Grammar instruction and the acquisition of gustar-type verbs by English-speaking learners of Spanish

ELIA 4, 2003, 255-279 GRAMMAR INSTRUCTION AND THE ACQUISITION OF GUSTAR-TYPE VERBS BY ENGLISH-SPEAKING LEARNERS OF SPANISH María Dolores López Jiménez Universidad de Sevilla Spanish gustar-type verbs form part of a group called psych verbs (Belletti and Rizzi 1988). These verbs pose potential learnability problems for English-speaking learners of college-level Spanish since the most frequent and unmarked word order with gustar-constructions is OVS in contrast to the obligatory SVO pattern in English. This study addresses two questions: a) Does instruction promote the acquisition of gustar-type verbs? b) If so, are there task effects1? A total of 24 upper-level beginners (first year, second semester) of Spanish as an FL participated in this study: 12 formed part of the treatment group and 12 were included in the control group. Two threepart tests were administered in a 3-week period. The preand post-teaching tests consisted of a multiple-choice task, a scrambled sentences task and a free production task, and were distributed during the first and third week, respectively. Teaching which consisted of grammar explanation and practice took place the third day of class of the first week. A total of six days of class elapsed between both tests. Results indicated that the treatment group outperformed the control group in the scrambled sentences and the multiplechoice tasks, but not in the free production task in the post-teaching test. A 1 This term is used to refer to the variability in language use evident when learners are asked to perform different tasks. Estudios de lingüística inglesa aplicada 256 M. D. López Jiménez ELIA 4, 2003, pp. 255-279 task effect was found: production of target-like forms by both the experimental and control groups decreased as production became less controlled. Key words: Psych verb, focus on form vs. focus on formS instruction, task effect, cyclical syllabus. 1. Introduction The incentive for this study is twofold: first, gustar-type verbs present a recurrent problem to English-speaking learners of Spanish; second, not much research has been done on the acquisition of this type of verb (cf. Montrul 1997). The Spanish gustar-type verbs form part of a group of verbs called psych verbs. Apparently, they show arbitrary mappings between their semantics (thematic roles) and syntax (syntactic positions) (Montrul 1997). These verbs contain a theme and an experiencer. The syntactic position of subject may be occupied by an experiencer or a theme. According to Belletti and Rizzi (1988), there are three classes of psych verbs in Italian: 1) the temere (“fear”) class: Gianni teme questo, “Gianni fears this”; 2) the preoccupare (“worry”) class: Questo preoccupa Gianni, “This worries Gianni”; 3) the piacere (“like”) class: A Gianni piace questo,“To Gianni pleases this”. Class I psych verbs (“fear” in English) are transitive: the experiencer appears in the subject position, whereas the theme takes the object position. Class II (“frighten”) presents the opposite mapping; in this case the subject and the object are performed by a theme and an experiencer, respectively. Class III are, according to Belletti and Rizzi (1988), unaccusatives. The experiencer appears in subject position but takes the case marking of an indirect object. The verb requires agreement with the theme, which remains in an object position. Either the experiencer or the theme may appear in Grammar instruction and the … 257 ELIA 4, 2003, pp. 255-279 preverbal or postverbal position, a unique characteristic of this third class of psych verbs. English lacks this third class. In Spanish, dative experiencers are used with gustar-type verbs (Italian piacere “to like”), that is, fascinar “to fascinate”, importar “to be important, to matter”, alegrar “to make someone happy”, convencer “to convince”, disgustar “to displease”, molestar “to bother”, just to name a few. According to Montrul (1997: 192) there are arguments in favor of experiencers being treated either as objects (indirect object) or subjects of the verb: “arguments in favor of experiencers as objects rest almost exclusively on the oblique case marking. By contrast, arguments for subjecthood are supported by their behavior with binding and control”. In this study, gustar-type verbs are treated as follows: 1) (A mí) me gustan las manzanas indirect object (experiencer) subject (theme) “I like apples” As indicated by the parenthesis, the prepositional phrase is optional. Its presence indicates an emphatic or clarifying function. However, the presence of the indirect object is always required; otherwise, the gustar-type construction becomes ungrammatical. Spanish gustar-type verbs always take