This paper discusses some possibilities and limitations of laboratory research methods fortesting theories of second language acquisition. The paper includes a review of 20 experimentallab studies. The review focuses on the motivation for conducting lab studies, the use of artificialor semiartificial language structures, and various design features (including pre- and posttesting,number of subjects, random subject assignment, between- and within-subjects comparisons,treatment materials and procedures). The paper calls for lab studies addressing issues central toSLA theory (“learning” vs. “acquisition”) and ends with somemethodological recommendations, concerning the length of experimental treatments, the use ofgrammaticality judgment tasks, the measurement of reaction times, and the use of retrospectiveinterviews.

This study investigated the effects of text modification and task on incidental learning offoreign language vocabulary by reading. The focus of the study was on how different cue types(appositives and single and multiple-choice marginal glosses) and a task (translation) wouldaffect processing of input, initial learning, and retention of the meaning of target words in a text.Two hundred thirty-one students in Japanese universities were randomly assigned to 10conditiongroups. After a pretest, a reading task, and a cloze test, they were given a surprise vocabularytest. A week later, two unexpected delayed posttests were also given. The results indicated that(a) both single and multiple-choice marginal gloss conditions performed significantly better thanthe no cue and the appositive conditions, (b) no statistically significant difference existedbetween the single and the multiple-choice gloss conditions nor between the appositive and theno cue conditions, and (c) the translation task had no effect.

This paper reviews current research on language and gender and discusses theimplications of such work for gender-based research in second language acquisition. Recentwork in sociolinguistics, generally, and language and gender research, more specifically, hasrejected categorical and fixed notions of social identities in favor of more constructivist anddynamic ones. Thus, in this paper I elaborate a conception of gender that has not generallyinformed research in the field of second language acquisition, and point to more recent work inthe field that theorizes and investigates gender as a construct shaped by historical, cultural,social, and interactional factors.

This study was designed to extend previous research on the relationships amongintelligibility, perceived comprehensibility, and accentedness. Accent and comprehensibilityratings and transcriptions of accented speech from Cantonese, Japanese, Polish, and Spanishintermediate ESL students were obtained from 26 native English listeners. The listeners werealso asked to identify the first language backgrounds of the same talkers and to provideinformation on their familiarity with the four accents used in this study. When the results of thisstudy were compared with the Munro and Derwing (1995, Language Learning,45, 73–97) study of learners of high proficiency, speaker proficiency level didnot appear to affect the quasi-independent relationships among intelligibility, perceivedcomprehensibility, and accentedness; however, the relative contributions of grammatical andphonemic errors and goodness of prosody differed somewhat. Ability to identify thespeakers' first languages was influenced by familiarity.

This study examines how two types of word knowledge, word associations and grammaticalsuffix knowledge, change over time both receptively and productively. Ninety-five secondary andpostsecondary Japanese students were tested on three word associations and inflectional andderivational suffixes for each of 20 verbs, once near the beginning of their academic year andonce near the end. The results showed their average vocabulary gain was 330 words. Thestudents showed rather poor knowledge of the allowable suffixes for the verbs, especially thederivative suffixes. Likewise, the subjects did not show very good mastery of the verbs'word associations. Even for verbs rated as known, the students as a group were able to produceonly about 50% of the word associations possible on the test as judged by native speaker norms.Word association knowledge and suffix knowledge were shown to correlate with each other andwith total vocabulary size. The subjects overall had from 19 to 25 percentage points morereceptive knowledge than productive knowledge.

