Input Matters in SLA

Thorsten Piske, Martha Young-Scholten, Thorsten Piske, Martha Young-Scholten

• Publisher: Multilingual Matters
• Year: 2008
• ISBN:
  9781847699312
• Topic: Sprachen & Linguistik
• Subtopic: Psycholinguistik

This volume bridges the knowledge gap between second language acquisition researchers and second language pedagogy professionals in its focus on a topic of mutual interest: input. The reader-friendly contributions from seasoned researchers including Stephen Krashen, Bill VanPatten and new voices offer a wide range of existing and new perspectives on the matter of input. A rare feature of the book is that it includes extensive coverage by experts including James Flege and Alene Moyer of the acquisition of the sound system of a second language, where input seems to matter most. Those who are just making their acquaintance with second language acquisition research or updating their knowledge will find the editors' introductory chapter on past and current issues in the field particularly useful.

Part 1

Matters of Input

Chapter 1

Input Frequency and the Acquisition of the Progressive

ANDREAS ROHDE

Introduction

The progressive form has proved to be very salient in native English-speaking INPUT so that it is generally expected to figure very early in L2 learner data. However, the fact that the progressive form has a number of different functions may render the acquisition process rather complex and the mere fact that the progressive is correctly formed in L2 acquisition does not entail that its uses are in fact target-like. In the first part of this chapter, the development of the progressive form is sketched for four German children acquiring English during a six-month stay in California. Two main functions of the progressive are under scrutiny: as a marker of grammatical aspect with both present and past reference, and as a marker of future tense without marking grammatical aspect. The data are compared with input data from American children the German children were in contact with. The distribution of the functions over the six months varies considerably, suggesting that each function of the progressive is tackled separately. In the second part, the focus is on LEXICAL ASPECT or AKTIONSART and the ASPECT HYPOTHESIS. In relation to the results of the first part, it is investigated to what extent the PRODUCTION data reflect the distribution of –ing inflected verb types in the input data with regard to the verbs’ inherent verbal aspect.

The Role of Input in Second Language Acquisition

In the days of behaviourism, both L1 and L2 language acquisition¹ were mainly seen as a process of the learner’s imitation (Lado, 1957; Skinner, 1957). This is why the input any language learner was exposed to was of primary importance. L2 learners’ output was viewed as a more or less faithful mirror of the language which NATIVE SPEAKERS of the TARGET LANGUAGE provided (Gass & Selinker, 2001: 259ff). Once language acquisition was increasingly regarded as a creative construction process (Dulay & Burt, 1974b), researchers became more interested in the internal mechanisms of the learner and in the developmental sequences that could be identified for linguistic structures (for L1 acquisition see Bloom, 1970; Brown, 1973; for L2 acquisition see Bailey et al., 1974; Dulay & Burt, 1973, 1974a, 1974b; for L2 negation see Wode, 1976; for L2 questions see Wode, 1981). The input the learners received was only of marginal interest, all the more so because the POVERTY OF THE STIMULUS was and is regarded as a characteristic feature of the input, leading to the logical problem of language acquisition (White, 1989). But input (also known as PLD = PRIMARY LINGUISTIC DATA) has been considered essential by generative approaches for some time (see Carroll, 2001; Schwartz, 1993; Schwartz & Gubala-Ryzak, 1992). Non-generative approaches from the 1970s on have also considered input (and social exchange) as crucial (e.g. Krashen’s INPUT HYPOTHESIS, Swain’s OUTPUT HYPOTHESIS, Long’s INTERACTION HYPOTHESIS, CONNECTIONISM – for an overview see Mitchell & Myles, 2004).

The present study’s background is a functional non-generative approach to L2 acquisition, that is, it is not assumed that linguistic knowledge is available from the outset. Rather, this knowledge is determined by general learning mechanisms ‘operating on the rich data provided by human interaction’ (Ellis, 1994: 369). In this study, learner input proves to play a two-fold role. First, it will be shown that both input frequency and saliency trigger the use of the progressive form by German learners of English. Second, however, the polysemous nature of the progressive in the input makes it difficult for the learners to subconsciously attribute a clear-cut function to the progressive form, leading to a rather complex developmental sequence which does not lead to L2 mastery at the end of a six-month stay.