the third person (singular/plural) and allow the theme to be either in preverbal or postverbal position, whereas the indirect object is always preverbal: 2) a. (A mí) me gustan las manzanas; b. Las manzanas me gustan (a mí); c. *Las manzanas gustan me. 2. Goals The purpose of this study is to investigate the effects of instruction on the acquisition of Spanish gustar-type verbs by native speakers of English at the college level. These verbs pose potential learnability problems because the most frequent and unmarked word order for this type is OVS. This order 258 M. D. López Jiménez ELIA 4, 2003, pp. 255-279 differs from the typical subject-verb-object order in Spanish or from the obligatory subject-verb-object (SVO) pattern in English. VanPatten’s (1984) study of 59 English learners of Spanish showed that noun-verb-noun (NVN) is interpreted as agent-action-object. Thus, when students are given scrambled sentences like a mí/gustar/las buenas notas “To me to-like good grades”, they tend to convert them into Yo gusto las buenas notas “I like good grades”, a non-acceptable construction in Spanish. This study explores the following questions: a) Does instruction promote the acquisition of gustar-type verbs? b) Are there task effects? A number of studies have been carried out in order to investigate the efficacy of instruction, comparing instructed learners who were under an experimental treatment with those who were not. Cadierno’s (1995) study investigated the efficacy of traditional form-focused instruction as opposed to processing instruction. A total of 80 subjects, enrolled in second-year university-level Spanish classes at the University of Illinois, participated in this experiment. Traditional instruction consisted in presenting the learners with explanations about the form and position of direct object pronouns within the sentence and giving practice on how to create sentences with these pronouns. Processing instruction involved teaching the learners how to interpret OVS strings and having the participants respond to the content of OV strings. The pretest and the three posttests contained both interpretation tasks and written production tasks. The results revealed that subjects who received processing instruction showed gains in comprehension and production, whereas those who experienced traditional instruction improved production only. Grammar instruction and the … 259 ELIA 4, 2003, pp. 255-279 3. Methodology 3.1. Participants A total of 24 upper-level beginners (first year, second semester) of Spanish as an FL participated in this study: 12 formed part of the treatment group (subjects #1 to #12) and 12 were included in the control group (subjects #13 to #24). The subjects chose their own nickname to identify themselves in the different parts of the testing instruments. 3.2. Instrument The material consisted of two three-part tests: a pre-teaching and a postteaching test. The three parts of each test consisted of a scrambled sentences task, a multiple-choice task, and a free written production task. The scrambled sentences task contained fourteen sentences to be done in ten minutes. Each indirect object pronoun (me “me”(1sg.), te “you” (2sg.), le “him/her” (3sg.), nos “us” (1pl), les “them”(3pl)) occurred twice, except for the third person singular and plural indirect object pronouns, which presented four occurrences each (eight total). Instances of the third person indirect object pronoun were doubled because teachers observing students’ homework suggested that le and les tend to be more problematic from an acquisitional point of view. They are superficially quite similar compared to the opposition me (1sg.) versus nos (1pl.). The second person plural indirect object pronoun, os, was not included since this form is not taught in S150 (first year, second semester of Spanish, that is, upper-level beginners). The only clue that the students had to produce the indirect object pronoun was a prepositional phrase (e.g. a mí “to me”, a mis amigos “to my friends”): Preocupar/ a mí/ las emisiones de los coches viejos “Worry/to me /the emissions of old cars”. The presence of this prepositional phrase is optional in gustar-type constructions, though that is not the case for the 260 M. D. López Jiménez ELIA 4, 2003, pp. 255-279 dative pronoun. The combination of all the indirect object pronouns was distributed in such a way that instances of the same pronoun did not appear in two consecutive sentences. The multiple-choice task contained ten sentences that provided the context for the multiple-choice answers. The sentences were constructed such that neither an indirect object pronoun nor a gustar-type verb was present (e.g: Los profesores de historia tienen mucho interés en las pirámides de Egipto, “Professors of History are