A major requirement for theory development in L2 vocabulary acquisition is anadequate model of L2 lexical organization and processing. At present, there is no widelyaccepted model of this kind, but recent research has demonstrated the potential usefulness ofLevelt's (1989, 1993) L1 speech processing model (de Bot & Schreuder, 1993;Green, 1993; Poulisse, 1993). This paper explores the relevance of the model as adapted for L2lexical processing of written texts, with particular attention to its mental lexicon components(concepts, lemmas, and lexemes) and organization. The model is first discussed in terms of itsrelevance to L2 lexical access, retrieval, and acquisition processes. Findings of an introspectiveL2 vocabulary study follow, detailing how university ESL learners reported dealing withunfamiliar words while carrying out L2 reading tasks. The information provided on their lexicalinferencing strategy and varied knowledge sources used in the process provide support for thelexical components and organization proposed by Levelt. It is argued in this paper that theprocess of inferring the features of unknown words in a reading passage can be described interms of lemma construction, as defined in the model.

This paper illustrates the advantages of laboratory research into SLA by describing twostudies of acquisition of second language syntax. The first addresses the question of whetherhuman morphological abilities can be understood in terms of associative processes or whether itis necessary to postulate rule-based symbol processing systems underlying these skills. Wedemonstrate that acquisition of L2 morphology shows frequency effects for both regular andirregular forms and that the acquisition course of learners' accuracy and reaction time canbe accurately simulated by connectionist systems. The second concerns a bootstrapping accountof SLA whereby processes of chunking in phonological memory underpin the acquisition oflong-term memory for utterances and abstractions of grammatical regularities. It shows thatphonological short-term memory is particularly important in the acquisition of long-distancedependencies. In both cases, it is argued that these aspects of SLA reflect associative learningprocesses. When SLA research is properly focused on acquisition, laboratory research allowsinvestigation of the learners' exposure to evidence, their processes of perception andlearning, and their resultant language representations.

This paper reports on two studies that addressed the issue of ultimate attainment by latesecond language learners. The aim of the studies, which included a carefully screened group ofhighly successful Dutch learners of English in their designs, was to determine whether or not latesecond language learners who had achieved a nativelike performance in the pronunciation of asecond language could be identified. Speech samples provided by two groups of learners, one ofwhich consisted of highly successful learners only, and a native speaker control group were ratedfor accent by native speakers of English. The ratings obtained by some learners were within therange of the ratings assigned to the native speaker controls. Such results suggest that it is notimpossible to achieve an authentic, nativelike pronunciation of a second language after aspecified biological period of time. Examination of the learning histories of the highly successfullearners lead the authors to argue that certain learner characteristics and learning contexts maywork together to override the disadvantages of a late start.

This paper provides evidence from a study of ESL speakers that cross-language transfereffects interact, sometimes in very subtle ways, with developmental effects in the construction ofL2 syllable structures. It extends Broselow and Finer's (1991) proposal for the MinimalSonority Distance Parameter-Setting (MSD) model in L2 syllables by (a) adding evidence from alanguage that unambiguously exhibits transfer in complex onsets and (b) examining syllablecodas to see whether the MSD parameter can be extended to account for the distribution of errorpatterns there, as well. The results of the study generally support the predictions that the MSDmodel makes for consonant cluster difficulty in specific syllable positions; however, it does notmake predictions about the patterns that are found in learner error types. We argue thatoptimality theory (Prince & Smolensky, 1993) provides a more explicit account of theinteractions between transfer and developmental effects in L2 syllables, allowing us to interpretwithin a singular grammatical framework not only why learners have difficulty withspecific ESL structures but also how they resolve it.

One of the differences between Chinese and English is that the former allows both nullsubjects in finite sentences and null objects, but the latter allows neither. This cross-linguisticvariation is believed to be related to the underspecification of I and topic drop in Chinese but notin English. This paper reports on an empirical study investigating the unlearning of null subjectsand null objects by 159 Chinese learners in their L2 acquisition of English. In L1 acquisition, ithas been found that English-speaking children display an asymmetry by frequently allowing nullsubjects but rarely null objects. The results of this study indicate that there is an asymmetry inChinese learners' L2 English, which, however, is opposite to that found in English L1acquisition: Chinese learners are able to reject the incorrect null subject in English, but unable todetect the ungrammaticality of the null object. It is proposed that the unlearning of null subjectsby Chinese learners of English is triggered by the evidence in their input indicating thespecification of AGR(eement) and T(ense) in English, and that the difficulty in the unlearning ofnull objects is related to the lack of informative evidence to unset the [+ topic-drop]setting in Chinese learners' L2 English.

