Mansoor Al-Surmi
University of Central Missouri, USA
Bio Data:
Mansoor Al-Surmi is an Assistant Professor of English at the University of Central Missouri. His research interests broadly include investigating theoretical and practical issues in the areas of Corpus Linguistics, Assessment, SLA, and CALL.
Abstract
This paper investigates the influence of the discourse narrative structure on verbal morphology in L2 learners' interlanguage temporality system. The aim was to retest the Discourse Hypothesis predictions regarding of the influence of discourse structure on verbal morphology use in oral narrative in an English as a Foreign Language context. The discourse hypothesis predicts that L2 learners will use past forms predominantly in the foreground of the narrative while non-past forms will be used in the background. Data obtained from 36 learners was randomly chosen from a pool of pretest productions by Thai L2 learners of English. Participants were asked to narrate a strange dream after looking at six pictures. Results revealed that participants show more use of the past forms in the foreground than the background while they use more non-past forms in the background. Learners’ systematic errors in tense marking could be understood in the light of the results of the present study. The paper concludes that the discourse hypothesis is supported and that English as a Foreign Language learners exhibit similar use of tense and aspect to English as a Second Language learners.
Keywords: Discourse hypothesis, tense and aspect, EFL, Thai learners, temporality, interlanguage
Introduction
The investigation of expressing temporality in second language acquisition research has led to a focus on the surface forms of tense and aspect morphology and the underlying temporal semantics in learners’ language production (Bardovi-Harlig, 1999, 2000). As mentioned by Bardovi-Harlig (1999), the acquisition of temporal subsystems has been studied by examining the various linguistics devices and the distribution of verbal morphology employed to express temporality. Tense, grammatical aspect, and lexical aspect have become crucial facets of verbal morphology investigations. These parts have been studied within two main trends in second language acquisition: the meaning–
oriented approach, which examines how semantic concepts are expressed through different linguistic devices, and the form-oriented approach, which examines verbal morphology distribution as a marker of the underlying interlanguage semantic system (Bardovi-Harlig, 2000).
The form-oriented trend has focused on identifying and tracing the morphological forms and their distribution in interlanguage in order to determine the underlying system of expressing temporality in second language acquisition. Two main hypotheses under form-oriented trend have emerged: the aspect hypothesis (AH) and the discourse hypothesis (DH). Table 1 shows a comparison of the main principles of the two hypotheses. The AH argues for the influence of the inherent semantics of lexical aspect of verbs and predicates on verb morphology, whereas the DH supports the influence of discourse organization and storyline orientation. The AH is based on a lexical aspect theory, which suggests that lexical verbs have different inherent semantic properties.
These properties are classified according to Vendler's (1967) model of lexical aspect (as cited in Housen, 1998). Vendler's model divides lexical verbs into four types: states,
activities, accomplishments, and achievements based on whether the verb indicates that it is either stative or dynamic, punctual or durative, and telic or atelic (i.e., it has an endpoint). According to Andersen and Shirai (1996), tense aspect morphology is acquired initially with achievement and accomplishment verbs, then with activity and state verbs at a later stage.
Table 1
A Comparison of the Main Principles of the Aspect and Discourse Hypotheses Aspect Hypothesis Discourse Hypothesis Influence Inherent semantics of lexical
aspect of verbs and predicates
Discourse organization and storyline orientation
Prediction Learners will choose tense-aspect morphology according to whether lexical verbs are states, activities, accomplishments, or
achievements.
Learners will choose tense-aspect morphology according to whether lexical verbs occur in the
foreground or the background sections of the discourse.
Basis Lexical basis Discourse basis
On the other hand, the DH is based on the proposed influence of narrative structure on the learner's verbal morphology system (Hopper, 1979). Dahl (1984) defined narrative discourse as a text “where the speaker relates a series of real or fictive events in the order they took place” (p.116). Narrative discourse, according to Dahl, consists of the main-line of events and the interrupted background information. These two components are referred to as foreground and background respectively in discourse literature.
