V+N miscollocations in the written production of university level students

ELIA 8, 2008, pp. 91-116 V+N MISCOLLOCATIONS IN THE WRITTEN PRODUCTION OF UNIVERSITY LEVEL STUDENTS Magdalena Zinkg ä Uni e sidad Nacional del Comahue, A gen ina maguizinkg a @je band.com.a The s udy o colloca ion – he es ic ed way in which wo o mo e wo ds combine ac oss ex s – has ecen ly gained much ele ance. The acquisi ion o hese combina ions has p o ed an impo an obs acle o non-na i e speake s o a o eign language as e inced in eco ded unusual combina ions o wo ds. This pape analyses nons anda d colloca ions p esen in he w i en p oduc ion o Spanish-speaking uni e si y s uden s o English as a o eign language (EFL), in sea ch o ecu en pa e ns and s a egies ha may explain hese inapp op ia e colloca ions as a way o con ibu e o EFL eaching and lea ning. The esul s show a conside able in luence o lea ne s’ mo he ongue, which induces hem o ansla e li e ally om L1 in o L2 one o bo h elemen s in any gi en colloca ion. Seman ic o e lap be ween app op ia e o ms and possible synonyms o ei he he base o he colloca e appea s o be ano he ac o leading o e o . The s udy also poin s o some e bs which in his pa icula con ex migh equi e consis en and explici eaching. Key wo ds: second language acquisi ion, o eign language eaching, colloca ions, e bs, nouns, da a base, w i en p oduc ion El es udio de la colocación – la combinación es ingida de dos o más palab as en di e sos ex os – ha cob ado mucha ele ancia en las úl imas décadas. La adquisición de es as combinaciones se con ie e en un obs áculo impo an e pa a hablan es no-na i os de un idioma ex anje o, Es udios de lingüís ica inglesa aplicada 92 Magdalena Zinkg ä ELIA 8, 2008, pp. 91-116 quienes combinan palab as de mane a poco usual. Es e abajo analiza las des iaciones colocacionales egis adas en la p oducción esc i a de alumnos uni e si a ios de inglés como lengua ex anje a cuya lengua ma e na es el español pa a explo a es uc u as ecu en es y las es a egias que las gene an, y con ibui , de es a mane a, a la enseñanza y el ap endizaje del inglés como lengua ex anje a. Los esul ados señalan la impo an e in luencia de la lengua ma e na, que impulsa a es os alumnos a aduci , en o ma di ec a, uno o ambos elemen os en las combinaciones del español al inglés. O o ac o que conduce a e o es es la supe posición semán ica en e la base o el colocado en una colocación y posibles sinónimos. El es udio indica ambién la necesidad de una ins ucción explíci a y sis emá ica de cie os e bos en es e con ex o educa i o en pa icula . Palab as cla e: adquisición de una segunda lengua, adquisición de una lengua ex anje a, colocaciones, p oducción esc i a, e bos, nomb es, base de da os. 1. In oduc ion Mul iple s udies ha e explo ed he acquisi ion o colloca ions in lea ne s o a o eign language which ha e p o ed qui e a di icul aspec o he language o lea n, as well as a ell- ale sign o a lea ne ’s non-na i e use o language. The ollowing s udy is based on a manually-compiled co pus o V+N miscolloca ions ga he ed om he w i en p oduc ion o 102 Spanish- speaking s uden s aking an English language cou se in Teache and T ansla o T aining p og ammes a a uni e si y in A gen ina. Analysis o his co pus h ows ligh on he essence o hese de ia ions o a ypical combina ions and he na u e o he mis akes hey e ince. This pape p esen s e idence e ealing an impo an p opo ion o nega i e ans e om hei mo he ongue. V+N miscolloca ions in he w i en … 93 ELIA 8, 2008, pp. 91-116 2. Colloca ion Colloca ions in ol e mo e han andom co-occu ence o wo o mo e lexical i ems ac oss di e en ex s and ac oss di e en gen es, as in make he bed and lay he able. Fo easons o space, no all he di e en de ini ions pos ula ed o he e m will be explo ed he e; su ice i o say ha he e a e as many as he e a e au ho s who ha e esea ched he opic. The issue has been s udied om mul iple heo e ical amewo ks, he mos ele an o which a e B i ish Con ex ualism (Halliday, Sinclai , Hoey, and Phillips), Lexical Seman ics (C use) and di e en b anches o Lexicog aphy. Many dic iona ies ha e been published which include o deal pa icula ly wi h he combina ion o wo ds, and al hough hese will no be commen ed on indi idually, i is wo h men ioning ha bo h he Spanish and he English lexicog aphic mo emen s ha e widely explo ed and analysed he issue o colloca ion in specialised monolingual dic iona ies, some o which a e o ien ed o o eign language lea ne s o hese languages. Wi h he pu pose o es ic ing he possible in e p e a ions o wha a colloca ion is, and o make clea wha is unde s ood by his e m he ein, a numbe o cha ac e is ics will be p esen ed as essen ial o he concep . We shall e e o ‘colloca ion’ in hose cases whe e he e is an a bi a y and syn agma ic link be ween a leas wo lexemes ( e bs, nouns, ad e bs, adjec i es), o example, commi a c ime, bliss ully unawa e; he e is a cons an and ecu en uni y obse ed ac oss mul iple ex s; i s meaning can be in e ed om he elemen s ha compose i , e en hough one o hem may be used in a igu a i e sense (bu no bo h); a ce ain deg ee o a iabili y (subs i u ion) is possible a one o wo poin s in he colloca ion (exe + powe / con ol/ in luence / au ho i y); a ce ain deg ee o g amma ical a iabili y ypical o he syn agma ic ela ion be ween he componen elemen s is allowed (e.g., The powe exe ed by ha g oup was oo s ong); 94 Magdalena Zinkg ä ELIA 8, 2008, pp. 91-116 he appea ance o o he close-class (p eposi ions, a icles, e c.) o open-class wo ds ( e bs, nouns, ad e bs, adjec i es) be ween he elemen s in he colloca ion does no a ec he link be ween hem (e.g., he exe ed all his au ho i y o…); a ce ain deg ee o a ia ion ela ed o pa s o speech is allowed (i he e exis s a link such as colloca ion be ween a e b and a noun, he colloca ion is s ill accep able i he e b is nominalised, e.g., an exe ion o powe ); he combina ion e inces s o ed knowledge in he memo y o na i e speake s o a language as pa o he meaning o any wo d ( ha i colloca es wi h o he s like x o y); he exp ession o ideas and in e ac ion be ween speake s is aided ough p e-selec ion o lexical al e na i es; he combina ions p esen a ce ain speci ici y, i.e., a pa icula seman ic bond be ween he meanings o he lexical i ems con eyed h ough he choice o wo ds (all nouns accompanying he e b o exe sha e a seman ic associa ion wi h he concep o powe ). These cha ac e is ics may help in he ecogni ion o a colloca ion, ypically classi ied in o wo b oad g oups: lexical and g amma ical. Those combina ions which in ol e wo open-class wo ds co espond o he o me , while he la e combine an open-class wo d and a pa icle o p eposi ion. This pa icula s udy ocuses on one ins ance o lexical colloca ions, namely hose in ol ing a e b and a noun which pe o ms he unc ion o i s di ec objec , o in e nal a gumen . 3. O e iew o S udies on he Acquisi ion o Colloca ion in Fo eign Language Se ings Many au ho s ha e in es iga ed he puzzling issue o colloca ion in a o eign language se ing, basically awa e o he ac ha compe ence in his espec enables lea ne s o pe o m almos as p o icien ly as na i e speake s o he V+N miscolloca ions in he w i en … 95 ELIA 8, 2008, pp. 91-116 language (Paquo , 2005). The main conclusions d awn om a ious s udies in o he acquisi ion o English colloca ions in lea ne s o di e en L1 backg ounds a e p esen ed below 1 : Polish lea ne s o English success ully ex ended he meaning o L2 colloca ions on he basis o an equi alen combina ion in L1, whe eas Ge man lea ne s o English made swi ching mis akes and blends be ween L1 and L2 (Biskup, 1992). The wide he meaning scope o a lexical i em, he mo e he in e e ence o L1 in he colloca ions ha lexical i em pa icipa ed in. Simila ly, he mo e synonyms a wo d had, he mo e di icul y lea ne s ound o p oduce a es ic ed colloca ion (Biskup, 1992). Ad anced EFL lea ne s’ colloca ional compe ence did no ma ch hei gene al compe ence in he o eign language; lea ne s ound