This article presents a study of corrective feedback and learner uptake (i.e., responses tofeedback) in four immersion classrooms at the primary level. Transcripts totaling 18.3 hours ofclassroom interaction taken from 14 subject-matter lessons and 13 French language arts lessonswere analyzed using a model developed for the study and comprising the various moves in anerror treatment sequence. Results include the frequency and distribution of the six differentfeedback types used by the four teachers, in addition to the frequency and distribution of differenttypes of learner uptake following each feedback type. The findings indicate an overwhelmingtendency for teachers to use recasts in spite of the latter's ineffectiveness at elicitingstudent-generated repair. Four other feedback types—elicitation, metalinguistic feedback,clarification requests, and repetition—lead to student-generated repair more successfullyand are thus able to initiate what the authors characterize as the negotiation of form.

A consistent finding in second language acquisition research has been that in the early stagesof acquisition learners often receive simplified input. This finding has led researchers to questionwhether or not simplified input has a facilitative effect on the acquisition process. This studyexamines the effects of simplified input in early L2 acquisition by experimentally manipulatinglanguage input to two groups of learners and then assessing their acquisition longitudinallywithin a controlled laboratory setting. The impetus for the study described here wasGivón's (1990) competition hypothesis that posits that, in early L2 acquisition,vocabulary and grammar compete for memory, attention, and processing capacity. Because onecan communicate with vocabulary in absence of grammar but not vice versa, it was proposed thatlearners receiving pidgin input would acquire vocabulary more efficiently than learnerschallenged with the dual task of acquiring vocabulary and grammar simultaneously. Furthermore,it was hypothesized that once vocabulary processing skills were automated, learners wouldacquire grammar more rapidly. Results of a variety of measures reveal that the dual task ofacquiring vocabulary and grammar does not hinder either and that the longer the exposure togrammatical input the greater the advantage in real-time grammar processing abilities.Converging evidence from all measures of language learning provides strong support for theusefulness and viability of laboratory study of SLA.

This empirical study investigates the relationship between language aptitude components andsecond language (L2) outcomes among learners whose intensive L2 exposure began at differentages. The learners in this study are 65 11th-grade students in continuing early and late Frenchimmersion programs. Evidence is found to support the main hypothesis that in late immersionstarting in adolescence there will be a positive relationship between L2 outcomes and ananalytical dimension of language aptitude, whereas in early immersion beginning in grade 1 apositive relationship will hold between L2 outcomes and memory ability. A further hypothesisthat early immersion students will have higher language aptitude as a result of their early L2exposure is not supported by the findings.

Three years ago we ushered in a new rubric of SSLA, Replication Studies. Webelieved—and still do—as did many researchers and consumers of second languageand foreign language (S/FL) acquisition research that the problems of reliability andgeneralizability are best resolved through attempts at replication. We realized that the system ofrewards in academe inhibited investment in research that could be viewed as lacking originality.However, we have been gratified by the response from workers in our field who perceived thepotential contribution of replication studies and have availed themselves of this newvenue.

In this paper we address the need for replication studies in the field ofsecond language acquisition and discuss the problems surrounding standards of reportingresearch. A particular focus is on the level of detail necessary for replication to take place. Thelack of uniform standards is noted with specific examples of problematic reporting taken fromdescriptions of second language learners' proficiency levels and measures of linguisticaccuracy in second language writing research. We propose ways that we believe will lead tomore thorough reporting of research and that will therefore allow others to engage inreplication.