Hopper (1979) defined foreground as “the parts of the narrative which relate events belonging to the skeletal structure of the discourse” while he defined background as the
“supportive material which does not itself narrate the main events” (p. 213) but supports, amplifies, or comments on the events in the main foreground narrative. The foreground includes events that succeed one another in the narrative as they actually happened in the real world (i.e., the events in the foreground are sequential) while the background includes events that are not sequential but rather concurrent with the events in the foreground (Hopper, 1979). Hopper presented the following example to illustrate this distinction:
Example 1
We went back to the camp, and ran away during the night, and we traveled several days, we passed through several villages, and in all of them we did not have to pay tribute
Foreground Background We went back to the camp
and ran away during the night we passed through a few villages we traveled several days and in all of them we did not …
Notice that the background events are not sequential in relation to the main story line but concurrent to the third event in the foreground and that they present supportive information to the main sequence of events. Table 2 shows a comparison of the main characteristics of foreground and background.
{
Table 2
The Main Characteristics of Foreground and Background
Foreground Background
Function presents main events. presents supportive events.
Timeline moves time in discourse forward. does not necessarily move time.
Narrativity includes sequential events. includes concurrent events to main events.
Punctuality usually includes punctual events. usually includes events reporting durative, repetitive, or habitual events.
Completeness usually includes completed events. usually includes ongoing events.
The sequentiality of events in the foreground entails a forward movement in time (Dry, 1981). A “narrative of consecutive events creates for the reader or hearer an imagined time stream as a dimension of the narrative world in which the events occur” (Dry, 1983, p. 19). Reinhart (1984) described the temporal criteria of foreground as follows:
1. Narrativity, or temporal continuity: Only narrative units, i.e., textual units whose order matches the order of the events they report, can serve as foreground
2. Punctuality: Units reporting punctual events can serve more easily as foreground than units reporting durative, repetitive, or habitual events.
3. Completeness: A report of a completed event can serve more easily as foreground than a report of an ongoing event. (p. 801)
These criteria identify and distinguish foreground from background, which has different
characteristics and functions. According to Hopper (1979), background events are not as sequential as foreground events and provide supportive information, which elaborates on or evaluates prior, simultaneous, or following events in the foreground. These criteria and definitions were used as the operationalized terms in the analysis of the data of this study.
Discourse Hypothesis
Narrative structure is one of the factors that second language research correlates with the acquisition of tense and aspect morphology (see Bardovi-Harlig, 1992, 1995, 1998;
Flashner, 1989; Housen, 1994, Tre´vise, 1987 among others). Godfrey (1980), when studying discourse and tense-aspect morphology in interlanguage, observed that the use of tense verbal morphology cannot be interpreted without considering discourse structure.
Native speakers distinguish between foreground and background of narrative by using tense-aspect morphology (i.e., native speakers use simple past to narrate the successive events in the foreground). This use raised the question of whether second language learners would acquire tense morphology based on the distinction of the foreground and the background. It also raised the question of whether the route and sequentiality of events would really entail the consistent use of one form of verbal morphology.
In a case study investigation of the use of tense morphology based on the grounding of a narrative by a Japanese immigrant to the US, Kumpf (1984) suggested that there is a relationship between the use of verbal morphology in interlanguage and the grounding of narrative. Hence, the interlanguage DH evolved to predict that learners use emerging verbal morphology to distinguish foreground from background in narratives. The DH predicts that learners will use more past tense than non-past forms in the foreground
while non-past forms will be predominant in the background (Bardovi-Harlig, 1992).
Furthermore, as the learners' interlanguage system develops, it is expected that there will be target-like performance and the past tense will be prevalent in both foreground and background.
