Scribd
Upload
182 views17 pages

Causatives Oromo

language Oromo

Uploaded by

JosephAffonso
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
182 views17 pages

Causatives Oromo

language Oromo

Uploaded by

JosephAffonso
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
You are on page 1/ 17
LEXICAL AND SYNTACTIC CAUSATIVES IN OROMO. SrastryDuansky — MaKiARosaL Loner ——_PAUL-NWaAN University of California, Indiana Universiy, Indiana Universiy “Senta Cras ‘Berens ee ee Ta anes ‘Scie ners comin my aso ome vreau ft ceive SSpetee howe oa peatiee topes ote seen oneal Ss, at ee tt ‘Sint ese eager heleon cle Se” 1. 1n Oromo formerly known a Gall) Cushitc language of Ethiopia snd Kenya, the formation af causes is sccomshed though the alain of ‘bound mtorpheme (wih surface variants st, ad is) and the adton ‘fan appropriate argument (1) aman ni daaat, rile botox? "The milk bold” eran canna daanfie “on milk boiteausaon “Terfa mad the mk boi (Le, “Ter oiled the mi’ ‘The affhation ofan addtional sand an argent to an already eausativized form reels in mite suv. (2) tevin ganteessaa aannan daanfystse ow iil oikeatcnt-so “Terfa made Gamtesa oil the mil. ‘As was shown in Lovet 1987, verbs can be classed according 10 whether thei inal causative forms require one “x mompheme or wo (the vowel A! between the two's ins sade by a rule of epenthesis) 4s Morin pl 197, We ovld fo tak Addn Toten ar hi ovale hp and ‘Sofia an a at nas Una We wo trae et egece ar reo ‘Big, Grenbe Care, Wiis Gee ul nant Tule- Wenlock eStats Ea A oe eb sec er oan vs ne mise @) gop bedey) | Dag “rink* fot ake dy? Data "make drink Hayward (199633) noted that this division correlates with che number of at nls inthe resultant coorave consrcton. Thus, causatves TEER gor ans vers or from what Hayward term ‘ageive nian Herre afi wth two. morphenes, while those formed from "nonsge- {ie haransives are affixed with one ~+ morph. (a) a. Degli akedrink” ( Deekamsi "make any’ Pom “betbeakty” > foyyie ‘make heathy” Furthermore, the subject of tanive verbs of perception Chest see), care aim account are understood have the tema ole “expeincer™ TEST han "agen aso form thir eavatve with 1Wo -s morphemes. (Oh arg "see! > aris “make see 2 thetic, he sl eee a ee re soy, Te we fo sooltpi ast he maa nen see 28 1.2, Second, one nds tat the addition of an -1 morpheme toa causivized form doesnot always correspond tothe addition of another agetive nomial ‘The adetional sin 8, fr example, serves often the nature ofthe caus sion rather than to add further cause. 8) tefacn urbe raf. ost boy sleepenu-ac “Tert pt the boy to slesp (eg by rocking him)” be tecfeaen purbae ravine ost boy sleepexusie-acs "Teria made the boy sleep (by ving him a sleeping pil)” 2.1, The fs problem can be handed by looking atthe numer of mor hemes in causative consretions in tems of ammatcal eetion caer has ff thematic roles, That is, We ae stl operating inthe spit of Haywards Seqroach, bu we are suggesting that Che correlation realy beeen =» mote [hemes sid grammeticel subjects rather than between - morpemes nd he Ina agents, Our proposals that the verbs which form ther causatives via Single ~s morpheme ae all uaccusative, fe. verbs whose single argent i iniay ect objet; bY contrast, verbs which form eausatives Wath (0 = ‘morphomes are citer tensive (verb that select anil subject and dict ttjec) or unegstive (verbs that select an ntl subject oly). Using the re Istonalnotatian of for sujet, 2 fr dec object abd P for predicate, the ‘elatonal networks of clauses cooaining cach ofthese thee classes of verbs se represented in A teen aannan Dus (ease) Yow mi drink-sce ZS “Te rk the ile eae oe 1 tafuon fie (eerie