i di icul o pa aph ase colloca ions hey could no p oduce (Bahns & Eldaw, 1993). Ad anced EFL lea ne s we e insu icien ly awa e o he phenomenon o colloca ion in gene al (Bahns & Eldaw, 1993; Fa ghal & Obieda , 1995). Fa ghal and Obieda (1995) claim ha hei A abic lea ne s “ end o espond o he open choice p inciple (…) a he han he Idiom P inciple 2 ” (p. 326), making use o lexical simpli ica ion h ough synonymy, a oidance, ans e and pa aph asing. The success ul use o o e lapping colloca ions (which combine “appa en openness a iabili y- wi h ac ual es ic ion”) and he subsequen p oduc ion o clus e s o e b-noun combina ions could be a ma ke o p o iciency among na i e w i e s and p esen di icul ies o non-na i e w i e s (Cowie & Howa h, 1995, p. 84). Typically, lea ne s o English ended o use he mos open o ee ype o colloca ions, which p esen ed almos no challenge o hei ph aseological compe ence, a oiding he ype o language hey migh conside p oblema ic (Cowie & Howa h, 1995, p. 89). 96 Magdalena Zinkg ä ELIA 8, 2008, pp. 91-116 Lea ne s in Cowie and Howa h’s expe imen eso ed o he ex ension o a lea n -by-hea es ic ed colloca ion used as he basis o new, inapp op ia e combina ions. The p ocess o combining he igu a i e senses o abs ac e bs and nouns in academic w i ing was pe cei ed as c ea i e by na u e, and he e o e, p oduc i e 3 , which led o miscolloca ions. Non-na i e F ench speake s o English ended o eso o some ad e bial ampli ie s in excess in compa ison wi h na i e speake s o he language, as a c ea i e s a egy o minimise he chances o making mis akes because he app op ia e ad e bial combina ions we e no salien o non- na i e speake s o English, i.e., hey did no pe cei e hem as colloca ions (G ange , 1998). Non-na i e speake s o English made use o epe i ion when hey ound hei colloca ional knowledge insu icien o exp ess in ended meanings (Huang, 2001). Japanese lea ne s o English u ned o a g ea ex en o delexical e bs (make, do, ha e, pu ), and as hei le el o gene al p o iciency in he o eign language inc eased, he nominal ph ase ha colloca ed wi h he selec ed e b go mo e complex in e ms o modi ica ion (Ba ield, 2001). Japanese lea ne s ended o uni o mly mis ake alse e b-noun colloca ions (expe imen al dis ac o s) o eal colloca ions (Ba ield, 2001). F ee combina ions we e easie o comple e and ansla e by Taiwanese lea ne s o English han pu e idioma ic exp essions (which allow li le o no a ia ion), leading o a g ea numbe o inco ec a ian s (Huang, 2001, p. 9). Dis ance be ween L1 and L2 esul ed in mo e incomple e colloca ional knowledge and nega i e ans e (Huang, 2001; Mahmoud, 2005). When he e was insu icien knowledge o he e b componen o a colloca ion, ecogni ion o he colloca ion was impeded (Ba ield, 2003), V+N miscolloca ions in he w i en … 97 ELIA 8, 2008, pp. 91-116 indica ing ha e bs play a de e minan ole in he p oduc ion o hese colloca ions. Taiwanese college EFL lea ne s’ use o lexical colloca ions posi i ely co ela ed wi h hei o e all online w i ing luency (Hsu, 2007). Mos o hese au ho s ha e done esea ch in o he acquisi ion o English colloca ions by lea ne s o di e en linguis ic backg ounds and ag ee on he ypical s a egies ha may gi e ise o awkwa d combina ions, namely epe i ion and eliance on he belie ha es ic ed combina ions a e ee . Lea ne s may also o e gene alise he meaning scope o hose synonymous i ems pa icipa ing in a es ic ed combina ion o eso o o e simpli ica ion and ans e o colloca ions om L1. Howe e , he e is no eco d o such a s udy in Romance-language-speaking lea ne s o English. In he ollowing sec ion, an analysis o he p oduc ion o a ypical V+N colloca ions by Spanish lea ne s o English is p esen ed. 4. The S udy The s udy consis ed o wo phases: i s ly, he esea che was o c ea e a co pus o Ve b-Noun miscolloca ions eco ded in ad anced EFL lea ne s’ w i en assignmen s a an A gen inean uni e si y; secondly, he co pus was o be manually analysed in sea ch o ypical pa e ns ha would e eal