The Progressive Form in English

The progressive form in English has a number of different functions which range from marking grammatical aspect to marking future tense to expressing stylistic nuances.² In the following, two main functions of the progressive form in English are briefly discussed as they play a major role for the data these learners produce. The first concerns the progressive as a grammatical aspectual category, the second is the progressive’s function as a marker for future tense.

The progressive as a marker of grammatical aspect

The prototypical and most frequent function of the progressive is to mark grammatical aspect, that is, give information about the internal structure of the action or event expressed. According to Quirk et al., the main semantic features of the progressive are:

imperfectivity: the action or event are not complete.

duration: actions and events are not punctual.

boundedness: the duration of states expressed in the progressive is limited. (Compare ‘We are living in London’ in contrast to ‘We live in London’). (Quirk et al., 1985)

As a grammatical aspectual category, the progressive can be used across all tenses. In the past tense, for example, the progressive is often used when the background of an event (1) or an implicit reference point is given (2):

(1) I was reading when the doorbell rang.

(2) She was writing her first novel at the time.

In connection with the present perfect, the progressive can be used to make subtle semantic differences which, however, are not important for the present study as this combination is not featured in the learner data used for the analysis.

The progressive as a marker of future tense

There are two uses of the progressive with future reference. The first one regards the periphrastic CONSTRUCTION going to + V, which expresses ‘future fulfilment of present intention’ (Quirk et al., 1985: 214). This function is not discussed as it is a specific use where the -ing inflection appears in the contracted form gonna, which as frozen form is likely not analysed by the learner as going to.

(3) I’m going to see Deirdre in Berlin tomorrow.

(4) She’s going to leave in a couple of days.

Accordingly, the second use of the progressive with future reference can be referred to as ‘future arising from present arrangement, plan, or programme’ (Quirk et al., 1985: 215). Here, the progressive form is used without any additional verb:

(5) I’m finishing my work after dinner.

(6) Megan and I are leaving for Berlin tomorrow.

It will be shown that this latter use of the progressive for marking future tense is quite prominent in the learners’ production discussed in this chapter. It has to be noted here that German, the L1 of the learners analysed in this chapter, does not mark grammatical aspect with an auxiliary. In German, the difference between habituality (‘I read a book’) and ongoing processes (‘I am reading a book’) is not marked morphologically. The context usually makes clear whether an action is habitual or in progress. If the difference has to be made explicit, temporal adverbs, such as nun (now) or gerade (just now) have to be added (‘Ich lese ein Buch’ vs. ‘Ich lese gerade/nun ein Buch’).

In the input, -ing is phonetically more SALIENT than the other verb inflections -s and -ed, as -ing (be it in its full form or as [IN])³ always consists of two sounds, thus changing the prosodic structure of the verb. In addition, forms of the auxiliary be used with the present participle makes the progressive form apparently easy – that is, salient – for the learner to identify in the input. As a consequence, -ing figures very early in production data from learners. This observation has been supported by numerous studies, from the morpheme order studies of the 1970s (see Introduction above) to more recent studies in L2 PROCESSABILITY THEORY (Pienemann, 2006; Pienemann et al., 2006). However, it cannot be concluded that the progressive form is target-like just because -ing inflected verbs occur in L2 data from early on. We will in fact see that the developmental sequence for the progressive form is a long drawn-out process.

The Aspect Hypothesis

As stated above, the progressive primarily represents a grammatical aspectual category in English. Grammatical aspect has to be clearly distinguished from the lexical or inherent aspect of the verb (this phenomenon is also referred to as aktionsart [aktions = action; art = manner’] (see Andersen & Shirai, 1994; Comrie, 1976). According to Vendler (1967), the following types of lexical aspect can be distinguished:

(i) Achievement – that which takes place instantaneously, and is reducible to a single point in time, this point being the necessary endpoint or goal (e.g. start, recognize, die, reach the summit,⁴ etc). These verbs or predicates are referred to as ‘telic’ (Greek telos = ‘aim’).

