanenawa        Sharanawa; *Marinawa        Shawanawa (= Arara)        Yawanawa        *Yaminawa‐arara (not same as Shawanawa/Arara)        †Nehanawa        c. Amawaka          Peruvian Amawaka (intermediate between this subgroup and Chama  subgroup, perhaps as a result of areal contact)          †Nishinawa (= Brazilian Amawaka)        †Yumanawa (also very similar to Kashinawa of the Ibuaçu R)        d. †Remo of the Môa River (resembles Amawaka)        e. †Tuchiunawa (resembles Yaminawa dialects)  11    Appendix B. Vestigial prefixes in Marubo. ————————————————————————————————————– vestigial prefix: më- ‘hand-forearm’ mëbi ‘hand’ mëntsisi ‘(finger/toe)nail’ mëtunti ‘index finger’ (finger is mëbi rëbu) mëruşhku ‘wrist’ or ‘knuckle’ mëtaşhë ‘forearm’ (iwi) mëan ‘(tree) branch’ mëşhuki ‘crawl’ mëki ‘work’ (I might be pushing it here) mëshmiri ‘right (side)’ mëkiri ‘left (side)’ vestigial prefix ma- ‘head, hill, top’ mapu ‘head’ matu ‘hill’ machi ‘top (of hill, house, etc.) machuti ‘crown of head’ manşhun ‘antler’ maiti ‘headband’ (there may be a verb mai ‘wear on head’) maşhkuri ‘cut hair close to head’ manë ‘fill to top’ vestigial prefix pë- ‘upper arm puyan ‘upper arm’ pëşhu ‘shoulder’ pëi ‘wing, feather mani pëi ‘(plant) leaf’ vestigial prefix ta- ‘foot’ taë ‘foot’ tapuşhë ‘ankle’ (iwi) tapun ‘(tree) root’ tariti ‘shoe’ (there may be a verb tari ‘wear on/cover foot’) vestigial prefix bë- ‘eye, face, forehead’ bëru ‘eye’ bëmanë ‘face’ bëtunti ‘forehead’ bëshpi rani ‘eyebrow’ bëun ‘tear(s)’ bëku ‘rheum (sleep of the eyes), eye pus’ bëpan ‘bangs’ bëisti ‘mirror’ (there may be a verb bëis ‘look at one’s reflection’) (yura) bëshuya ‘blind (person)’ vestigial prefix kë- ‘lip, edge’ këşha ‘lip’ këni ‘beard, moustache’ këmu ‘saliva’ (waka) kësu ‘(river/stream)bank’ kënwan(-ka) ‘sharp’ këntu(-ka) ‘dull’ vestigial prefix të- ‘neck’ tëşhu ‘neck’ tëpun ‘front part of neck’ tënsan ‘throat’ tëntu ‘Adam’s apple’ vestigial prefix pu- ‘abdomen’ pustu ‘belly’ pubi ‘abdomen flesh’ puku ‘intestines’ vestigial prefix rë- ‘nose, tip’ rëkin ‘nose’ rebu ‘tip, headwaters’ ————————————————————————————————————– Notes: 1) I attempted to elicit productive prefixation of nouns, verbs, and adjectives by translation requests in Portuguese or Matses (the latter from bilingual Matses-Marubo informants). Nothing that looked like productive prefixation was given in the translations to Marubo. Generally the full noun was used, or else a vocabulary item that did not involve anything tht looked like a body-part prefix was used. 2) Prefixation has not been described for Katukina in Katukina. Though it is possible that researchers may have missed it, but there is a good possibility that Katukina, being so similar to Marubo, also lacks prefixation. Significantly, Kasharari is unknown with respect to whether it has productive prefixes, as are the languages in my Poyonawa subgroup (Poyanawa, Iskonawa, Nukini, and Nawa (not to be confused with Parkenawa/Yora), all of which are obsolescent. 3) More such vestigial prefixes would surely be uncovered with a longer vocabulary list. 12    Bonus: Bora, según Thiesen 1996, pp. 56-57 Muchos de los verbos no solamente indican la acción o el efecto de la acción, sino también indican la manera o el instrumento usado para hacerla. Estos verbos llevan formas que aparecen como prefijos de la raíces. Se nota que do- indica que la acción ha sido hecha con la mano, di- indica que la acción ha sido hecha con los dientes, ca- indica que la acción ha sido hecha con algo puntiagudo, ki- indica que la acción ha sido hecha con algún instrumento cortante, pi- indica que la acción ha sido hecha con algún instrumento como un serrucho, ta- indica que la acción ha sido hecha con el pie y wa- indica que la acción ha sido hecha a golpes. Por ejemplo: dópújuhjáco romper algo frágil dóhdahíro quebrar algo en pedazos (con la mano) (con la mano) tápújuhjáco romper algo frágil táhdahíro quebrar algo (con el pie) (con el pie) wápújuhjáco romper algo frágil wáhdahíro cortar o trozar (a golpes) (a golpes con algún instrumento) cápújuhjáco romper algo frágil (con algo puntiagudo) dípújuhjáco romper algo frágil díhdahíro quebrar algo en pedazos (con los dientes) (con los dientes) kíhdyahíro trozar algo (con algún instrumento cortante) píhdahíro trozar algo (con serrucho) Notas de Fleck: Solamente 3 de los 7 prefijos refieren a partes del cuerpo. ta- es el prefijo para ‘pie’ en lenguas Panas. y (según Antoine) también en Reyesano y Araona. Según Thiesen y Thiesen 1998, los sustantivos en Bora para pie son túhaá y méjúhaá; para mano son hójtsií y méhójtsíí; y para diente son méhwájíí y peenújtso (incisivo); o sea, parece que los prefijos en Bora no son relacionados formalmente a sustantivos en Bora. Frank Seifart (comunicación personal) dice que estos prefijos no son muy productivos y se pueden usar juntos con el sustantivo que denomina el instrumento; Thiesen y Thiesen (1998) no incluyen estos prefijos en sus listas de afijos o como lexemas.