he sou ce o hese e o s and he unsuccess ul, compensa o y s a egies ha hese lea ne s make use o when hey need o con ey meanings ha in ol e colloca ions. 4.1. Pa icipan s, Da a-Collec ion P ocedu es, and Da a Desc ip ion Pa icipan s in his esea ch in ol ed 102 lea ne s o English aking ei he he English IV o English V cou ses du ing he academic yea 2003/ 04 wi hin he Teache and T ansla o aining p og ams a he Uni e sidad Nacional del Comahue, A gen ina, who ag eed o submi hei w i en wo k o inspec ion and analysis. The le el o p o iciency o hese lea ne s in he o eign language was assessed h ough X_Lex: The Swansea Le els Tes s, a 98 Magdalena Zinkg ä ELIA 8, 2008, pp. 91-116 ocabula y ecogni ion es (Mea a, Mil on, & Lo enzo-Dus, 2005) ha “measu es s uden s’ lexical and s uc u al ocabula y knowledge a he mos equen le els in English” (Mea a, Mil on, & Lo enzo-Dus, 2005, p. 1) 4 . The mean adjus ed sco e ob ained was 4188.56 o English IV pa icipan s, and 4612.12 o English V lea ne s, espec i ely (s anda d de ia ion: 420.37 in bo h cases). The i s g oup’s p o iciency le el is cha ac e ized by he p og am as ”compe en ” and “in e media e le el o pe o mance”, and is equalled o ha o a CAE cou se in he es desc ip o (UCLES exams) o he 3750-4240 ange. The mo e ad anced g oup o lea ne s in his s udy is desc ibed as ha ing a “ e y good le el o pe o mance” co esponding o CPE (UCLES exams) wi h sco es anging om 4500 o 4740. The da a we e collec ed om 13 di e en p ac ical assignmen s, including eading comp ehension asks, essays and e iews (a o al o 270,170 wo ds). Table 1 shows he dis ibu ion o hese asks ac oss cou ses as well as he opics explo ed in each o hem. Table 1 Da a collec ion ins ances by ask- ype and cou se Ac i i y Type Topic Cou se Reading comp ehension + essay Fashion IV Reading comp ehension + essay Swimming IV P oposal Menus a he can een IV Reading comp ehension + summa y Maxwell & he laws o Elec omagne ics IV Reading comp ehension + essay Child-ca e cen es IV Essay Tele ision and iolence IV Reading comp ehension + essay Oes ogen: science and na u e IV V+N miscolloca ions in he w i en … 99 ELIA 8, 2008, pp. 91-116 Ques ions on a no el O yx and C ake IV Reading comp ehension + summa y Mo he kind V Repo A es i al V Reading comp ehension + summa y Mu phy’s Laws V Re iew (s uden s’ choice) V Reading comp ehension + essay The Mayas V To build he co pus, all ins ances o possible inapp op ia e e b- noun colloca ions ollowing he basic s uc u e in (1) we e ansc ibed and analysed - he b acke s indica e he possible omission o elemen s. (1) [(De ) + (Adj) + N] + V + (De ) + (Adj) + N This p ocedu e was ollowed by he eache o he cou se and an ex e nal e alua o . Examples like hose in (2) and (3) cons i u ed a la ge pa o he co pus. (2) [The emale body] emi ed [ e e sal clea (sic) signs] (3) [you] a e aining [you muscles] O he g amma ical ealisa ions o he same pa e n, such as passi e o ms (4), a nominal ph ase modi ied by a ( educed) ela i e clause (5), a o- in ini i e pu pose ad e bial ph ase con aining a ansi i e e b (6), bo h ge unds (7) and o he nominalisa ions o ansi i e e bs accompanied by hei di ec objec s, we e also included in he da a-base. 106 Magdalena Zinkg ä ELIA 8, 2008, pp. 91-116 elabo a e s uc u es, while on he o he , poin o a possible eason o he o ma ion o hese a ypical combina ions. When he canonical o de is in e ed o when he sequence is in e up ed by a p e-modi ie o he noun, lea ne s may be led o choose an inco ec elemen o he in ended colloca ion, echoing Nesselhau ’s (2005) discussion o syn ac ic pa e n and i s co ela ion wi h lea ne s’ di icul ies wi h colloca ions. 5.4. Mos F equen Ve bs The analysis o he esul s so a has cen ed on he equency and s uc u al cha ac e is ics o he combina ions as a whole in he co pus. In his sec ion, howe e , he occu ence o e bs pa icipa ing in he miscolloca ions is analysed. Th ough s a is ical analysis i has been es ablished ha delexical e bs ha e been ex ensi