very interested in Egyptian pyramids”). Underneath each sentence there were four possible answers with this structure: indirect object pronoun (singular/plural) plus a gustar-type verb (singular/plural). When the third person indirect object pronoun (singular/plural) and a gustar-type verb (singular /plural) were tested at the same time, all four choices contained a third person form. Each pronoun (me, te, nos) appeared in two of the four choices, whereas in the other two options a third person (singular/plural) indirect object pronoun was used. The verb of each option was the same, varying in number (singular/plural) only, e.g.: Yo estoy nerviosa por mi examen (“I am nervous because of my exam”) a. Les preocupa (“It worries them”) c. Le preocupa (“It worries him/her”) b. Me preocupan (“They worry me”) d. Me preocupa (“It worries me”) The verb number of each le/les combination was decided upon the verb number of the correct answer. The free production task consisted of three passages. For each of the three passages the students had to report the author’s belief about ecology using a gustar-type verb (three sentences total). There were also two lists of gustar-type verbs and a glossary provided at the end of the three readings. The reason for presenting the three tasks in this order, scrambled sentences, multiple-choice, free production, was to avoid monitor use in the Grammar instruction and the … 261 ELIA 4, 2003, pp. 255-279 scrambled sentences task, since the second task contained the target structure. The multiple-choice task could not have increased the monitor use on the free production task since it was given on a different day. 3.3. Procedure The two three-part tests were administered in a 3-week period. The pre-and the post-teaching tests were distributed on the first two days of class of the first and the third week, respectively. The instructional treatment for the experimental group took place on the third day of class of the first week. Six days of class elapsed between the pretest and the posttest. The scrambled sentences and the multiple-choice tasks were distributed on the same day (the first day of class of the first and the third week for the preand post-teaching tests, respectively), whereas the free production task was administered on the following day of class (the second day of class of the first and the third week for the preand post-teaching tests, respectively). Each test had to be distributed on two different days due to Department policy. The teaching of this grammar point was presented in a fifty-minute class under the topic ¿Somos una sociedad consumidora? “Are we a consumer society?” taken from the textbook ¿Qué te parece? (Lee, Young, Wolf, and Chandler 2000a). This textbook follows the tenets of the Communicative Approach to Spanish language teaching (cf. Lee and VanPatten 1995). The teaching plan consisted of a five-minute introduction of the topic based on the question ¿Qué te gusta comprar? “What do you like to buy?”, followed by a fifteen-minute focus on form grammar explanation under the heading “Gustar and Similar Verbs”. Doughty and Williams (1998: 3) make a distinction between two types of grammar teaching, that is, a focus on form type and a forms-in-isolation type: “a focus on form entails a prerequisite engagement in meaning before attention to linguistic features can be expected to be effective…[whereas] the traditional notion of formS always entails isolation or extraction of linguistic features from the context or from communicative activity”. Following this 262 M. D. López Jiménez ELIA 4, 2003, pp. 255-279 differentiation, the approach to grammar in this study was based on a traditional notion of formS. Grammar instruction was also accompanied by twenty-eight minutes of grammar exercises from both the workbook and the textbook titled ¿Qué te parece?. Three different tasks were practiced in class: a five-minute multiple-choice activity (structured input activity), an eight-minute scrambled sentences activity (structured output activity), and a fifteenminute free production exercise (structured output activity), which contained four reading passages based on the opinion of four different Spaniards about ecology and environmental issues. The last two minutes of class were spent on a summary of the main points treated in class that day. According to Lee and VanPatten (1995), structured input activities should precede structured output activities. This approach to grammar instruction differs from traditional approaches based on “an almost exclusive focus on production […], when grammar is presented, every textbook has hundreds of manipulative and controlled practices that have the learner “creating” output with particular forms or structures. We questioned the utility of these practices for getting linguistic information into the developing system” (Lee and VanPatten 1995: 118-119). The control group did not receive specialized instruction on gustartype verbs. Apart from not receiving specialized instruction on gustar-type verbs, the control group followed the same syllabus as the other students from S150. 