See ' ‘Tein! tujna He . enmana dont coacuiv) i Sipe otras 2 "The mi boled” iy eanen doar 19, the inal direct object of dango” advances to subject to satisfy ‘the universal equrement that every basic clause have afta suec (ihe Fal 1 Law: Permutter 98D, Johnson & Postal 188+ undertyinly, daa "boil and other verbs ofthis las are subjetess. While the ntansive seatenees SSoand 9c are srutualypaall on the srface,undelyingly they ae p= sumed tobe grammatical distinct. Note that i our analysis verb such as Deckans "be aray’ which takes two =r morpberies inthe casative, ers from a verb such 3 fy. "be heathy’, which takes only one s, not i any intrinsic semantic agenvity of te surface subject bins language specie ‘alegorizaion ay an unereatve as opposed aa unaceusaive Yer (00) a, texan Deckan-e Now beangry-aci 1 “Terta was ary (eta, gamtessaatefaa Dechamsise) “Games made Tera ang verfaa——_Deckam 0. teen fae ont behest xo AN ste iy" i (cE games io fei) [e \ ‘Sanccin ae ers iy. terfa fas “The causative construction under examination here is taken to be 2 anion “onstrston inthe sense ct Davies Roxen 988 I the formation of 2 ca ‘ative the morpeme bic sel prediat, is introduced into the clase ‘long wit an addtional subject argument. The subject of the embeded prod feat (when preset is unilormly reveled ae a vet stject The embedded +s cms sir sil poet by Cy 0 he een ef etc Prettiest ht oer ‘ition omy i.e Pdeio Racb aongwih aee t Seon pat esnagrnntamn ailes (rocoto tempat nce rey oer peek ‘ore fa han not ee ered by mes an pee se ‘Pam emanates ie pea ee nd mein. ec Seige ae & oa eh, Ucn ico be ‘Stam nee beak bay we predicate andthe embedded dlect object of «transitive prodiate lose thee ‘elationa tat and become Chomeurs (co) this onsstent withthe Sts Uniqueness Law, which provides that no two dependents can bear the Same stunmatial elton in the same stratum Urson 4 Posi 1980; see also Dubinsky 1988 and 188 foe the appropriate revision ofthis lav) a terfaa gamicessog—aannan Dag tt mike” atk eau tefua—purboa fet ©. =U) ON teyfaa—ganman daft wile tol “caw We cam atrbute the realization of each s morpheme in casativ to the presence of subject nthe lise, Thus, sings the verbs Deg-“vink and fi “run each introduce thir own subject nominal (in siton to that aed By the causative morphome ise, their cateatives show up with Wo sales, Inthe cate ofthe constuction frmed with the unaseusaive predicate dean ‘ol oly th eausaive predate introduces a sbjet, and therefore, only ‘oe ix shows up 22. Thisamalyss doesnot adress the sosond problem for Haywards trea ‘ment of causativen, amy the existence of casatve constructions which sd fan 5 but do.noe add an agersive argument. The explanation ere lies in an tnderstanding of the cavsarivesrensvr constriction, wish, withthe cx ception of Lloret 1987, hasbeen negicted in the Oromo Iiterature. Attention {othe causaiventensiveconsricton, moreover, provides important evidence that allows one to dtingish more acuraely the nature of causasne phe nomena in Oromo, While si usually described (and usually fonction) 38 ausativemorpheme ted the numberof undeiying grammatical subject an, as was showa above in serve a an tense (i). (02) a, terfaen tlre gurba aannan Dugstivise Now boy alk digkcavcne-von “Terfa made Tolta make the boy drink the mil” whuwn gurbus aamen Dapeticire, ‘ow boy make dehccat-Nt-AGR “oltu mae (forced) the boy to dink the milk" ‘Therear, however, suet imitions onthe se of-sinits intensive faction. While tho cousatives ofall transitive verbs and all uncrative intransitive low the ation of te ier ony sxe unaccusative verbs (verbs that form thee ‘aust witha singles) do so. For example, te additon of