ely in ol ed in he a ypical combina ions o he co pus. Mo e han a qua e o he o al colloca ions in he da a (28%) in ol es one o he ollowing delexical e bs: do, make, ake, ha e, p oduce and hold, all o which ha e also been eso ed o by Japanese lea ne s o English in he inco ec o ma ion o V+N colloca ions (Ba ield, 2003). In his s udy Ba ield e lec s on lea ne s’ endency o use a delexical e b plus a nominaliza ion ins ead o he mo e equen combina ion ound in na i e speake s’ use o a ull, lexical e b eplacing he odd V+N cons uc ion. Examples in he co pus analysed in his pape a e: (8) his disco e ies we e no ecognised a he ime he did hem (9) he abulous con ibu ion he did o mankind These ins ances show an awkwa d ela ionship be ween e b and noun which could easily be sol ed i lea ne s had di ec ly eso ed o he e bs disco e and con ibu e. Fu he analysis o he equency dis ibu ion o e bs in he co pus e inces a endency o employ e y common e bs in English. Table 4 p esen s six speci ic e bs which appea in mo e han 6 miscolloca ions V+N miscolloca ions in he w i en … 107 ELIA 8, 2008, pp. 91-116 each, and o each o hem, he numbe o epe i ions ound in he co pus (Rep) acco ding o he S a is ix 8.0 es o equency dis ibu ion, as well as hei equency a ing in he Gene al Se ice Lis (GSL) 8 and in he B i ish Na ional Co pus (BNC) 9 o compa ison. Table 4 Dis ibu ion o he mos equen e bs Make and do a e by a he mos equen delexical e bs in ol ed in miscolloca ions, each epea ed 24 and 19 imes espec i ely. Many o he “e o s” in his da abase a e due o w ong choice o delexical e bs, especially his minimal pai , and could ha e been a oided i lea ne s had used he o he op ion. Thus, in (10) […] hey do a se ies o changes he combina ion becomes accep able i do is changed o make. Wha appea s bo h su p ising and wo ying is ha hese lea ne s ha e o he wise eached an ad anced le el in hei gene al p o iciency o he o eign language, as ce i ied by he sco es hey ob ained in he Swansea Le els Tes X-Lex (Mea a e al., 2005). These mis akes, howe e , poin o an impo an lack in hei colloca ional compe ence in ol ing mas e y o e y basic e bs, like make and do. Table 5 p esen s mo e e y equen e bs, hough hese ha e been epea ed ewe han i e imes, as calcula ed h ough S a is ix 8.0 es o equency dis ibu ion. These ha e been classi ied as “conside ably equen ” in opposi ion o he ones abo e, which pa icipa e in mo e han 6 colloca ions in he co pus. Ve b Rep GSL BNC Ve b Rep GSL BNC Ve b Rep GSL BNC do 24 98 18 make 19 37 46 Imp o e 9 604 865 hold 9 172 197 use 7 65 92 P ac ice 6 - 2658 108 Magdalena Zinkg ä ELIA 8, 2008, pp. 91-116 Table 5 Ve b Rep GSL BNC Ve b Rep GSL BNC Ve b Rep GSL BNC p oduce 4 444 332 ake 4 60 54 de elop 4 130 411 lose 4 417 362 ha e 3 9 8 enhance 3 - 2666 achie e 3 - 611 p o e 3 687 691 Design 3 262 11810 ain 3 - 855 ace 3 190 599 A ca e ul look a he equency a ing o he e bs in ables 4 and 5 in GSL and BNC p o ides an in iguing insigh in o he use o hese e bs. Mos o hese cases a e no only equen in he co pus o miscolloca ions analysed he e, bu a e all he mo e so in he language, mos o hem belonging o he 1,000 mos equen wo ds in English. The ac ha hese e y common wo ds in English appea in a ypical colloca ions p oduced by ad anced lea ne s o he o eign language is e idence o he une en de elopmen o lea ne s’ gene al linguis ic and colloca ional compe ences. As has been s a ed elsewhe e, Bahns and Eldaw (1993), Ba ield (2003) and Fa ghal and Obieda (1995) ha e ound his o be he case in hei in es iga ions. The da a in his s udy indica e ha he e bs used a e o so basic a na u e ha i migh be assumed ha hese lea ne s a e a an elemen a y le el in e ms o hei handling o colloca ion. On he o he hand, he da a migh also be in e p e ed as e idence o a s a egy lea ne s eso o when aced wi h he ac ha hey a e no con iden in he use o he al e na i e lexical e b ha would o he wise comp ise he meanings o bo h he delexical e b and he in ended noun. This pa