Motivation and Demotivation in the Korean Military Context

TABLE OF CONTENTS

List of Tables

1. Introduction

2. Literature Review
2.1. The Socio-Educational Model
2.2 The Quantitative Methodology of the Socio-Educational Approach
2.3 Criticism of the AMTB/Socio-Education Model
2.4 The Agenda for Change
2.5 The Need for Qualitative Enquiry
2.6 Attributional Processes
2.7 Demotivation
2.8 Ownership of English
2.9 English as a Lingua Franca
2.10 Integrativeness in World Englishes Contexts
2.11 The ‘Ideal L2 Self’
2.12 Future Research
2.13 Research Aims

3. Method
3.1 Semi-Structured Interviews
3.2 Ethical Considerations
3.3 Data Analysis

4. Results
4.1 Current Motivation
4.2 Integrative/Instrumental Orientation
4.3 Changes in Motivation over Time
4.4 The Role of Textbooks, Teachers, & Institutions
4.5 Motivation Strategies
4.6 The Role of Testing
4.7 Previous Learning Experiences & Motivation
4.8 A Duty to Learn English
4.9 ‘Doing Well’ & Success/Failure Attribution
4.10 Target Groups, Pronunciation/Style, and Ownership
4.11 Demotivation
4.12 Personal Factors & Workload

5. Discussion
5.1 Research Aim 1
5.2 Research Aim 2
5.3 Research Aim 3
5.4 Research Aim 4
5.5 Research Aim 5
5.6 Research Aim 6
5.7 Research Aim 7
5.8 Research Aim 8
5.9 Research Aim 9
5.10 Limitations of the Study
5.11 Future Research Implications
5.12 Conclusion

6. Bibliography

7. Appendices

LIST OF TABLES

Table 1 Pilot Prompts

Table 2 Main Motivational Factors

Table 3 Overall Motivational Factors

Table 4 Temporal Factors

Table 5 Effect of Teachers, Textbooks, and Institutions Table 5a Extracts

Table 6 Motivation Strategies

Table 7 Learner Perceptions of Testing Table 7a Extracts

Table 8 Previous Learning Experiences Table 8a Extracts

Table 9 A Duty to Learn English

Table 10 ‘Doing Well’ & Success/Failure Attributions

Table 11 Target Groups, Pronunciation/Style, & Ownership Table 11a Extracts

Table 12 Demotivation

Table 13 Personal Problems & Workload

1. INTRODUCTION

L2 motivational research has recently begun to move away from the traditional Gardnerian methodology. One major antecedent to this move is the problematic application of the Integrativeness concept in World Englishes contexts. The number of non-native speakers of English has recently outgrown that of native speakers, and theories concerning pronunciation, ownership, and the function of English in the world have come to the forefront of academic research surrounding linguistics and the English teaching profession. Further, there is an ongoing agenda to increase the amount of qualitative enquiry and diversify contexts in which research is carried out, such that new factors can be established and applicability of potential models in different contexts assessed. In an effort to contribute to this move in L2 motivation research, four Korean naval military officers enrolled at the Republic of Korea Naval Education & Training Command English School were subjected to semi-structured interviews of approximately 30 minutes. Question prompts were used to elicit responses pertaining to motivational disposition, demotivation, attitudes to English ownership and purpose, and any context-specific factors. All four learners have IELTS scores above 7 in each area, and have passed rigorous internal testing, both as part of entrance requirements to the school, and in their various professional capacities as liaison and interpretation professionals. They are clearly highly-motivated and successful learners, and much can be potentially learned from their success. Although the research aims of this investigation are both broad and exploratory, through rigorous planning, piloting, comparison with existing studies, and an open qualitative approach, a number of potential future research paths emerge from the results.

2. LITERATURE REVIEW

This literature review covers the progress of L2 motivation research, from social- psychological, Gardnerian roots to contemporary problems and approaches. The development of the model and its manifestation in the ‘Attitude/Motivation Test Battery’ (AMTB) is outlined, followed by criticisms in the literature aimed at the dominance of the model, the predominantly quantitative approach, and the AMTB instrument itself. A large body of literature from the 1990s, and the suggested adaptations of the model contained therein, is then covered through examination of literature from the cognitive-situated period. A number of qualitative studies are then reviewed, followed by an outline of the recently growing concept of demotivation research. Core items in the body of World Englishes literature are then covered, along with publications that challenge the applicability of the integrativeness concept in many modern contexts. Finally, the review briefly examines Dornyei’s (2005) ‘Ideal L2 Self’ model, which attempts to overcome problems with the integrativeness principle through a reclassification of motivation variables.

2.1 The Socio-Educational Model

The model sought to understand second language-learning in terms of the wider social-psychological tradition, which holds that attitudes towards targets influence overall target-orientated behaviour. Motivation was characterised in terms of four categories:

- Antecedent Factors: Including both biological (e.g. gender/age) and experiential (learning history) factors.
- Individual Differences: Learner variables.
- Language Acquisition Contexts.
- Learning Outcomes (Dornyei, 2001, p.52).

Recent application and revision of the model (especially AMTB implementation- see below), a view where “the socio-cultural milieu... [overrides] all aspects of the model” (Gardner & MacIntyre, 2003a, p.7) has emerged, such that “when considering the process of second language acquisition, it is recommended that close attention be directed to the social context in which the learning is taking place” (2003a, p.7-8). Gardner’s socio- education model, and the AMTB it inspires, has been widely applied. There has however been both criticism of the model, and a major concession from Gardner concerning the purpose and limitations of the model. In the former case, Au (1988) has challenged both the claim to generality of the Integrative Motive Hypothesis (that integrative motivation is positively related to achievement), and the claim that the Integrative Motive itself is unitary (Au, 1988, p.90-1). Moreover, and crucially considering the key hypothesis of Gardner (1985) that social-cultural milieu are causally overriding (see below), Au claims that “contextual considerations with all their vagueness may serve only to render the theory immune to disconfirming evidence” (1988, p.85), such that the central hypothesis is practicably un-testable and therefore unfalsifiable. In defence of this central thesis, Gardner & MacIntyre have claimed that it’s omission would lead to gross over-simplicity (2003a, p.7), adding that the pertinent agenda is to pursue research that “delineates the significant features of social milieu that influence the role of individual differences in language acquisition.” (2003a, p.8-9). They point to applications of the model in radically different contexts such as Kraemer, who indicated generalization was unproblematic in her markedly unique context. (1993, p.97-8). Gardner points to the need for further consideration of the model, noting that “there is no intention… that the model is the true or final one... [but rather that the] true test of any theoretical formulation is not only its ability to explain and account for phenomena... but also its ability to promote further developments and open new horizons” (Gardner 1985, p.166-7).