sto the eausaive irons "ake dry” ean ret i a double easatve or an Intensive, depending ‘on the amber of overt aguert (D3) a. terfaan —ganucessaa muda gorse. now ‘Rood dry ext-onusacn “Terla made Gamtesa mae the wood dy. 1h echen mabe rete ‘Now wood dy-cat-n-ack, “Teefa mad 7] the wood dr.” "he form ahs make break’ th otber hand, cannot form an intensive constrain with the addition of. Te affixation of = abs: mast nec scaly be accompanied by the aon of an argument Thus Tas ie ram ‘aa, whe Labi formed 14) a. eran — ole nooner Cabs Som pot breakeabicndnen “Terfa made Tolta make the po break, b. Mefaan twee Cab-siie. Sow pot bresieeat.er- scm "Terfa mad +11) the pot bresk” ‘The doube eave form Sabri "make someone ake somthing brea’ ‘on the ater hand, permis the addon of the wr, This, the afation of “in TS may, but eed ot ad a8 arunent (15) 3. terfeen —tolau gamteessua nace ab-sieive Now Pot break-cacscnv-cataan “Teafa mad ‘Tolta make Gamtosa mae the pot break.” by recoarn tlun torwee Cabs Now POL” Breakeat-cabanr “erfa made +] Tol ake te pot break. [tis importantio note th he intensities onyanntesifer of austives, cannot be alfned toa bare uncasaivized ver. The fax that mts cana be attached dieiy to snp sens, as posible wth polysemous causative! lntensivefrmations in Semitic. Moscat eal 164120), argues aginst the ‘ueatment of errand casas distinct homephonous morphemes (et Hamp's 1985.64 claim to this effec for Sans. An anal is more in Keeping with he ature of Oromo and eter Cusite languages is thatthe causative and the intensive are essentially the sume morpbere, but that they ae introduced by distinct processes and have diferent srammaticl consequences. The ca “ts added directly to any stem, whee the number of 's coreles with he number of subjects. The aon of ens toa causative verb requres the spacomitant addon ofa subject argument and poduees a multiple causative. ‘The cauatvesntensive constuction onthe other band, sengthens te male ofthe causation by reduplicating the found on a cassatve veri® Since the process of reluplicaton intensification does ot invlve sdding 2 ne inde Pendent cat marpeme, tt doesnot ental the addon of anew subject ar ‘Bent. This explains the semingly pucling now in intensive-canestive ‘onsiuctons between the ruber of + a the mer of grammatical stb Jess reroeeted (15). maken gop. ‘Woodson pac “The wood did” b. tok pops, ‘Wood-nan dey stan “The wood did) 1 a terfarn suchas Doane. ‘ou boy beatucie “terfa tet the boy , serfan gurbaa Daan-se. ost boy beara “tert beat 2) the boy.” (Our characterization ofthe caustivesintensive as 2 redupiative process finds further sippor in costings comparisons. Like Oromo, uner ‘of Souther Bantu languages among them Nauti, Sotho, Venda, Tsonga, and Shona—form thr causative by taecting vet sem mith the orem fr ves Doke 1984)” Some examples om Sotho wre provided 18 (Duke sk 140-40. (18) ufos ese” + uoisa “esse to hea’ Ide sit isa “set down’ Poolaea “il “> boolive aust kt In some ofthese languages the intensifier is ofen a reduplicaion of the cau can nd it eee nd teens Sa ae, satve morpheme. This isthe case in Neuat (Deke 1954105106, Sot (p. 14), and Ton pp. 194-35). i) Causative Inensive News sie ssa Sabo: “sa “sa ‘The fect that thi i education and nota case of concent homophony between the causative marker andthe intensive marker Becomes futher ap paren when one examines other Southern Banta languages, In Oshivasib, {or example, which has the causative morgheme ~ih we find the intensifier lk Vijoe 1978. In Svar here are thos coexsont for the intensife All of them reduplicauons is, le an al (Ziersoga 1952), Novis te ‘Phenomenon of redplication encoding ners