icula s a egy has been ound o be widesp ead among FL lea ne s ac oss he s udies e iewed abo e, and a e y use ul one in ha hese delexical e bs can be ypically accompanied by a g ea - and un es ic ed, in s uden s’ eyes - numbe o nouns. Unde he assump ion ha hese e bs lack a speci ic meaning, lea ne s o e -gene alise and combine hem wi h any noun unde he illusion ha he e is no es ic ion o he way hey can be used. These indings migh a es o lea ne s’ eliance on Sinclai (1991)’s Open Choice P inciple. Acco ding o Huns on (2002), e iewing Sinclai ’s wo k on his issue, V+N miscolloca ions in he w i en … 109 ELIA 8, 2008, pp. 91-116 meaning [o any g oup o sequence o wo ds] is made ei he by he ph ase as a whole, ope a ing in acco dance wi h he con en ional ph aseology, o (less o en) i is made by he indi idual wo ds, ope a ing in acco dance wi h g amma ical ules (p. 145). The esul s o his s udy seem o show ha i Spanish lea ne s’ knowledge o English colloca ions and hei inhe en es ic ions a e incomple e, hey will cons uc meaning by adding up meanings o indi idual wo ds (The Open Choice P inciple), igno ing he “whole ph ase” p inciple (The Idiom P inciple). In hei s udy o A abic lea ne s’ colloca ional compe ence, Fa ghal and Obieda (1995) ha e eached simila conclusions in his espec . Al hough in e es ing insigh s can be gained h ough he analysis o noun equency in he s udy o miscolloca ions, and in e ms o seman ic clus e ing o bo h nouns and e bs, o easons o space hese aspec s will no be discussed he ein (see Zinkg ä , 2005). Ins ead, he issue o he impac o lea ne s’ L1 on hese a ypical combina ions will occupy he ollowing sec ion. 5.5. The Role o he Mo he Tongue To u he in es iga e he mo i es o he a ypical colloca ions in he co pus, each elemen in he combina ions ( e b and noun) was gi en a ansla ion in o he mo he ongue by he esea che . In o de o ensu e he alidi y o hese ansla ions, ou ex e nal a e s whose mo he ongue is also Spanish and whose pe o mance in English is quasi na i e we e asked o accep o ejec he gi en ansla ion o he elemen s in each miscolloca ion. These a e s’ assessmen s esul ed in 61% o hese a ypical V+N colloca ions being uni o mly accep ed as gene a ed by nega i e ans e om L1. The esul s, which echo Hoey’s (2005) p edic ions o L2 acquisi ion o colloca ions, coincide wi h hose ob ained by Bahns (1993), Biskup (1992), Bonci (2004), Fa ghal and Obieda (1995), G ange (1998), Mahmoud (2005) and Ma yńska (2004) as ega ds he impac o he mo he ongue, in his pa icula case, Spanish, on he p oduc ion o inapp op ia e combina ions be ween e bs and hei in e nal a gumen s in lea ne s o English. 110 Magdalena Zinkg ä ELIA 8, 2008, pp. 91-116 A u he s ep in he explo a ion o his issue was aimed a de e mining which cons i uen is esponsible o he de ia ion in hese cases. To his pu pose, each o he miscolloca ions in he da a base was p o ided wi h an accep able al e na i e o exp ess he in ended meaning. Fo ins ance, o he combina ion [ex inguish] + [disease] in (12) [...] so as o p e en , cu e, o ex inguish a disease (In . 20) he dic iona ies consul ed o e he lemma ERADICATE as an accep able colloca e o DISEASE. This p ocedu e was epea ed wi h e e y miscolloca ion in he co pus o assess whe he he noun, he e b o bo h in he o iginal e sion needed o be modi ied. 70.5% o he a ypical combina ions in he co pus can be made accep able by eplacing he o iginal e b wi h one sugges ed by he lexicog aphical wo ks consul ed. On he o he hand, he combina ions ha could be made igh by changing he o iginal noun only ep esen 19%. In 9% o he cases i was ha d o de e mine which elemen had been selec ed inco ec ly, o wha message o idea had been in ended by he in o man s in he i s place. These da a indica e ha e bs a e mainly esponsible o he inco ec ness in hese a ypical V+N colloca ions, while showing ha pa icipan s in his in es iga ion ha e insu icien knowledge o he igh colloca es ( e bs) he bases (usually nouns) o he colloca ions