The beginnings of the integrative/instrumental distinction can be identified in the work of Lambert (1955), where examination of two high level English/French bilinguals led him to speculate that “their excessively high competence in French… was quite likely motivated by their somewhat different involvements with the language.” (Gardner & MacIntyre, 2003a, p.1). Gardner & Lambert (1959) paved the way for modern L2 motivation research (Dornyei & Ushioda 2011, p.40). They applied a test battery (including aptitudinal and attitudinal tests) to 75 Montreal 11th grade learners seeking to confirm their theses that:

- Factors other than linguistic aptitude affect results.
- Research indicating a link between L1 acquisition and desire to integrate into the L1 linguistic community (e.g. Mowrer, 1950) is applicable to L2 motivation.

Their testing indicated that “maximum prediction of success in second language acquisition was obtained from tests of... verbal intelligence, intensity of motivation to learn the other language, students’ purposes in studying that language [as well as]… linguistic aptitude” (1959, p.277-282). Testing applied the Attitude/Motivation Test Battery (AMTB) and employed a crucial distinction between:

- Integrative orientation: “a willingness to be like valued members of the language community” (1959, p.271).
- Instrumental orientation: “pertaining to the potential pragmatic gains of L2 proficiency” (Dornyei & Ushioda, 2011, p.41).

They found that “integratively orientated students are generally more successful... than those who are instrumentally motivated” (Gardner & Lambert, 1959, p.271), leading them to conclude that it is likely that “a strong motivation to learn a second language follows from a desire to be accepted as a member of a new linguistic community” (1959, p.271). It is crucial to note that such research, which was heavily influenced by social-psychological theory (e.g. Dornyei, 2001, p.47-48), was focused in the quite unique Anglophone-Francophone context. The integrative/instrumental distinction serves for Gardner to highlight the extent to which motivations linked to the former type generally indicate more success in L2 learning. Crucially, this involves the claim that “the individual’s attitudes towards the L2 and the L2 community, as well as his or her ethnocentric orientation in general, exert a directive influence on one’s L2 learning behaviour... [which is] in line with the traditional stance in social psychology... that someone’s attitude towards a target influences the overall pattern of that person’s responses to the target (2001, p.48).

In Gardner (1985), links between motivation and success were investigated. Five attitudinal and three aptitudinal measurements were correlated (1985, p.39). Further, the key concept of Integrative Motive was explored. This comprises three elements:

- Integrativeness
- Attitudes toward the learning situation
- Motivation (1985, p.153).

The first category involves integrative goals, interest in the practice of language-learning in and of itself (Dornyei & Ushioda 2011, p.42), and desire to interact with the target social- grouping (Gardner & MacIntyre, 1993, p.158-9). The second category pertains to “the students’ reaction to formal instruction,” and the third to “a combination of the learners’ attitudes, aspirations, and effort with respect to learning the language” (1993, p.159). An important aspect of Gardner’s approach to understanding L2 motivation is the close link between motivation and goals (labelled ‘orientations’), the latter being, strictly speaking, motivational antecedents. (Dornyei, 2001, p.48). In his 1985 research, Gardner reviewed early multivariate studies, later elementary/secondary school studies, and university level studies. Studies covered in the review overwhelmingly involved English-speaking students in Canada acquiring French as an L2. Findings indicated that a majority of considered research showed “very meaningful relations among attitudes toward the second language community, the language-learning context, and motivational attributes” (1985, p.82). The ‘Integrative Motive’ then can be understood as being employed by Gardner as a composite term for this group of attitudinal elements of motivation, “defined as a motivation to learn a second language because of positive feelings toward the community that speaks that language” (1985, p.82-3). Also suggested by the review was a link between motivation and interest in continued L2 study and proficiency (1985, p.83). This model, in various forms, has been applied in numerable contexts, and “factor analytical studies examining data from… various parts of the world have again and again... [attested] to the fact that L2 motivation is generally associated with apositive outlook towards the L2 group and the values the L2 is linked with, regardless of the nature of the actual learning context” (Dornyei 2001, p.50-1). Even in distinctly monolingual contexts such as Hungary, integrativeness has been shown as a key motivational component (Dornyei & Clement, 2001).

2.2 The Quantitative Methodology of the Socio-Educational Approach

The roots of L2 motivation research are dominated by group-centered qualitative enquiry. This should come as no surprise, given that “the founders of the field, Wallace, Lambert, Gardner and their students and associates... were social-psychologists trained within this research paradigm” (Dornyei & Ushioda, 2011, p.212-3). Consequently, the overwhelmingly predominant methodology employed has been questionnaires, with the four main approaches being correlational studies, factor analysis studies, structural equation modelling, and experimental studies (Dornyei & Ushioda, 2011, p.213). The upshot of this qualitative approach is a tendency to focus on the quantifiable, necessitating “the expression of L2 motivation as a mathematical index, numerically compatible in correlational and multivariate statistical analyses with other variable indices” (Ushioda, 1994, p.77-8). Quantitative approaches have almost exclusively employed Gardner’s ‘Attitude/Motivation Test Battery’, which is now examined in detail..