4. Analysis and results The three elicitation tasks were analyzed separately. In this section a total of six summary tables, one for each task of each group, is presented: Grammar instruction and the … 263 ELIA 4, 2003, pp. 255-279 270 M. D. López Jiménez ELIA 4, 2003, pp. 255-279 The occurrence of non-target-like forms could be justified by means of a series of learner strategies: for example, L1 transfer, avoidance, L2 overgeneralization, and simplification. In terms of transfer, the cases discussed below belong to one of the three categories into which Lott (1983) classifies transfer, that is, transfer of structure (an L1 feature is used in the place of a target language feature). The introduction of ser and estar in front of a gustar-type verb (in the form of an infinitive, a conjugated form, or a derived adjective) (for example, es importar, es importante, es importa, son importan, están importan) is probably due to the learner´s L1 influence. The students who inserted forms of to be might have been plugging in translations of English words such as are important, is important rather than using the appropriate construction (gustar-type verbs). Indeed, in Spanish the sentence Los libros me importan should be translated into "The books are important to me". In addition, some students tend to assign number and person to the verb on the basis of the indirect object pronoun that precedes the gustar-type verb. The most frequent and unmarked word order for this construction is object-verb-subject (OVS) (Me gustan mis libros, "I like my books"). This order differs from the obligatory subject-verb-object (SVO) pattern (I like my books) in English. Thus, the students transferred their L1 pattern to Spanish. In terms of avoidance, I will follow Seliger´s (1989) position: in other words, avoidance only takes place if the learner has shown knowledge of the targeted form/structure. Thus, subjects #1 and #8, for example, avoided the use of a pronoun in eleven out of the fourteen obligatory contexts in the pretest. However, subject #2´s and subject #4´s performances in the pretest could not be considered a case of avoidance since they did not produce any pronoun at all in the fourteen obligatory contexts. Another strategy is overgeneralization: L1 and L2 learners make errors like goed (instead of went) and cabo (instead of quepo "I fit"), which are considered extensions of some general rule to items not covered by that rule in the L1 or the L2 . In this study some students (for example, subject Grammar instruction and the … 271 ELIA 4, 2003, pp. 255-279 #11) produced the reflexive pronoun se "self" instead of the indirect object pronouns le "him/her" and les "them". The pronoun se "self" is highly frequent in Spanish since it may function as a reflexive pronoun, as a marker of impersonality (impersonal se), and may replace le and les when they precede lo, la (third person singular direct object pronouns), los, las (third person plural direct object pronouns). The pronoun se is also morphologically closer to le and les than any other pronoun. There were other students (for example, subject #1) who resorted to the singular form to mark the verb. Subject #1 chose to mark the third person of gustar-type verbs with the singular form in nine of the ten obligatory contexts in which number had to be indicated. Nevertheless, the third person singular was required only in five out of the ten obligatory contexts. The rest required a plural marking for the verb. The last strategy, structural simplification (see Ellis´ (1982) differentiation), could be found in those cases where the students used the base form (an infinitive) of a gustar-type verb instead of a conjugated form (subject #2 and subject #9) 6. Concluding remarks Instruction promotes the acquisition of gustar-type verbs. The treatment group, although it scored lower and/or equal to the comparison group in the pretest, outperformed the control group in the scrambled sentences and the multiple-choice tasks in the posttest. The type of instructional treatment, explicit grammar teaching, the students received did not help them to be aware of the targeted structure in freer contexts. Instruction is just one variable among many, and does not seem to be a privileged variable at all. Instruction does not have the power of altering acquisition orders, although it very likely increases its rate. There are other factors (such as the type of task, individual learner