peculiar to Aftian hnguages ‘Moravesic (19783212), in a general dlsussion of redupeatve costo: tions. cts recuplcave intensifier in Turkish, Sundanese, Ata, Teg, ‘Thai and Tagalog 3. Given thatthe ris aacossrilyrostictd to casative verbs, since tis formed by edapiating an existent causative morpheme, one might question ‘whether those appareatcaeatves fo which the latensifer cant be attached (eo. He) are ual causative. Suppose tha the causative predates, in a> dition to fret attaching o early any clause also combines derivatinally te sense of Lieber 1981 with limited cass of vers This predicate ike 29 ter, has a excl specfed valence. It minimally requires that fs clase ‘contain at leat a subject and a dret ject We might represent thi valence iin apt “Ths if causative -s combines i the lexicon with verb sch 8 the rsuing derived ver abs il have the sructare in 31 2D wiv eb 1s) Since its the head ofthe derived word, any lexical features the afi 5 ‘wil be feature of the devived form Sie the only relevant lexical feature ‘uved by is valence, it functions devivaonally as a ansveng al Gecving & transtive verb c'aby- break fom the inarsive ver ea~ ies Fabs ak LBS ttn een anche nn pr cent ‘eraton ney rfc ei seal eee cet ets “ey en Wh ae te at rw aly 2 Cabs 1021 cab.) Paiel In22, the unaccusatve ver ah “rel i seen tobe categorized for anita 2 rect objec only. When we derivtionaly attach the cau morpheme si Subeategorizion requremeats (an nal I ad 2) a imposed onthe derived form es in accordance with Lisber's Feature Percsstion Convencon (FPO) The meaning of the derived form i herve determined by the embedded sem é'ab in accordance with FPC I. ‘Tus, the formation of gogs- ‘make dey" involves the sestacrcaxation ‘of causatve toa cause, but the formation of abs-"beak ans) involves the trxeat afaton of easative “so a verb stem, Accordingly, 23's a auntie non construction as expresod by is relational network, while 36 is simple transitive clause continng&snale rede" 3). tefean mata gogree meten nten FP “Terfa mace the wood ary. aa so 4 mise te Figs wom pot break-ack <1) sobaeg ET poets pe eae This analysis claims that gogs-can te an intensifier because itis syntactic ost] ‘Accondingy, wings) transitive vor stem sch as &'ugaasshake results inthe fom Pagaours "make shake". Now if one examines th ex “Sntvized form ofthe propose tansivied verb abs "break (ass) itis [impale to determine whether the resultant frm & abs contains the suffix “ian which hs been phovolgicalyredoced (as would be expected of tran Sie stom) or whether contains wr (as would be the ease wih syntactically aust orm Q2) 2abe + ly > Cabs breakirnas} cat Gabe + ais > Cabal freaicea-” “caw Fortunately, there do ext forms which allow the claim tobe teste. If is ‘added toa verb stem ending in, becomes fin necortance witha general Pulatoatcate ule Lloret 1987). The affization of st nal ver roots este n examples such a in 38. 9 gal “le "tare heme "bing home” o + os shut "spend then “make spend the igh” 1m nr framework, elt behaves like aby “brea (6s) and is taken tobe x tvattvzed verb stem, with-r adden the eicon. By contrast Pu, which ishavestike gos -dy somethings causative form to which-s bas been Uikdedsymictealy. Since the seduction rale 31 oaly apples (othe je poston of sufi As doesnt apply in bese eases, and thes the pedition rade in 30 san be ested (he, one ca tll wheter one or wo ~smorphemes ave becn aed). Consistent wih our expectations, we tnd tht the ver ore {ele-does indeed behave ikea simple asitive ver stmin hat easativizes by the afiation of ss. (24 gate + slr + gals “pring home" ‘cat make bring ome (ass vet ster) ate. tes ote “The verb form bit, onthe oter