The AMTB is the quantitative operationalised manifestation of the main elements of Gardner’s socio-education model, and includes widely varying constituent scales:

- Attitudes towards French Canadians/Europeans/Language-learning
- Foreign Language Interest
- Integrative/Instrumental Orientation
- French Class Anxiety
- Parental Encouragement
- Motivational Intensity
- Desire to learn French
- Orientational Indexes
- Evaluation of French Teacher & Course (Dornyei 2001, p.53).

Gardner & MacIntyre (1993) examine the reliability of the AMTB (Gardner’s original 1985 version) in measuring the designated variables set out by the socio-educational approach. They examine four questions:

Question 1: Do the subscales test the presumed variables?

As seen from the outline above, the range of subscales is vast. A number of concerns had been raised concerning potential problems with the measurement of affective variables in quantitative approaches such as the AMTB. These focused on the areas of self-reports (e.g. Oller & Perkins, 1978a), where “a substantial portion of reliable variance... may be attributed to extraneous factors such as the desire to look good in one’s own eyes... or the eyes of others... or simply to be consistent in responding to questions of related content” (1978, p.417-418). They further indicated that posited correlations between certain attitudes and proficiency were only partially borne out in research results (Oller & Perkins, 1978, p.90-91). Gardner (1980) defended the quantitative pursuit of affective variable data, noting that “the bulk of research evidence does demonstrate a relationship between second language proficiency and attitudinal/motivational variables” (1980, p.268). Gardner & MacIntyre echoed this with specific reference to the construct validity of the (1985) AMTB, stating “the results support the conclusion that the subtests measure what they are intended to measure... and that they correlate meaningfully with measures of second language achievement” (1993, p.188). Given these findings, and the varied applications in different contexts of the AMTB, the test has indeed been shown to engender high levels of construct and predictive validity (Dornyei 2003, p.52).

Question 2: Can the variable class distinctions be justified empirically?

Gardner & MacIntyre also questioned the homogeneity of AMTB constructs. Despite some concerns relating to the similarities of variables, factor analysis “provided very strong support for the higher order constructs... [as] the four major constructs, Integrativeness, Attitudes Towards the Learning Situation, Motivation, and Language Anxiety form clear factors” (Gardner & MacIntyre, 2003, p.188). Further, subsequent studies undertaken in a range of environments also reinforce this conclusion; as Dornyei & Ushioda note: “[varying studies] have again and again produced a factor made up of all, or many of... [the AMTB] components” (2011, p.42-3).

Question 3: Are the studies internally consistent?

Gardner & Lambert, citing the variance in measurement strategy (e.g. 2003, p.161) explored possible links between measurement strategy and results. Despite earlier socio- educational approaches using mixed procedures (e.g. Osgood, Suci, & Tannenbaum, 1957), later studies utilised the Likert procedure for analysing assessment strategies. Considering applications of the latter type, Gardner & MacIntyre stated that “at least as far as internal consistency reliability is concerned, Likert assessments provide comparable measures” (1993, p.162), and that “the subtests adapted from the AMTB [predominantly of the Likert format]... correlate more highly with the objective measures of proficiency than do the other forms of measurement” (1993, p.189).

Question 4: Is the integrative/instrumental distinction inclusive of all possible motivations?

This question area is key, because significant difference exists between links found (or not) between orientation type and achievement, most notably between Gardner & Lambert 1959 (significant link) and Gardner & Lambert 1972 (few significant correlations) (Gardner & MacIntyre 1993, p.162). Further, mixed results are found when individual orientation items are applied and Cloze test methodology is utilized (1993, p.163). Crucially, the main conclusion (despite linking measurement strategy to many of the varying results among comparative studies), was that “affective variables play a significant role in second language-learning” (1993, p.191). As will be seen below, despite a long-standing influence on practically all L2 motivation research (e.g. Dornyei 2001; Dornyei & Ushioda, 2009) criticism of the AMTB (and Gardner’s model promoted thereby) snowballed at the turn of the 1990s. However, to date the AMTB is still the sole standardized published L2 motivation test (Dornyei 1991, p.52-3), a sign perhaps not only of the allure of a seemingly universal instrument in a profession with a plethora of contexts, but also of the absence of genuine, ‘bottom up’ research exploring alternative theories. As seen below, the main body of ‘criticism’ of the model is actually fine tuning of certain elements within a shared paradigm, and certainly does not constitute a full-blown, Khunian paradigm shift. Little wonder then, that the key instrument of that paradigm endures.

2.3 Criticism of the AMTB/Socio-Educational Model

Crookes & Schmidt (1991) identified two major problems with the Gardner-driven body of L2 research. They noted a severe limitation in the state of L2 motivation as it stood, namely that the relatively limited definition of motivation provided by a certain school of applied linguistics, and the accompanying social-psychological locus upon attitudinal and social aspects, falls far short of the extension and depth of the term ‘motivation’ as used in the L2 teaching profession (1991, p.469). To a great extent, their proposed research agenda to address these points defined the major subsequent trends in L2 motivation research. They suggested a move away from questionnaire/AMTB methodology, proposing use of observational measures, ethnographic work, action research, introspective measures, and experimental studies (1991, p.502).

Skehan (1989) points out that almost all L2 motivation research prior to writing was heavily influenced by the work of Gardner and his associates (e.g. 1959; 1972; 1985). Crookes & Schmidt concur, noting that “motivation has consistently been linked with attitudes towards the community speakers of the target language” (1991, p.470-1). They note the five key hypotheses shared by all such research:

- An integrative motive positively effects SL achievement
- Cultural beliefs influence both the development of an integrative motive and the degree of relatedness between it and SL achievement
- Integratively motivated learners succeed because they are active
- Integrative motives are the cause of SL achievement (1991, p.473)

They directly challenge the assertion that integrative motivation is empirically superior, noting various conflicting results in meta-analyses (e.g. Oller, 1981b; Au, 1988). Furthermore, they note the suggestion by a great number of contributors that the causal relationship between Integrativeness and achievement may not be justified (1991, p.473- 474). Crookes & Schmidt note that “despite the tripartite distinction between cognition, motivation, and affect... [Gardnerian research has] tended to group affect, especially attitudes, and motivation together” (1991, p.471).