(ii) Accomplishment – that which has some duration, but has a necessary endpoint or goal (e.g. run a mile, make a chair, build a house, write a book, etc). These verbs and predicates are also referred to as ‘telic’.

(iii) Activity – that which has duration, but without a necessary endpoint (e.g. run, walk, play, sing, etc). These verbs and predicates are referred to as atelic.

(iv) State – that which has no dynamics, and continues without additional effort or energy being applied (e.g. see, love, hate, want, etc). These verbs are referred to as atelic (Shirai & Kurono, 1998: 247 ff.) (see Table 1.1).

The Aspect Hypothesis (AH) predicts that verbal inflections in both early L1 and L2 acquisition redundantly mark the lexical aspect inherent in the verb or predicate rather than tense or grammatical aspect. These predictions of the AH have been shown in L2 acquisition for a variety of languages and language combinations (Andersen & Shirai, 1994; Rohde, 1997; Salaberry & Shirai, 2002).⁵ This study will exclusively focus on the development of the progressive form, thus the other verbal inflections will be ignored.

In L2 English, the -ing inflection⁶ is prototypically affiliated with activities and accomplishments whereas -ed is mainly found with achievements, and the 3rd person singular – s is predominantly associated with states (Housen, 2002; Rohde, 1996, 1997).

Data and Procedure

The data of four children, aged four to nine, are reviewed. These children spent six months in a small town in California and acquired English in naturalistic contexts without any formal instruction involved.⁷ Their L2 development was documented in the form of diary data/spontaneous notes. Additional data come from selected tape recordings that were made to complement the diary data (Rohde, 1996, 1997; Wode, 1981). The input data discussed in this chapter are taken from seven selected tape recordings that have been exhaustively transcribed to include interlocutors. The data comprise spontaneous speech from nine American children aged six to nine.

Two developmental sequences will be shown for the German children’s L2 acquisition of the progressive form with respect to its function. In the first part of the study, the development of the progressive form with its functions as both a grammatical aspectual category and as a means of marking future tense is presented. In the second sequence shown, the link between the -ing inflection and the highlighted lexical aspect is investigated within the framework of the Aspect Hypothesis. It will be shown that each function of the progressive form is tackled separately so that the learner data only reflect the input distribution to a limited extent. It will be suggested that due to the fact that the children highlight different functions of the progressive in each month of L2 exposure, the predictions for the developmental sequence of -ing inflected verbs according to the Aspect Hypothesis are not entirely met and may have to be slightly modified. The data will also be discussed with regard to the Distributional Bias Hypothesis (see later section), which makes predictions about the distribution of lexical aspect in L1 speech.

The Learners: The Development of Uses of the Progressive Form. Results and Discussion

In this section, the data of the six-month stay are presented in chronological order divided into monthly samples of L2 exposure, starting with the second month as there are no instances of the progressive form documented earlier.⁸

Month 2

(7)	H/D	I’m pitching really fast.^9
(8)	H/D	I’m stealing (H is stealing a base in a baseball match).
(9)	H/D	It’s car coming.^10
(10)	H/D	Hey, look John. I’m riding my bike like this.
(11)	L/T	Heiko’s sleeping.
(12)	H/T	Where are you kicking?
(13)	H/T	He play on your team.

In (7), the 9-year-old Heiko goes outside to practice pitching. Strictly speaking, this utterance refers to a future event, that is, something that the boy intends to do. (8) is taken from a baseball match where the form stealing had been heard in the input numerous times before. (9) gives an example of a verb that at first exclusively appears in the progressive. (10), (11) and (12) are examples which include typical verbs appearing in the progressive. However, all three verbs are also used in their base forms in a target-like fashion. (13) is an instance w...