The study of DH includes early case studies (e.g., Flashner, 1989; Kumpf, 1984) and late larger studies (e.g., Bardovi-Harlig, 1992, 1995, 1998). In studying the temporal system and universality of interlanguage, Kumpf (1984) argued for the idea that discourse-function perspective applies to both native language and interlanguage analysis: "it is the discourse context which creates the conditions under which the forms appear" (p. 132). Kumpf focused on tense-aspect morphology and how the narrative discourse influences its distribution in interlanguage. He examined the personal narrative of a low-proficiency English learner of Japanese L1 background, who had been in the States for 28 years. The data of this case study was collected through conversation interviews with the subject. The results showed that completed actions in the foreground were expressed with the base verb form while in the background most verbs were marked for tense, especially the stative verbs. Kumpf then concluded that the basic interlanguage system is influenced by the distinction of foreground and background in narrative.
The influence of discourse structure on verbal morphology has been raised in those studies that focused on the process of expressing temporality in interlanguage. However, the above studies did not focus solely on the influence of discourse narrative on verbal morphology, but rather on all means of expressing temporality. Flashner's (1989) study of tense morphology distribution in the foreground and background of narratives by 3 Russian adults, therefore, seems to be the first study that focused only on this distribution.
Flashner examined the spontaneous personal narratives of 3 Russians who had lived in
the United States for more than 2 years and had different levels of English proficiency (1 beginner, 1 intermediate, and 1 advanced). Results showed that all 3 speakers’ past forms indicated perfective actions and correlated with narrative foreground while non-past forms, especially base forms, denoted imperfective function and correlated with narrative background. Flashner concluded that "the notions of background and foreground provide a gross distinction that enables us to examine the use of tense and aspect in narrative monologues" (p. 95).
Taking discourse influence as the theoretical basis of the study, Bardovi-Harlig (1992) focused more closely on discourse structure and tense morphology in narratives using a larger number of participants than earlier studies. Examining the narratives of 16 intermediate learners of English in an intensive program, Bardovi-Harlig tested the prediction of the influence of discourse on tense marking. Data was collected through written stories and oral retell. After listening to the story two times, the participants were individually asked to retell it orally to one of the interview team members and then write the story down. The purpose of collecting oral and written versions of the narrative was to provide two ways to test learners' developing knowledge of tense use. Previous studies used only oral narrative and did not investigate written narrative. Results showed that in both the oral and written narrative the foreground was associated with a high rate of appropriate use and dominance of the past tense while the background had lower rate of appropriate use of past tense and consisted predominantly non-past forms especially present tense, which was virtually absent in the foreground. Bardovi-Harlig argued that the use of tense by learners of English as a second language could best be understood from a narrative structure perspective.
Focusing on the distinction between the foreground and background in narrative and the
effect of this distinction on the development of learners' tense aspect interlanguage morphological system, Bardovi-Harlig (1995) replicated her previous study but this time with even larger number of learners: 37 beginner, intermediate, and advanced learners of English in an intensive program. The aim of choosing different levels of proficiency was to evaluate the difference and predict the development in the tense-aspect morphological system of the learners. In this study, the participants were asked to watch a silent film and then retell the story to one of the interview team members; later, the participants were asked to narrate the story in writing. Results indicated that the simple past tense appeared in the foreground first with a much higher rate than in the background at all levels while the background was mainly occupied with base form and simple present. The simple past in the advanced level was dominant in the background but never reached the same high level of use in the foreground. Moreover, while the foreground had both past tense and base forms, the background included past tense, base, progressive, as well as present forms. Bardovi-Harlig concluded that the interlanguage pattern of tense-aspect distribution reflects simple function of the foreground and multiple functions of the background.
The results of these studies support the influence of discourse structure on interlanguage verbal morphology system. However, they report quite contradictory findings. The DH states that, at early stages of interlanguage development, learners will use more past tense forms in the foreground and more non-past forms in the background.
Table 3 shows sample studies that supported and did not support the DH.
Table 3
Sample Studies that Supported and Did Not Support the DH
Supported Did not support
• Flashner (1989) found that perfective actions were expressed in past forms and correlated with foreground while imperfective actions were expressed in non-past forms and correlated with background.