Crookes & Schmidt also point out that by the time of Gardner (1985), the integrative motive “is no longer equivalent to attitudes towards the target language community... [or] a score on the [AMTB]”. Rather, the term is applied inconsistently across a number of studies to various factor analysis reductions of attitudinal/integrative composites of AMTB subscales (1991, p.475-6). In addition, many commentators point out the inconsistently applied terms in research applied before this period (e.g. Skehan, 1989).

Crookes & Schmidt point to two features for an improved research agenda:

- A move from the terminology and conceptualisations of social-psychology, such that greater congruence is achieved between research terms and terms used by teachers (1991, p.468-70).
- A move from the ethnolinguistic focus to a concern for situated variables (Dornyei & Ushioda, 2011, p.46)

Even during the 1990s there was then discomfort with the terminology and consistency of application, a point that would form the basis of Dornyei’s ‘Ideal L2 Self’ model (see later section.) However, during the 1990s a shift from ethnolinguistic to situated factors would form the basis of research and debate.

2.4. The Agenda for Change

As Dornyei & Ushioda succinctly put the matter, “the common theme underlying the new [agenda]... was the belief that motivational sources closely related to the learner’s immediate classroom environment have a stronger impact on the overall L2 motivation complex than previously thought” (2011, p.47-48). This new agenda was not a Khunian paradigm shift as such, but rather, as Dornyei notes, it is a matter of expanding the existing structure “by incorporating additional factors that... offer increased explanatory power [for teachers]. Crookes & Schmidt’s dual-faceted call for both wider scope of factors and more easily applicable terms borrowed its framework from Keller (1983):

- Interest (Intrinsic motivation, inherent curiosity etc.)
- Relevance (Feeling that instruction relate to needs, goals etc.)
- Expectancy (Perceived chance of success.)
- Satisfaction (Pertaining to extrinsic and intrinsic rewards.) (Adapted from Dornyei & Ushioda, 2011, p.49-50).

It is important to note that broadly speaking, Crookes & Schmidt’s 1991 critique did not aim to completely overhaul the socio-educational framework of Gardner. Rather, their project is “characterised more in terms of broadening the existing theoretical framework through integrating cognition motivation concepts, rather than in terms of discarding social- psychological concepts altogether” (Dornyei & Ushioda 2011, p.47). As they put it, they “would certainly not dispute that language-learning takes place within a social context, nor that socially grounded attitudes may provide important support... for motivation... [or that] there are no interesting relationships among social contexts, individual attitudes, and motivation... [but rather that] this particular approach has been so dominant that alternative concepts have not been seriously considered” (Crookes & Schmidt, 1991, p.501).

Given this new agenda for expansion upon existing principles, it is perhaps unsurprising that there was a busy period of discussion surrounding possible adaption, future research directions, and potential new factors. Dornyei (1994a) proposed an extended model that aims to be appreciative of the diversity of L2 motivation. His ‘process model’ is essentially a three level framework:

- Language Level: A composite term pertaining to cultural, community-orientated, intellectual, and pragmatic aspects of L2 motivation (Dornyei & Ushioda, 2011, p.51). It is indeed an extension of the now familiar Gardner-inspired socio-education approach, in that it “can be described by two broad motivational subsystems, an integrative and an instrumental motivational subsystem, which... consist of loosely related, context-dependent motives” (Dornyei, 1994a, p.279). The former subsystem pertains to learner-specific affective L2 related motives, which accounts for attitudes to both L2 learning and target cultures, and related affective factors. The latter subsumes extrinsic motivational aspects.
- Learner Level: This is similar to Tremblay & Gardner’s (see below) ‘Self-efficacy’ construct, but includes additional elements of causality and perceived competence. Its main components are ‘need for achievement’ and ‘self-confidence’. The former is explicitly linked to learner-specific intrinsic elements, while the latter encompasses “various aspects of language anxiety, perceived L2 competence, attributions about past experiences, and self-efficacy” (Dornyei, 1994a, p.279).
- Learning Situation Level: This contains course-specific (Interest, relevance, expectancy, satisfaction), teacher-specific (affiliative drive, authority type, socialisation conditions), group specific elements (goal-orientedness, norms/rewards, cohesion, goal structures.)

In addition to the Crookes & Schmidt’s (1991) contribution, and that of Dornyei outlined above, a third persuasive contribution took place in the form of Oxford & Shearin (1994). They point to a problematic feature of Gardner’s socio-educational model where the main emphasis is not “elaborating on the possible range of motivational antecedents... but on determining whether motivation has been aroused and specifying the learner consequences of this arousal in relation to the impact of other non-motivational factors” (Dornyei 2003, p.106-7). As Gardner puts the matter, “the source of the motivating impetus is relatively unimportant, provided that motivation is aroused” (Gardner, 1985, p.169). Oxford and Shearin explicitly challenged this conclusion, noting that “while this conclusion might be true for researchers, quite possibly the source of motivation is very important in a practical sense to teachers... [as] without knowing where the roots of motivation lie, how can teachers water those roots?” (Oxford & Shearin 1994, p.15). In fact, all three papers “[were similarly] ambivalent in regards to Gardner’s theory... [as] there was the feeling that the existing social-psychological construct is not as applicable in some areas of the L2 learning process as in some others” (Dornyei 1994, p.515). In addition, all three of the papers outlined above share an anxiety concerning “the recognition that the last fifteen years have brought about a major shift in mainstream [i.e. non-L2 specific] psychological theories, which... could and should be reflected” (1994, p.515-6). Gardner actively participated in this phase of academic discussion, debate and review. In Gardner & Tremblay (1994a), it is acknowledged that moves to expand the agenda of L2 motivation are positive (1994a, p.359). They challenge the assertions that the Gardner-driven social- psychological dominance has been limiting or is limited, and show that the Gardner-driven body of work is consistent with the suggestions for closer ties between research and L2 language teaching practice. They also challenge the widely expressed view that the social- psychological tradition has led to the underplaying of alternative methodologies and theories (1994a, p.362-3). Further, they go on to validate the suggested three-level approaches of both Dornyei (1994a, ‘process model’) and Crookes & Schmidt (1991).