ypically ake. Once again i is wo h no icing ha ou o he 70.5%, a signi ican numbe o w ong e bs belonged o he delexical ca ego y, and he “mis ake” could be co ec ed by selec ing i s coun e pa (e.g., make s. do). The esul s e ince an in e media e s age in he acquisi ion o colloca ions, since he e is e idence ha p o es hese lea ne s lack he linguis ic ools o p oduce es ic ed colloca ions, which in u n migh lead o hea y eliance on L1 o p oduc ion o V+N combina ions. The pa icipan s in his in es iga ion ha e p oduced colloca ions ha in ol e delexical e bs o an unusually impo an ex en as compa ed o na i e speake s’ pe o mance, e en choosing he w ong one among a se o possibili ies. Wha appea s signi ican ly wo ying is he ac ha he miscolloca ions V+N miscolloca ions in he w i en … 111 ELIA 8, 2008, pp. 91-116 eco ded in ol e ideas, concep s and ela ions ha a e o common use in basic exchanges in he o eign language. 6. Conclusions The analysis o he da a base ob ained h ough 13 w i en assignmen s submi ed by 102 Spanish-speaking lea ne s o English a he Uni e sidad Nacional del Comahue, A gen ina, shows he di e en s a egies hey ypically eso o wi h espec o hei colloca ional compe ence. Mos o he de ia ions a e he esul o nega i e ans e om hei mo he ongue, (61%) o , in some o he cases, o he exagge a ed use o inapp op ia e delexical e bs. The w ong choice o bo h nouns and e bs al e na i ely is ecu en in hese a ypical colloca ions, esponsible o 70% o he miscolloca ions in he co pus. O he usual s a egies employed a e epe i ion o one same e b combined wi h nouns which do no colloca e, simpli ica ion e inced in he eliance on he Open Choice P inciple when lea ne s igno e he es ic ions some combina ions a e subjec o, and ex ension by analogy, in hose cases whe e hey assume ha i a noun sha es meaning elemen s wi h ano he , hey bo h colloca e wi h he same e b. The esul s also poin o ecu en pa e ns in he s uc u es lea ne s use o con ey he e b-noun ela ion. These conclusions indica e a clea misma ch be ween lea ne s’ gene al compe ence in he o eign language and hei colloca ional compe ence, simila o he esul s in Schmi e al. (2004) wi h espec o a possible co ela ion be ween size o he ‘indi idual wo d lexicon’ and he ‘ o mulaic sequence lexicon’ (p. 64). Al hough he da a and conclusions he ein p esen ed ha e li le po en ial o gene alisa ion o wide educa ional con ex s, many o he esul s ob ained may be made ex ensi e o simila si ua ions in which ad anced lea ne s o a o eign language p esen limi a ions as ega ds he w i en p oduc ion o e b-noun colloca ions. The insigh s o e ed by his s udy poin o an impo an need o he de elopmen o hese lea ne s’ colloca ional compe ence h ough speci ic awa eness- aising ac i i ies as o he es ic ed na u e o he phenomenon in ques ion. Mo eo e , as some au ho s sugges (Lewis, 2000; Ba ield, 2003), special emphasis should be placed on hose a eas whe e L1 and L2 di e in 112 Magdalena Zinkg ä ELIA 8, 2008, pp. 91-116 meanings and colloca ions. Bea ing his linguis ic dis ance in mind should help eache s an icipa e possible occasions o nega i e ans e , and hus selec he combina ions o be wo ked on in he English class. In addi ion, a poin should be made o ocussing on colloca ions in o eign language eaching om he e y beginning a e y ea ly s ages in he lea ning p ocess. Thus, a basic colloca ional compe ence could be de eloped, and on i s basis, mo e es ic ed and complex meanings buil and d awn a en ion o in subsequen , mo e ad anced cou ses. Enabling lea ne s o pe cei e language as being made up o ecu en g oups o wo ds should be one o a eache ’s mos impo an aims and migh e en ually help s uden s in he p oduc ion o he o eign language, leading hem o ely mo e on he Idiom P inciple han on he wo d-by-wo d cons ual o meaning, hus esembling a na i e speake ’s na u al pe o mance in he use o he language. V+N miscolloca ions in he w i en … 113 ELIA 8, 2008, pp. 91-116 No es 1 These indings ha e been a anged ollowing a ch onological o de . 