Hand, casaivizes by the adton of single ‘i morpheme, conficning he sats ae sYotacclly produced casave (35) bul os bd “mcke spond the nigh” cA) “muke someone make someone (causniized nem) spend the nih ‘These sbvervatons are corroborated by the fact that bus~ can also be interpreted asthe Intensive of ule depending upon the numberof propriate srguments present (25) a, tern tlie bubeire ove spend the nightextsavrace “Terfa mae +] Tol spend the night” b. eraarn olwn gurboa bub ie. No boy Spend.the igh.cxs-oal-n "Terfa made Tolta make the boy spend the night “The adition of sis 10 gol however, cannot be an intensive, since the fntensive-sredupcation requires ihe presence of syntactically affixed = causative morpeme to operate on GI) a. “erfaan—tlau gitstive ox bring home-we-aoR “rerfa brought in] Tol: home: 1b. terfaan olen grbaa poe aive Now boy brig hone-cal-aae “erfa made Tol ring the boy Rome 4, We nave examined the eausntive consinicton in Oromo an have found that the observed morphological patterns can be accounted for by reference 10 (a snl rammatiea relations and () excl vs stare alison. I the fis cas, we Mewes the causative predate asa Bound morpheme “sand showed thatthe number of-+ afte in a causative contraction cor ‘espends othe aumbe fis inital logical subjecs, Thus, eausaives formed ‘with transitive or unerpative stems exhibit wos morphemes, and thos formed With unaccustve stems contin only one. orter on he bsisof to typeof causative intensification pallens lial ‘and perphvstic, we determined that te causative worpheme ~s combines Deasvarioat2y witha subct of the unacewsatve stems 1 frm fansive sesbs Unie Japanese, which has both lexical and syntactic causstives formed wit the sae stem (cf Shibtan 1976, MeCesey 197) tbe Oromo predicates ‘tht combine destvanally with -s do nat do so syntactically. The Japanese ‘er far home (it can form he lexical cautaive kar sen Rome (rnans) or the syatacseauslive Reeraer make go home (38) a, Mleko wa moo Kae or already went home *Micko already wen! home . Taree wt Micka 0” haya hast. Yor acccarly seathome “Tao set Mieke home eat" & Terao we Mirko haya kaeroset, Tor" Acc ealy pohome-cat-renr “Ta made Nicko go hone eal.” By contrast th intransitive Oromo ver ga. return hors" can only combine wits dervtionally to form te transive ver stem gal “ring home Ta depanese thre are sera perms of ranstisizing morphology sue hat lexical easatves canbe seen not to contain the synaci ausative podiate * {ANGUAGE, VOLUME 6, NUMER 86) “lore (cf. Jacobsen 1982:197-206) The examples in 39 iustate some ofthe ‘morphological prterns exhibited in Japanese tanskvelincheatie pis, ©) Inransiive ‘Transitive stern ben yal "bas aa “open ‘ere “ope! = fageru rane Tone “goo row extinguish? ‘dort “be supe! ‘dorokaen “sumpie Jn Oromo, on the oter hand, the tansitvising ax und the causative ais aretha same, Ths lena afation of + war shown to progice 3 change ‘he argument srt ofthe rerallant ver stem. Such a eet ii cord nce wih the Proction Prneiple (Chomsky 198:29, which sales that a= {ument siete chaages cat only occur inthe process of wor formation Further, these argument structure changes ae exnely what woud be expected ‘when one combines the eatsaive «lexically with Yer sem according 1 isbers 1981 Feature Pereaation Conventions ‘This lexicallyntactic distinction ws supported by the behavior af he cau- satvedoteniier, We characterized the intensier asa redoplcation of the ‘ausaive morpeme, Remembering tat ve ver forms lected With sn- [Ser morpheme could e intense, while otbers could not. 13 and 1), ‘seated this ference othe effi: fing excl i he fst case fad syntactically inthe second. The reason that as