Where they deviate from the three articles is that they consider the Gardner-research misunderstood and already providing for the integration of various motivational theories in a second language learning measurement instrument. (Gardner & Tremblay, 1994, p.364) They also point out that by and large suggestions made in Dornyei’s (1994a) article are not backed up with empirical research (1994, p.368). Oxford (1994) perhaps provides the most sobering summary of this ‘debate’. All three of the ‘provocative’ original articles point out need for future research in various areas, whilst the Gardner & Tremblay response indicates a need to back up suggested benefits with research. This circumstance seems to imply general agreement, and Oxford herself admits that while writing her cooperative contribution she “was standing on Gardner’s own shoulders” (2004, p.513). The debate can be understood as a hashing out of finer points against a background of consensus regarding the need for expansion and research in the area of L2 motivation. As Dornyei sums up, “L2 motivation research has undoubtedly gained new momentum... a number of new ideas have been suggested and first reactions to these expressed. What we need now... is construct validation and pertinent empirical research... we can foresee in future L2 motivation research a dynamic interplay of established motivational concepts grounded in a social- psychological approach, and constructs rooted in other psychological fields and approaches” (2004b, p.522).

2.5 The Need for Qualitative Enquiry

Ushioda (1994) notes that in quantitative approaches, motivation tends to be operationalised “as an affective variable implicated in second language-learning achievement” (1994, p.78). However, there are severe limitations to this approach, as the three central components of the Gardnerian theory, as well as the qualitative difference between integrative/instrumental orientations, are expressed in “differences of degree rather than in quality” (1994, p.78). Pointing to the work of Ames (1986), Ushioda notes a researchable link between qualitative accounts of motivation, positive thinking, learner autonomy, and strategies (1994, p.79; also Ames, 1986, p.238). The outcome is that effective motivational strategies hinge on “preservation of a positive self-image and sense of competence... largely achieved through selective attribution processes, whereby successful performance is ascribed to personal ability, but failure or poor performance to external or temporary causes” (1994, p.78-9).

In Ushioda’s 1991 (Ushioda, 1991) study, open-ended, two-stage interviews were utilised. This is mainly due to the “exploratory” nature of the research agenda (Ushioda 2004, p.80- 1). Results from transcribed and content-analysed interviews with 20 university-level learners of French indicated qualitative motivational differences. Overall, 63 coded variables appeared on average only 4.3 times per interviewee, with a small section of the sample providing a disproportionate number of the overall code yield (2004, p.81). From the coding emerged the following descriptive categories:

- Academic Interest
- L2 Learning Enjoyment
- Past L2 Learning Experience
- Personal Satisfaction
- Desired Levels of L2 Competence
- Personal Goals
- Feelings About French-Speaking Countries & People
- External/Course-Related Pressures and Incentives (2004, p.81)

Most frequently cited variables included earlier positive experience and positive goals. This led Ushioda to conclude that such temporal aspects warranted further investigation; she notes that “attempts to classify learners as simply instrumentally or integratively orientated on the snapshot basis of responses to a closed set of items may overlook subtle differences in how individual learners prioritise future goals... [such that] some learners may become classified as instrumentally motivated, yet in fact have little clear perception of what their specific goals might be” (2004, p.81-2).

2.6 Attributional Processes

In both of the systems proposed as extensions to the social-psychological model, explicit attention is paid to the role of causal attributions (Dornyei & Ushioda, 2011, p.55). However, despite the obviously increased relevance of expectancy of success in L2 (compared to L1) acquisition, wider research on the application of attribution theory (Weiner 2007) to L2 acquisition is limited. Dornyei & Ushioda (2011, p.55-57) link this to the dominance of quantitative methods in L2 motivation research (as seen above and derivative from Gardner AMTB led roots of this field). The sheer complexity of causal attributions make them difficult to investigate using the self-report questionnaire method (2011, p.55). This has led many contributors (e.g. Skehan, 1989; Tremblay & Gardner, 1995) to call for more qualitative research to be conducted in this area.

In Ushioda’s 1996 & 1998 investigations (Ushioda, 1996; 1998) a two-stage qualitative interview method was employed to investigate attributional patterns with Irish learners of French. The main findings were that maintenance of positive self-concept and potential for success in spite of negative experiences depended upon:

- Attribution of L2 outcomes to internal factors
- Attribution of L2 outcomes to temporary & surmountable factors (Dornyei & Ushioda 2011, p.55-56)

Interestingly, these are perfectly in line with input from wider educational psychology (2011, p.56). Ushioda notes that by the time of Gardner & MacIntyre (1992), significant changes had emerged in the role of success as a cause (1996, p.239-40). This change “reflects a notable change from the theoretical position… that strong motivation is a cause rather than a result of L2 learning achievement” (1996, p.240). Their position indicated a dynamic account of causation, for which a simple instrumental/integrative distinction is insufficient. Ushioda suggests that what is required in order to explore the nature of such cyclic changes, is a more ‘introspective’ methodology (1996, p.241). Examining motivational changes arising since first stage interviews, she found that personal goals can “crystallise, strengthen, or change at different stages... [such that] intrinsically motivated learners may also develop an instrumental motivation, or lose some of their intrinsic motivation in the face of exam-orientated motivation” (1996, p.243). It was also found that two non-L2 related factors, competing study priorities and personal emotional crises, were significant factors (1996, p.243-4).

