Table of Contents
2 Constructions in Second Language
3 Efficiency of Construction Teaching
4 Application to Second Language Teaching
4.1 Current State of Second Language Teaching
4.2 Necessary Changes
4.2.1 New Tenets for Construction Based Second Language Acquisition
4.2.2 Usage-Based Teaching Methods as a Basis for Construction Teaching...
4.3 Identification of Constructions Relevant for Teaching
For more than 20 years now, the concept of constructions has been playing a more and more important role in theories of language acquisition and language use. In the 1980s Fillmore, Kay and O’Conner (1988) were the first linguists interested in constructionist approaches; and with her two books Constructions: A Construction Grammar Approach to Argument Structure and Construction at Work: The Nature of Generalization in Language Goldberg (1995, 2006) eventually paved the way for this alternative view on grammar. By now, Construction Grammar has become a well- accepted descriptive and processing model that is based on a substantial body of scientific publications (Gilquin & Knop 2016: 3). However, the big interest of Construction Grammar research in first language acquisition and native speakers’ language use contrasts the little interest in the branch of second language acquisition. It is only recently that linguists have approached the question whether second language learners’ linguistic competence relies on constructions as well. Hence, it is no surprise that there has been only little interest in application of Construction Grammar in second language teaching as well. In my opinion, this is a wasted opportunity. If Construction Grammar is widely accepted in the field of first language acquisition, it is also necessary to transfer this concept to second language acquisition and teaching in order to create suitable teaching materials and methods. Thus, this paper is supposed to advocate an applied Construction Grammar in second language teaching.
After this introduction, I begin with a short review on experimental evidence that proofs the existence of constructions in second language use. The results of this experiments do not only show that constructions exist in the language learners mind. They reveal furthermore that language learners produce and store constructions that are unique to the process of second language acquisition and cannot be found in native language. The third chapter is dedicated to the potential that lies in second language teaching based on the principles of Construction Grammar. A glance at Asia shows how construction teaching may improve student’s second language skills. The fourth and last part of this paper contains a short assessment of the current state of language teaching with focus on text-books used in Germany showing that Construction Grammar has not found its place in the curricula yet. However, this chapter also deals with the question to what extent the current pedagogical approaches need to be changed or extended to include construction teaching and how constructions relevant for teaching can be identified.
2 Constructions in Second Language
One of the earliest studies that concerned constructions in second language use was conducted by Liang (2002) who replicated Bencini’s and Goldberg’s (2000) sentence sorting experiments with a testing group of Chinese learners of English. The study revealed that the foreign language learners sorted by constructions rather than by verbs. Furthermore, it appeared that the constructional sorts depended on the learners’ second language skill since the more proficient test persons were more likely to sort according to constructional meaning. This result was confirmed by Gries and Wulff (2005, 2009) in further sorting experiments with advanced German learners of English. In their experiment, it even turned out that foreign language learners rely more heavily on the respective constructions than the native speakers in Bencini and Goldberg’s sorting experiment (Gries & Wulff 2005: 192).This is remarkable since such a result suggests a huge importance of constructions in the process of foreign language learning. Furthermore, Gries and Wulff were able to verify that their participants were as sensitive to syntactic priming for ditransitive and prepositional dative argument structure constructions as their native speaking counterparts (Gries & Wulff 2009: 180). Recently, the effect of syntactic priming was also tested by Baicchi (2016) with a group of Italian English Learners. Their test persons could be primed as well which demonstrated that “the mapping between syntax and semantics occurs also in the mind of Italian learners of English” (Baicchi 2016: 228). These results were of special interest because two of the used constructions do not belong to the Italian construct-icon. This eliminated the likelihood of transfer which means that the test persons’ perception of L2 construction based solely on their acquired knowledge of that language - the knowledge of constructions. In 2007, Jiang and Nekrasova performed an online grammatically judgment task with English Learners and native English speakers. Both groups responded faster and more accurate to the formulaic utterances than to the non-formulaic phrases. At a similar result arrived Conklin and Schmitt (2007) with their measurement of reading times for formulaic and non-formulaic phrases in native and non-native speakers. In accordance with the theory of Construction Grammar, both groups read the formulaic sentences more quickly. Another replication of one of Bencini’s and Goldberg’s experiment was conducted by Valenzuela Manzanares and Rojo López (2008) with Spanish learners of English. They performed an acceptability judgement task and combined it with corpus analyses of native corpora and learner corpora. This experiment, again, confirmed that foreign language learners acquire mental representations of constructions in their L2. And again, the tested constructions either do not exist or have a different form in Spanish. So both L1 and L2 learners are sensitive to the frequencies of occurrences and storage of construction and it is assumed that they learn these statistics implicitly from usage (Ellis & Wulff 2015: 412). Following those results one may assume that the grammatical and lexical knowledge form a continuum in the readers’ minds as well. This continuum of construction ranges from “entrenched and conventionalized formulaic units [...] to loosely connected but collaborative elements.” (Ellis & Wulff 2015: 412)
Nevertheless, it would be an oversimplification to put construction learning in L2 on a level with construction learning in L1. Both are related but separate matters (Ellis & Wulff 2015: 412). In contrast to first language acquisition, second language acquisition consists of constructional and re-constructional processes (Ellis 2013: 366) that are strongly influenced by such effects like transfer or blending. The phenomenon of transfer, that was observed in many experiments, illustrates the presence of L1 elements in second language use. Foreign language learners constantly come with their own constructions while learning an L2 (Gilquin & Knop 2016: 6). This has positive and negative effects on the process of second language acquisition. This becomes especially evident within closely-related language pairs. Della Putta (2016) tested the effect of transfer with a group of Spanish-speaking learners of Italian. The constructions in question were syntactically similar in both languages but differed semantically. As expected, the learners referred back to L1 constructions in order to analyze the semantic content of a formal but not functional counterpart in L2. Such an transfer-generated misinterpretation produces errors that are usually highly fossilized and hard to correct (Della Putta 2016: 238). The alleged familiar construction needs to be re-constructed in order to acquire more second language proficiency. However, transfer is not always bad, it might be useful as well. In cases of constructions that are similar in form and function, transfer effects support the learning process. Another effect that needs to be considered in L2 acquisition is blending which is defined as the combination of two inputs in mental space that results in a third mental space, the blend (Waara 2008: 52). Such a blend, for example, is the following construction which appeared in Waara’s analysis of constructions with get in utterances of Norwegian learners of English.