differences, language learning strategies, etc.) that together with the instructional treatment are responsible for the final outcome. In the event of further studies, there are desirable changes to be taken into account: first, it would be necessary to resort to lower-level 272 M. D. López Jiménez ELIA 4, 2003, pp. 255-279 learners in order to avoid a ceiling effect in the multiple-choice task of the pretest; second, the number of obligatory contexts for the free production task would have to be expanded in order to make the data from this type of task more robust, and the results more reliable; third, it would be interesting to distribute a second posttest to investigate whether the instructional effects are long lasting or not. From a pedagogical perspective, there are several teaching implications that would also have to be considered in further studies In this study, the grammar point was introduced and explained in Spanish following one of the Department´s rules: all Spanish classes have to be conducted in Spanish and no translation into English is allowed in class. Perhaps a presentation and explanation of the same grammar in both languages, Spanish and English, would help the students to obtain better results in the posttest. It would also be wise to expand the twenty-eight minutes spent in class practicing each of the three tasks. In this way, one could avoid having the presentation and practice of gustar-type verbs comprise the fifty-minute class. In addition to this, the practice of each of the three tasks would take place on three different days. With these changes, it would be possible to study the effect(s), if any, that spending more time on the practical part of the instructional treatment could have on the acquisition of gustar-type verb constructions. Finally, it would be advisable to incorporate a complex construction like this in a cyclical syllabus as opposed to a linear one. A cyclical syllabus offers two advantages: the recurrence of the item (in this case, the grammar point) throughout the syllabus allows 1) a continual review of the item studied, and 2) an expansion of the item a step further. Grammar instruction and the … 273 ELIA 4, 2003, pp. 255-279 References Bardovi-Harlig, K. 2000. Tense and Aspect in Second Language Acquisition: Form, Meaning, and Use. Malden, MA: Blackwell Publishers. Belletti, A. and Rizzi L. 1988. "Psych verbs and theta theory". Natural Language and Linguistic Theory 6: 291-352. Cadierno, T. 1995. "Formal instruction from a processing perspective: An investigation into the Spanish past tense". Modern Language Journal 79: 179-193. Doughty, C. and Varela E. 1998. "Communicative focus on form" in C. Doughty and J. Williams (eds.). Focus on Form in Classroom Second Language Acquisition. Cambridge: Cambridge University Press. Doughty, C. and Varela, E. 1998. "Communicative focus on form" in C. Doughty and J. Williams (eds.). Focus on Form in Classroom Second Language Acquisition. Cambridge: Cambridge University Press. Doughty, C. and Williams J. 1998. "Issues and Terminology".in C. Doughty and J. Williams (eds.). Focus on Form in Classroom Second Language Acquisition. Cambridge: Cambridge University Press. Ehrman, M. 1990. "The role of personality type in adult language learning: an ongoing investigation" in T. Parry and C. Stansfield (eds.). Language Aptitude Reconsidered. Englewood Cliffs, N.J.: Prentice Hall. Ellis, R. 1982. "The origins of interlanguage". Applied Linguistics 3: 207223. Ellis, R. 1996. The Study of Second Language Acquisition. Oxford: Oxford University Press. Lee, J. and VanPatten. B. 1995. Making Communicative Language Happen. New York: McGraw Hill. Lee, J., Young, D., Wolf, D., and Chandler, P. 2000a. ¿Qué te parece? Intermediate Spanish. New York: McGraw-Hill. 274 M. D. López Jiménez ELIA 4, 2003, pp. 255-279 Lee, J., Young, D., Wolf, D., and Chandler, P. 2000b. Manual que acompaña ¿Qué te parece? Intermediate Spanish. New York: McGraw-Hill. Lott, D. 1983. "Analyzing and counteracting interference errors". English Language Teaching Journal 37: 256-261. Montrul, A. 1997. "Spanish gustar psych verbs and the unaccusative se construction: the case of dative experiencers in SLA" in A. PérezLeroux and W. Glass (eds.). Contemporary Perspectives on the Acquisition of Spanish: Developing Grammars. Somerville, MA: Cascadilla Press. Pienemann, M. 1989. "Is language teachable? Psycholinguistic experiments and Hypotheses". Applied Linguistics 10: 52-79. Pienemann, M. 1998. Language Processing and Second Language Development: Processability Theory. Amsterdam: Benjamins. Seliger, H. 1989."Semantic transfer constraints in foreign language speakers´ reactions to acceptability" in H. Dechert and M. Raupach (eds.). Transfer in Language Production. Norwood, N.J.: Ablex