In Williams & Burdens’ 1999 investigation (William & Burden, 1999), 36 English learners of French (randomly chosen from 3 schools, 50% male/female, and split equally divided into Year 6/7/9 & 10-12) were interviewed. Individual content-analyses (by each researcher) were combined and discussed until agreement was met on overarching constructs and descriptive terms. Interviews were semi-structured and one-to-one, lasting about 25 minutes in each case. They set out to investigate the little-examined role of attributions in second language-learning success. They utilized a semi-structured interview method aimed at uncovering data for the following research questions:

- How to learners of different ages conceptualise the notion of doing well in their attempts to learn a foreign language?
- How to learners of different ages determine how well their learning of a language is progressing?
- What reasons do learners of different ages attribute to their perceived successes and failures in learning a language?
- What actions do learners of different ages see as necessary in order to do well in learning a foreign language? (1999, p.195)

They found significant correlations between age and complexity of attributions with regards to all research questions. As they put it “individual attributions for success and failure appear to be formed by a complex interplay among internal feelings and developmental stage, external influences, and social context… [meaning] that it is important to understand, therefore, the way in which individuals make sense of external influences to shape internal attributions... [including] the way teachers teach, teachers’ aims and their beliefs about learning and the nature of education” (1999, p.199). They content- analysed their transcribed interview data in a grounded method, ensuring the possibility of open interpretation at the results stage with no pre-imposed conditions. Importantly, Williams & Burden acknowledge that their study is of limited scale, and is a basis for, rather than an attempt to provide, a descriptive framework. Crucially for this inquiry however, they show the value of in-depth qualitative study and the potential to show (as they indeed managed to, see 1999, p.199) that the range of factors attributed to success and failure by learners is by no means clear at this stage in terms of either number or importance. They indicated that larger samples, temporal changes in attributions, and gender could be factors in fruitful lines of similar enquiry (1999, p.200).

2.7 Demotivation

Recently a new trend in L2 demotivation research has emerged. Demotivating influences can be broadly understood as “specific external forces that reduce or diminish the motivational basis of a behavioural intention or an ongoing action” (Dornyei, 2001, p.143). It is dissimilar in application to the motivational psychology concept of ‘amotivation’ (e.g. Deci & Ryan, 1985), which primarily concerns the absence of motivational antecedents. Demotivation is a concept largely borrowed from the field of instructional communication (Dornyei & Ushioda, 2011, p.140-1), where it has been widely researched using both quantitative and qualitative methodologies (e.g. Gorham & Christophel, 1992). Ushioda (1998) examined the role of self-conception of motivation, demotivation, and temporal aspects of motivation amongst a sample of 20 university-level Irish learners of French. The second-stage element focused on demotivation elements; it was found that “almost without exception... these demotives related to negative aspects of the institutionalized learning context, such as particular teaching methods and learning tasks” (Dornyei & Ushioda, 2011, p.146). She used semi-structured, open-ended interview methods to identify that “responses overwhelmingly targeted negative aspects of the institutionalised learning framework, rather than personal factors such as falling grades or negative self-perceptions of ability... [such that] projecting the responsibility of their loss of motivation onto external causes... [learners] limit the motivational damage and dissociate the negative affect they are currently experiencing” (Ushioda, 2001, p.86).

Oxford (1998) conducted content-analysis of essays written by American high school students who responded to prompts. Major response-themes were:

- Teacher-learner personal relationships.
- Teachers’ attitudes towards course and materials.
- Teacher-learner style-clashing
- Classroom Activities

Dornyei (1998) was an exploratory, qualitative investigation aimed at learners identified as being demotivated. (Dornyei, 2001, p.150). Structured interviews were employed, but some pre-prepared questions were also used. Demotivating factors were categorised, and ordered in terms of importance with respect to each subject. In order of frequency and importance, the main findings were:

- The Teacher: Personality/Commitment/Competence/Methods
- School Facilities: Group Size/Staff Inconsistencies
- Experiential failures and diminishing confidence
- L2 Attitudes (Language)
- Compulsory nature of study
- Additional L2 interference
- L2 Attitudes (Community)
- Fellow Learner Attitudes
- Coursebook Issues (Dornyei 2001, 152-3)

In Korean contexts, Kim (2011) employed a mixed-methods approach to explore the reasons for demotivation in Korean elementary school students, using questionnaires to assess 6,301 grade 3 to 6 learners, and interviews with 17 English teachers. He found that demotivation increased as learners progressed through the grade system (e.g. 2011, p.22). He also noted the importance of the fact that in a Korean context, the impact of extra- school private-school and tutoring classes and the high expectations and long working hours affected demotivation attributions (2011, p.22). This shows that as we extend the range of factors in preliminary enquiries in an attempt to address a wider range of factors, it seems to follow that difference between contexts becomes more important (as can be seen from Kim’s 2011 work). In the related field of investigating ‘willingness to communicate’ (WTC), Wen & Clement suggest that a “culture-specific interpretation of the WTC construct provides an enriched concept” (2003, p.34). Their work points to the Confucian roots of Chinese educational environments as one of many factors determining WTC performance. Of course, limitations run vice-versa; as Kim acknowledges: “the results are limited to the EFL context in Korea and cannot be generalized to other EFL contexts... an active co-construction of knowledge is the main characteristic of qualitative verbal protocols, and the interpretation of the interviews… is no exception” (2011, p.26).