• Bardovi-Harlig (1992) found that a high rate of appropriate use and dominance of the past tense were related to foreground whereas lower rate of appropriate use of past tense and non-past forms dominance, especially present tense, were related to the background.
• Kumpf (1984) found that completed actions in the foreground are expressed with the base verb form (i.e., non-past) while in the background most verbs are marked for tense (i.e., past forms).
The results of studies testing a hypothesis are not always consistent and it is therefore sometimes necessary to replicate a study using a different context in order to reach robust and solid grounds that support and confirm the predictions of such hypotheses (see Polio
& Gass, 1997). Thus, the purpose of this paper is to retest the predictions of the DH in a different context and using different data. While all the studies mentioned above tested the DH in an English as a Second Language (ESL) context, to my knowledge, no study has tested the hypothesis in an English as a Foreign Language (EFL) context.
The present study uses data produced by EFL Thai learners in order to further expand our understanding of the role of discourse structure in second language acquisition and of
the possible effects of foreground and background distinction in narrative on an interlanguage verbal morphology system. Through retesting the predictions of the DH, the present study examines the effect of narrative structure on verbal morphology distribution in L2 EFL learners' interlanguage system. The DH hypothesis would predict that Thai L2 learners of English will use past forms predominantly in the foreground and non-past forms predominantly in the background.
Methods Participants
Participants in this study were 36 Thai learners of English who were selected randomly from a pool of participants in a larger scale study. This larger study used a dream narrative task as the pretest assessment of the participants' use of past tense (McDonough, 2007). Participants were all first-year university students enrolled in a Bachelor’s degree program in English at a large public university in northern Thailand. 4 males and 32 females participated in the study; their ages ranged from 17 to 20 years old, with an average age of 18.22 (SD = .637). Their length of previous study of English ranged from 7 to 15 years, with an average of 10.72 (SD = 2.13).
Materials and Procedures
The testing materials were communicative activities in the form of one-way information-gap tasks that the learners carried out individually in a language laboratory. Each test contained a dream narration task that elicited contexts for past time verbs. Learners were given a series of six pictures, which illustrated related events, and were asked to narrate a
dream based on those pictures. This elicitation task was performed at the beginning of the school year to assess the initial knowledge of temporal morphology in the learners' interlanguage before giving the participants any treatment (see McDonough, 2007). The tests were administered in a language laboratory using a pre-recorded audiotape that gave instructions and controlled the amount of time for each activity. Participants were given six pictures to look over in 30 seconds and then were asked to recount their strange dreams based on the pictures in 15 minutes. The learners were seated at individual carrels equipped with boom microphones. The test sessions took approximately 20 minutes to complete.
Analysis
The data obtained were coded first for foreground and background clauses. Clauses that were sequential and moved the narrative time forward were categorized as foreground while concurrent clauses that either described actions or presented supportive information were categorized as background. The categorization of foreground and background clauses were made independently of the verbal morphology and some clauses were excluded from the analysis for various reasons. The first clause of each narrative - "last night I had a strange dream" - was excluded because participants were asked to begin their narratives in this manner. Irrelevant clauses as in Example 2 were excluded too as the participants were not narrating the dream but presenting irrelevant information or questions.
Example 2
…I ride it and I break home again but my home is disappear. Do you think it's strange dream.
The analysis did not include clauses with missing predicators as shown in Example 3.
The clauses had no verbs (i.e., the participant did not supply the verb of the clause) and it was not predictable whether the learner would use past tense, present, or base form.
Example 3
a) It run very fast and I like it too much // it very strong // it face look frightened.
b) Suddenly, // a lot of water // to city // I try to swim.
Quoted speech was not considered either. However, the verb that introduced the quoted speech, if available, was counted. Example 4 (a) shows a case with a quotation verb while in 4 (b) there is no verb introducing the quoted speech.
Example 4
a) He said // you are the prisoner// I said No // help me please.
b) I just came to the door and knock the door. // Is there any body//
The grounding coding was followed by verbal morphology coding. Each verb in the
The grounding coding was followed by verbal morphology coding. Each verb in the