This study is a fine-grained analysis of extensive empirical data on the automatization ofexplicitly learned rules of morphosyntax in a second language. Sixty-one subjects were taughtfour morphosyntactic rules and 32 vocabulary items in an artificial language. After they hadreached criterion on a set of metalinguistic tests of grammar and vocabulary, they engaged insystematic, computer-controlled comprehension and production practice for 8 weeks.Comprehension practice consisted of choosing between pictures displayed on the computerscreen to match a sentence; production practice consisted of typing the correct sentencecorresponding to a picture. All subjects were taught the same rules and then practiced them, andall subjects had the same amount of comprehension and production practice, but which ruleswere practiced in comprehension and which in production varied between groups. Results showthat the learning of morphosyntactic rules is highly skill-specific and that these skills developvery gradually over time, following the same power function learning curve as the acquisition ofother cognitive skills. These results are consistent with current skill acquisition theory.

The authors begin their book with the assertion, “Many linguists today assume thattheirs is an empirical and deductive science, and that scientific progress in the domain of theirresearch is possible” (p. 1). They then proceed in an interesting, well-written, andinformative case study of the evolution of generative grammar to cast doubt on the veracity ofthis assertion. The central focus is the nature of the debates among researchers developing whatcame to be known as generative semantics and interpretive semantics. The book attempts todetail who influenced whom and who was interested in what. More particularly, Huck andGoldsmith ask to what extent data, analyses, and argumentation were critical to the debates. Thecentral question is: To what extent were generativist scholars moved one way or the other byrational considerations? The answer, not surprisingly, is: Not much. The real story is one aboutpersonalities, not about truth and reason. They attempt to explain why paying particular attentionto the personalities involved, the role of the linguistic institutions where the participants worked,and the research agendas of each is important.

This experimental study examines the extent to which 60 adult Japanese ESL learners wereable to acquire a rule regulating the argument structure frames of novel verbs of English afterexposure to grammatical examples of sentences containing the verbs. Training took place underconditions with no focus on form (implicit and incidental conditions) and with focus on form(enhanced and instructed conditions). The presentation of instances during training wasmanipulated as a test of predictions made by Logan's (1988, 1990, 1992) memory-basedinstance theory of automaticity. Results measured in reaction times show similarslopes for automaticity on trained examples in each condition but significant differences in theextent of learning, with the focus on form conditions outperforming the no focus on formconditions in transfer of learned knowledge to accurate judgments of new ungrammaticalsentences. Implications are drawn from the results regarding the acquisition of rule-based versusmemory-based knowledge from exposure to stimuli in each training condition and the influenceof this knowledge on decision-making about grammaticality during the transfer task.

Many SLA researchers have demonstrated an interest in language pedagogy (LP), yetthe relationship between SLA and LP is a problematic one, not so much because of thelimitations of SLA itself but because the two disciplines involve different“Discourses” (Gee, 1990). In this paper it is argued that an educational perspectiveis needed in order to examine how SLA can contribute to LP. Such a perspective suggests waysin which SLA can be appraised in a pedagogically relevant manner and, more importantly, whatkinds of applications may be fruitful. It is suggested that relevance is more likely to be achievedif SLA is used to address issues that practitioners nominate as important to them. Differentmodels of application are considered, reflecting different ways of viewing teaching. A behavioralmodel, according to which teachers implement those behaviors that research has shown to beeffective, is rejected. However, SLA can serve as an important source of information that canhelp to shape practitioners' theories of teaching (a cognitive model). Most importantly, itconstitutes a source of “provisional specifications” that practitioners can evaluatein their own contexts of action (an interpretation model). SLA also affords practitioners themeans for conducting their own investigations. In short, an educational perspective suggests thatfor SLA to influence LP, practitioners need assistance in transforming knowledge about L2acquisition into practice.

The book under review follows other research by Biber in the field of register variation (cf.Biber, 1988; Biber & Finegan, 1994). It further develops the thesis that multidmensionalanalysis of many different linguistic variables can help posit theoretical and conceptualprimitives for the analysis of register. It also explores a corollary that the primitive for registervariation in different languages can be productively compared.