At this stage it is important to speak of motivational research in light of changing attitudes towards English. As seen above, a key element of the ‘integrativeness’ concept involves attitudes towards both the language and the ‘target’ community of speakers. However, in light of a recent proliferation of literature in the area of ‘World Englishes’, many commentators have begun to question the validity of this concept in many contexts. In the 21st Century, ‘non-native’ speakers of English outnumbered ‘native-speakers’ (Graddol 1997, p.10-11). As a result, many commentators have begun to move away from geo- political classifications of English, with Modiano (1999) suggesting a model that affords status to proficient speakers of communities, such that goals of promoting English for global communication and international cooperation at all levels are upheld at the expense of “creating barriers, or upholding systems of membership and exclusion” (1999, p.27). Following the English Today debate, issues of standardisation, propriety, ownership, diversity in purpose, and (crucially) the future of English, are at the forefront of academic debate. The spread of English means that the range of English learning contexts has proliferated (Sridhar & Sridhar, 1986, p.3). Motivation research has to be explored with this hugely important process in min.

2.8 Ownership of English

Monochromatic views of American English or British English as global standards have dissipated. Attempts at quantitative justification of such ‘standard Englishes’ have been exposed as qualitative (e.g. Trudgill, 1999). The expansion on English in the outer circles and the nativisation/acculturation it entails has unsurprisingly challenged the heavily codified inner circle varieties (e.g. Kachru, 1985). The emerging ‘World Englishes Perspective’ looks to establish English as the ‘democratic’ property of a global community, such that as “English is an international language... no nation can have custody over it” (Widdowson 1994, p.385). Scepticism surrounding the potential for English to fulfil such a role has been expressed by Wierzbicka (e.g. 1997; 2006), who insists on the impossibility of extrapolation of word/meaning from cultural contexts. Similar sentiments have been expressed where English is connected to colonial factors in areas such as Africa (e.g. Ngugi, 1986). However, the mere absence of qualitative constraints from American or British English ‘standards’ is only a precursor to a developmental process of ownership. This can be understood as the journey from abrogation, a “refusal of the categories of imperial culture, its aesthetic, its’ illusory standard of normative or correct usage” to appropriation, where“the language is made to bear the burden of one’s own experiences… [and] adopted as a tool and utilized” (Aschroft, Griffith, & Tiffin, 1989, p.38-9).

2.9 English as a Lingua Franca

As shifts in attitude abound in the world of linguistic academia, the function of English, as used, in the world, continue to fly in the face of narrow accounts of English, and by implication any conception of an ‘English language community’. As Seidlhofer points out, whatever the political take on English proliferation, the role of English as a lingua franca is undeniable (2005, p.339); moreover “as a consequence of its international use, English is being shaped at least as much by its non-native speakers as by its native speakers... [as] the vast majority of verbal exchanges in English do not involve any native speakers of the language at all” (2005, p.339-40). Jenkins points out that recent snowballing of awareness in the academic sense will not necessarily result in any changes in pedagogical practice (2004, p.174-5). What is rarely, if ever, mentioned in the literature is the attitude of actual learners to English, and their conceptions of what ‘English’ is; after all, it is they who are investing such personal time and effort into the acquisition of the language! It would seem absurd if the academic community should undertake such a shift in thinking about the nature and purpose of English without such a shift in thinking having already been manifested in the community of learners. Accordingly, the need for a discussion of L2 motivation in terms of changing attitudes towards the status of English is implicated in any forthright attempt at gauging motivation from the learner perspective (after all, what other perspective is there?).

2.10 Integrativeness in World Englishes Contexts

Sridhar and Sridhar identified a ‘paradigm gap’ between SLA theory and the emerging multiplicity of Englishes and English-learning contexts (1986). Coetzee-Van Roy has extended this line of enquiry to indicate both that “second-language acquisition theories that rely on any assumption of integrativeness should not be applied uncritically to sociolinguistic contexts where learners are acquiring a variety of World English... [and] the concentric circle description of the sociolinguistic realities of World English speakers needs to be adjusted” (2006, p.437). As seen in the above review, reliance on the notion of integrativeness runs throughout practically the whole body of L2 motivation research. However, there has been wide ranging criticism of the notion (e.g. Dornyei, 2003, p.67; Nickels, 2005, p.234-5), primarily on the grounds of “the simplex views of the identity of second language learners and the incorrect assumptions made about the sociolinguistic contexts of many learners of English as a second language across the world” (Coetzee-Van Roy, 2006, p.441). Criticisms relating to the former pitfall advocate more complex views of language-learner identity. In World Englishes contexts, the implication is that “there is no attempt for the user to be like a native speaker of English” (Smith 1983, p.7). Crucially, considering the context of the present enquiry, Shaw discovered that in Asian contexts learners do not perceive themselves as “learning English so that they can change themselves and become more like native speakers” (1983, p.24). He goes onto to criticise the geo-political models of English and the relationships between groups of ‘native speakers’ and learners they presume. Rather, he suggests that the ‘target group’ may be a localised elite (1983, p.24). In conclusion, Shaw suggests that “the whole aspect of integrative motivation should be re-examined in terms of a desire among learners to join an indigenous group of English language speakers or a vague international one” (1983, p.442). In light of these and similar findings, the role of the notion of integrativeness in L2 motivation research has declined pointedly in recent years (Dornyei & Ushioda, 2011, p.85); Ushioda & Dornyei ask:“[As] the debate about the integrative concept has intensified and taken on a new turn, prompted by the burgeoning discussions... about the global spread of English, a basic question we have begun to ask is whether we can apply the concept of integrative orientation where there is no specific target reference group of speakers” (2009, p.2-3).