2 Re e ence o Sinclai (1991)’s Idiom P inciple and Open Choice P inciple is made in sec ion 4.4. 3 See Sinclai (1991)’s Idiom P inciple and Open Choice P inciple in sec ion 4.4. 4 This compu e p og am p o ides a ocabula y p o ile in e ms o i e di e en le els o ocabula y knowledge. An adjus ed sco e deduces he numbe o pseudowo ds om he o al sco es, hus aking es - ake s’ guessing in o accoun . 5 'de ' should be unde s ood as 'de e mine ' 6 Consul ed a www.collins.co.uk/Co pus/Co pusSea ch.aspx 7 Compu e p og amme downloadable om h p://unde na ion.com/c ack-s ep7- mic o/c ack-S a is ix+8.0-se ial-keygen- ee- ull-download- o en .h ml 8 The Gene al Se ice Lis is a lis o equen wo ds in English and can be consul ed a h p://jbauman.com/gsl.h ml 9 A hie a chy o he mos equen lemmas in his co pus can be consul ed a p:// p.i i.b on.ac.uk/bnc/lemma.al 114 Magdalena Zinkg ä ELIA 8, 2008, pp. 91-116 Re e ences Alonso Ramos, M. (1994-1995). Hacia una de inición del concep o de colocación: de J. R. Fi h a I. A. Mel’cuk’. Re is a de Lexicog a ía, 1: 9-28. Bahns, J. (1993). Lexical colloca ions: A con as i e iew. ELT Jou nal 47(1): 56-63. Bahns, J. & Eldaw, M. (1993). Should we each EFL s uden s colloca ions? Sys em 21(1): 101-114. Ba ield, A.W. (2001). Colloca ion design and de elopmen . Pape p esen ed a he 11 h Vocabula y Acquisi ion Resea ch G oup Ne wo k Con e ence, Swansea Uni e si y, UK. Ba ield, A. (2003). Colloca ion ecogni ion and p oduc ion: Resea ch insigh s. Tokyo: Chuo Uni e si y, Japan. Biskup, D. (1992). L1 in luence on lea ne s' ende ings o English colloca ions: A Polish/Ge man empi ical s udy. In P.J.L. A naud, & H. Béjoin (Eds.), Vocabula y and applied linguis ics (pp. 85-93). London: Macmillan. Bonci, A. (2004). Colloca ional es ic ions in I alian as L2: The ele ance o au hen ic language use o language lea ning [On-line]. A achno iles, Vol.3 Hos name: selc.ed.ac.uk Di ec o y:a achno iles/pages/ h ee_bonci.h m Cowie, A.P. & Howa h, P. (1995). Ph aseological compe ence and w i en p o iciency. In G.M. Blue & R. Mi chell (Eds.), Language and educa ion. B i ish s udies in applied linguis ics 11 (pp. 80-93). Cle edon: Mul ilingual Ma e s. Fa ghal, M. & Obieda , H. (1995). Colloca ions: A neglec ed a iable in EFL. In e na ional Re iew o Applied Linguis ics 28(4): 315-331. V+N miscolloca ions in he w i en … 115 ELIA 8, 2008, pp. 91-116 G ange , S. (2001). P e ab ica ed pa e ns in ad anced EFL w i ing: Colloca ions and lexical ph ases. In A.P. Cowie (Ed.), Ph aseology: Theo y, analysis, and applica ions (pp. 145-160). Ox o d: Cla endon P ess. Hoey, M. (2005). Lexical p iming. A new heo y o wo ds and language. Oxon: Rou ledge. Hsu, J. (2007). Lexical colloca ions and hei ela ion o he online w i ing o Taiwanese college English majo s and non-English majo s. Elec onic Jou nal o Fo eign Language Teaching [On-line se ial], Vol. 4 (2): 192-209, Di ec o y: e- l .nus.edu.sg/ 4n22007/hsu.h m Huang, L. (2001). Knowledge o English colloca ions: An analysis o Taiwanese EFL lea ne s. Texas Pape s in Fo eign Language Educa ion [On-line se ial], Vol. 6, Nº1. Hos name: u exas.edu Di ec o y: lesa/ p le/con en s7.doc Huns on, S. (2002). Co po a in applied linguis ics. Camb idge: Camb idge Uni e si y P ess. Lewis, M. (Ed.). (2000). Teaching colloca ion: Fu he de elopmen s in he lexical app oach. Ho e: L.T.P. Mahmoud, A. (2005). Colloca ions e o s made by A ab lea ne s o English. Asian EFL Jou nal [On-line se ial], Vol. 5. Hos name: asian-e l- jou nal.com Di ec o y: p a_Augus _05_ma.php Ma yńska, M. (2004). Do English language lea ne s know colloca ions? In es iga iones linguis icae [On-line se ial], Vol.XI, Poznán. Hos name: s a .amu.edu.pl Di ec o y: ~in eling/index.php?di ec =7 Mea a, P., Mil on, J. & Lo enzo-Dus, N. (2005). The Swansea le els es X_lex,Lognos ics CD-Rom [Compu e so wa e]. Exp ess Publishing. Nesselhau , N. (2005). Colloca ions in a lea ne co pus. Ams e dam: John Benjamins.