make) break” cannot, ‘be intensified is tht has Been added to b> dvivatonally to form a new Ver sem There f no independent causative morpheme inthis eonstraction fora duplicative vale o act on. Thus the form abs. cannot mean ake) Ireak [rte On the ether and, since goes: make dy" is 2 syntactically Prodacedcaurative, the wedupeaive ule CaN operate ns lone Yo produce [ogre "make dy [+1] ind pot sbould be raae with respect to the processes desrbed. We have noted thatthe causative moreme ran combipe excl With aliited Claes of predicates; thal sean combine syactcaly withthe outpt of Serivtion: and that te sytactiallyafixed causative morpheme can thes undergo reduplcation wo form a eausve-ntensve. cis pnerally recognized that redupicative morphology operates in the lexicon [Lisber 181, Moravesik 1978) I hiss te cane, then the reduplicatve rl of causative ifensifieation ‘would have 1 ake eet in he lexicon eventhough ts fed by the output of ‘yatactie les. The Orome phenomena discussed inthis paper thus provide ‘Suppor fr the chim made by Dorr 1985 that morpholgia ules ca pote Tay take effect at any stage ina grammatical derivation, Sonsn, Hada. 185. Sync alton a Kel affraton: Te ington for ‘hata angen Pape resented we L3A Apt Meng, Sats no, Mot Leva pvemnest aDaete Fr SPAS. See concepts an consetbences ofthe eco over nnn. ‘cambridge, MA MT ess Dav, WasineD. and Cao. Rese, 19S, Unions 4 mires case. Le ont CM, 194. Th sien Ba anges. London: Oxo Uaiversty Pes, Dona, Sony, 18. Union contin a Japanese: tied andy oe snd rare: Cone Uiversty deren PS Simm eas nant, Con afin wn. Sten in acral abgage an nals tbr.) Dave Reel to apy (Gnas, Gene B. 1, Ove nd coven acres in erating. CLS —ylPtS Orang of Wells. Th non Sec langues of Ethiopia, by Lionel end, 6-5. East Laine Alcan Ses Cner Micha Sac Unncrty frome aitionay Est Lami: aan Sen Cot, Michi Ske ‘unites Howe, Enc F198. Tate and cassie Indo-European, Pagers 00 the Pa ‘hss on Casas and Aety VEL 319), 0, Hoargara, RJ. 196. A question in Gromo merphophtoooe. olla Oretala 17.28 Jconaen, Weiss, 182. Tansy ine Japenese vera system. (Unrest of rs tn Reine iy ey use, Ser 1976, Fetal mbt and finch Ces The Sask sgrntve: Bou the Paso on Dashvns Shane CS Bee Jomo, Dad Pa Poa. 180 Are pra, Pension: Pinel Un Nenu re ‘Lies RHE 1. On the roto telex, MET etton Boo ti ni Us Ca : Loon Aaa, 8 ena of he cause ia Oromo, Journal Anca angus & guts 90 Manas Aine B64 On the ate of amma aes. Cambridge, MA: MT McCaot; ts D. 198, Conversational impact the lexicon. Sys and ‘Semis Pragati by Peer Cai 599, Neu Nan: Senos oe, Morava: Eom 178. Rebiplaivecontsctnns Unc ot ee ae Wel Wond tcc y Jonph Greener, 7-3 San Seed alert res, MongyoatrN0 M199, Granaaicn torso pracca ee lagu Gal. Rome ‘onda Mosca Sano; Anrow Svan: Eowato Ussznacrt and Wourr von Soe. {864 An todito ithe compere paar of he Senta Wis ceo, Harman News ace BM The fen ee cavaive) in Ha. Std in Cac fot Mente psi by & Wowtandl ne fais S526 Res wm owe 1968 A amma Hua Oromo. (Casi Language Sai 4) SS Ta Orc cain: amma witha ee mes, Roi: oun, Di Banal ns, Sy eines 13 Crentap- ‘osetia, dy Eth Morea andes With 15250 Neo ‘cdiemc nes ROE PACL NC Pots, 194, The Ladvncenentexhlvenes aw, Sadie ia ‘elton mere by Dai Primates and Cr Ros B13 Cnc Pusytom Brats Hens 199, Der Kau inden genes Knctaten ‘Spc, Aap eat Rost, Cae. 184, Te eefce between sero le dnl sami 1 This Sake ntl pana cy Darl Porte nl Cael ose, ST Chicge: Uncen Chas Pree Swanre, Maton 16 The amma of usa consustins A cospet "Sass aud Sen The oar of esr enncton: 8.) 8 asm Tone Wank of shina ‘unten: B.A pa of Sac, ohaetere Wits Urvty Yama), Prtori: Un So Paty ce Cove Cag ‘Sowers ar owt

You might also like