2.11 The ‘Ideal L2 Self’ and the ‘Ought-to L2 Self’

Dornyei has recently promoted the ‘L2 motivational self system’, which comprises three key components:

- Ideal L2 Self: L2-specific components of perceived ideal self. This subsumes elements which under Gardnerian systems would be classified as integrative/internalised instrumental motives.
- Ought-to L2 self: Perceived attributes one ought to possess in order to meet external expectations, and to avoid possible negative results. This also includes less- internalised/extrinsic instrumental motives.
- L2 Learning Experience: Immediate elements relating to teacher, curriculum, peer group etc. (Dornyei & Ushioda 2011, p.86)

This system borrows heavily from the ideal/ought-to distinction expounded by Higgins (1987; 1998). Instrumental motivation can be of two types:

- Promotion-focus: The ‘ideal-self’ involves hopes, aspirations, goals, and desired end- states.
- Prevention focus: The ‘ought-to self’ involves obligations, responsibilities, and avoidance of negative outcomes. (Dornyei & Ushioda, 2011, p.87)

A number of quantitative studies conducted in a range of contexts have suggested that this re-conceptualisation of integrativeness in terms of the ‘L2 self system’ is valid (e.g. Csizer & Kormos, 2009; Ryan, 2009). In many countries, English is perceived as the sole global language (Canagarajah 1999; Phillipson 1992). Applying the L2 self system in a case study of 2 Korean students, Kim (2009) notes the socially orientated strength of this view of English as a sole global language necessary for employment opportunities, both domestically and internationally (2009, p.290-1). Similarly, in Japanese contexts, such a view of English as a route into domestic and international social elites has been noted (Zeng, 1995). Kim noted that where English as a global language was more closely related into personal contexts by learners, it “may provide a powerful motivator for learning English” (1999, p.290), i.e. be internalized into the ‘ideal L2 self.’

L2 motivation research has largely remained within the Gardnerian paradigm, and despite calls for exploration of more situated variables and exploratory research, the body of literature is still overwhelmingly quantitative. Amidst wrangling over the finer points of essentially the same paradigm, a new challenge to the Gardnerian school has emerged: that of the ‘World Englishes’ view, whereby the sheer relevance of the notion of integrativeness is called into question, due to the absence in many contexts of any one ‘target group’ or ‘culture.’ An emerging focus on the concept of ‘demotivation’ also warrants consideration in any enquiry.

2.12 Future Research

At this stage it seems there are two important factors. First, genuinely non-presumptive, qualitative research should be undertaken in a range of contexts, such that new, context- specific factors can be identified, and existing research challenged where necessary. Second, the applicability of ‘integrativeness’ as a concept needs to be directly examined, particularly with reference to learners’ attitudes towards the target/function of English. Given the broad range of research objectives, and the exploratory nature of the work as a whole, this project has not attempted to verify any specified hypotheses. Rather, the aim has been to contribute to the modest yet both growing and insightful body of qualitative research in the field of L2 motivation. A number of recent commentators and researchers have noted the limitations of AMTB-inspired quantitative approaches and called for an expansion of research into previously underrated or undiscovered factors in L2 motivation. (e.g. Ushioda, 1994; Dornyei & Ushioda, 2011). Given the uniqueness of the L2 learning context and subjects, the aim is rather to elicit as much data as is viable concerning L2 learning motivation as perceived by subjects, and where appropriate, compare this with results from existing qualitative studies. Moreover, this study aims to respond seriously to calls for an expansion of factors under consideration in the present tradition of L2 motivation research and approach the data collected from a standpoint with as few preconceptions as possible. Kim (2011) has attributed many factors affecting demotivation to specifically Korean circumstances. Further, in areas such as Anxiety and Willingess to Communicate, significant differences in data have been attributed to the locality of East- Asian subjects (e.g. Yan, 2001; Yashima, Zenuk-Nishide, & Shimizu, 2004). Considering such findings, it seems reasonable to acknowledge the possibility of specific factors that cannot be predicted in advance relating to both the locality and profession of the subjects of this study. Further, qualitative approaches have been shown to be successful when approaching understanding of learner-specific personal factors in Anxiety (e.g. Yan & Horwitz 2008) and in L2 motivation/demotivation (e.g. Ushioda, 1996).

2.13 Research Aims

1. To establish the motivational orientation of the learners.
2. To investigate the role of teachers, textbooks, and institutions in learners’ motivation.
3. To investigate the role of motivational strategies.
4. To investigate changes over time in learners’ perceptions of their motivation
5. To examine the role of previous experience in learners’ motivation.
6. To examine the role of success/failure attributions in learners’motivation.
7. To examine demotivation in learners’ perceptions.
8. To examine any arising context-specific factors.
9. To thoroughly investigate learners’ attitudes toward function, style, and ownership of English.

3. METHOD

3.1 Semi- Structured Interviews

Interviews have been shown time and time again to be a method of investigation that can provide detailed data that is unobtainable through questionnaires. As Dowsett notes, “you can produce extraordinary evidence... that you don’t get in structured interviews or questionnaires... no matter how open-ended and qualitative you think your approach is” (1986, p.53). Semi-structured interviews have been successful in recent qualitative approaches to L2 motivation (e.g. Ushioda 1994, 1996; Williams & Burden, 1999), and have been applied in this study. The semi-structured interview allows a list of topics to be addressed, whilst new and original elements can be pursued as and when they arise (Dornyei, 2001, p.38). As Nunan points out, this method has the advantage of giving the researcher “a degree of power and control over the course of the interview… [and] a great deal of flexibility” (1992, p.150). Given that this study aims to examine findings in light of standing principles in L2 motivational academia, compare such findings to the existing body of qualitative L2 motivation research where such comparison is viable, and also discover any new factors relating to the subjects past and present L2 motivation, such an approach seems best suited. This is especially the case given that evidence exists in related areas of L2 research that locality-specific factors are significant to findings (as noted above with respect to Kim’s 2011 study).

Although open ended questioning and semi-structured interviews were employed, piloting was an indispensable component of the overall research strategy. To avoid ambiguous questions, piloting with samples is as important stage of the process (Nunan, 1992, p.151). Given the broad, exploratory aims of this work, very general and open-ended questions were employed using two subjects from the same class as the four final subjects. Questions were mainly compiled by referring to the main conclusions of studies into L2 motivation. However, novel responses were encouraged through follow-up of responses such that problematic aspects could be rectified and new question strategies established for the main study. The following questions were used:

[...]