(1) *Can I get a dance?
The test subject seemed to have mixed up two otherwise conventional constructions with obtain like in
(2) Can I get a coffee ? and
(3) May I have this dance?
This interplay between construction, transfer, blending and re-construction results in an interlanguage that is highly characterized by learner constructions. According to Waara, such is a construction “which is used in a slightly unconventional manner. Although usage does not result in a communication breakdown between participants, it deviates in some way.” (Waara 2008: 53) By blending and transfer learners use what they already have to facilitate their proficiency in L2. Such constructions, which make sense but sound odd, are unique in L2 acquisition and they must be seen as potential sources of error but also as useful indicators that provide insight into the learners’ conceptualization processes and thus into their state of language system.
The results of such studies suggest that L1 constructions, L2 constructions and learner constructions are present in the foreign learners’ mind. Consequentially, this requires special attention to construction teaching in second language acquisition. This, in turn, requires a revision of grammatical and didactical principles used in school. Furthermore, the psychological reality of constructions in the language learner’s mind warrants a redefinition of the concrete learning goal. According to Wulff and Gries, L2 accuracy should be seen as “the selection of a construction (in the Goldbergian sense of the term) in its preferred context within a particular target variety and genre” (Gries & Wulff 2011: 70). This should be the goal for every second language learner.
3 Efficiency of Construction Teaching
The need for an adjustment of recent methods in second language teaching in line with Construction Grammar becomes obvious from the various experimental results gathered in the preceding chapter. The results have shown that L2 constructions constitute psychologically real and fundamental entities in the learner’s mind and that they play an important role in the process of second language acquisition. Thus, the integration of Construction Grammar into foreign language didactics is not only advisable but inevitable. Concerning first language acquisition and use, constructionist models and their new insights are well-accepted in the scientific world. Consequently, one must not ignore such findings in the branch of second language teaching. Especially, because experiments have shown that such an approach may be promising.
Unfortunately, such experimental evidence for the effectiveness of teaching methods inspired by Construction Grammar is still rare. Asian researches, however, present promising results that speaks in favor of a construction-centered second language instruction. Sung and Yang (2016) tested the efficiency of construction- centered teaching within a group of 93 Korean learners of English. The construction in question was the transitive resultative construction which was considered as “one of the most difficult constructions for [Korean] learners of English as a foreign language” (Sung & Yang 2016: 89). The test persons were divided into two groups. The first group was provided with a construction-centered instruction that focused on the constructional properties of this construction. The second group was taught by means of a form-centered instruction that emphasized the construction’s syntactic composition. Both types of instruction consisted of two lessons with a length of respectively 25 minutes. Before and after the instruction, both groups underwent Korean-to-English and English-to-Korean translation tests with similar tasks. The pretests revealed that both groups had only limited command of the construction in question. After instruction, both groups showed improvements in their translation skills. The group that was provided with a construction centered instruction, however, demonstrated a more significant increase in language proficiency with regard to the transitive resultative construction. This is all the more astonishing as the students’ English proficiency was at the level of novice or novice-high while this construction is seen as one of the most challenging construction for Korean learners. Furthermore, the explicit instruction of this construction positively affected the translation proficiency in other constructions like the intransitive motion, the caused-motion and the ditransitive construction. According to Sung and Yang, this effect results from the network structure of constructions. A glance at the hierarchical network of argument structure constructions (figure 1) shows that those three constructions are more basic and closer to the prototype than the transitive resultative construction. This leads to the assumption that the explicit teaching of a more marked construction improves the handling of less marked constructions within the same hierarchical network. This effect is even stronger when the more marked construction and the less marked construction are directly linked to each other like in the case of the transitive resultative and the caused motion-construction (Sung & Yang 2016: 106ff.).
Abbildung in dieser Leseprobe nicht enthalten
Figure 1: Hierarchical network of argument structure construction (adapted from Goldberg 1995: 109)