Appraising English-Speaking International University Student Attitudes towards Their Elite Second-Language Education and Status

About a third of the global population learns English as a second language (L2). L2 students routinely read English literature. Many L2 readers are members of cultures with painful colonial pasts. The English language functions in part to support western cultural and economic domination. L2 students have a complex identity, as members of their own culture, but also as elites having options in the global economy. This study used content analysis, content clusters and Appraisal analysis to explore how three cultural groups with colonial histories responded to an English-language translation of a Bulgarian poem about an elite, L2-educated character in a colonial situation. Results suggest that bilingual youth do not see themselves as elite, and that authentic literature in translation helps them perceive status and positionality.


Introduction
Studying English shapes the identities of 2 billion L2 students worldwide (Crystal, 2000).Most L2 students do not try to assimilate to an English-speaking culture, for example British or American, but combine elements of local with aspirational aspects of international culture, creating hybrid identities (Levitt, 2001).Reading English-language literature is part of this process, as it is used to teach critical thinking and culture performance, vocabulary and grammar, idiomatic collocations and discourse structures (Paran, 2008).Literature motivates students to remain engaged in the difficult work of L2 reading (Mantero, 2002).L2 students often read translations of literature by world authors.This study explored how readers from three cultures with colonial pasts constructed the meaning of a text where the main character was both colonised and a beneficiary of L2 education.The text selected was 19th century Bulgarian poet Hristo Botev's poem "Stranger".

Literature Review
Reading is subjective.When learning to read words, children use their own experiences to constrain possible meanings, and meanings attributed reflect each individual's unique experiences of selected real-world referents (Clifton, 2000).Individual readers adjust meanings, building up an interpretation as they move through a text.Thus, reading is highly intertextual, particularly in a L2, as each word and phrase is reviewed in relation to previous knowledge in both languages (Van Oostendorp & Goldman, 1999).Reading activates memories and complex assemblies of previously constructed meanings, which readers use to process new texts (Carter & McRae, 1996).Reading is also cultural.In learning their mother tongue, children absorb extensive, culturally-specific figurative language (Glenberg & Robertson, 1999).For example, in English, time is conceptualised as a path moving ahead into the future, with the past receding behind where in Chinese it is represented as a vertical continuum, with the past being "above" or "high" in a sense denoting reverence (Dancygier & Sweetser, 2014, p. 172).Every language uses myriad culturally-specific metaphors, which organise how members interpret a text.Members of a particular culture are trained from infancy to use culturally normative values when attributing meaning in areas such as family, school, work, authority and success (Coulson, 2001).These meanings are repeatedly rehearsed in social life, ensuring that members of a culture hold similar views (Gergen, 2001).
There is a political dimension to L2 students reading English literature.English-language teaching (ELT), enmeshed in the processes of globalisation, is often seen as "serv[ing] a post-imperial capital agenda" (Philipson, 2001, p. 185).English-language education is valued as a ticket-of-entry into the global economy, with western institutions gate-keeping, setting standards for teacher training, and administering standardised tests (Seidlhofer, 2005).States eager for economic benefits enact national language policies using ELT industry norms, which reshape local cultures (House, 2003).This creates passive educational participants, positioning western native-speaker teachers as cultural brokers, valorising western cultural content, and inculcating a western mindset and values in students (Pennycook, 2014).ELT education often casts pedagogical constructs in western terms, undercutting their authentic instantiation in local contexts (Jenkins, 2013).The focus on standard over world Englishes in local curricula reflects its perceived role in the global economy, which still retains a monolinguistic western ethos and identity (Seidlhofer, 2001).For L2 students, then, English is a colonising force, and their expectation is that their L2 education will reward them economically for assimilation (Mauranen & Ranta, 2009).
This study explored the responses of Lebanese, Hong Kong and Bulgarian participants to a literary character who is positioned in ways similar to themselves.Content analysis was used to identify frequently-realised themes and nominal-group clusters.Appraisal analysis was used to detail subjective attitudes.Regularities across the subcorpora offer a way of understanding the global identities English users are forming.Differences highlight cultural variations among participant groups.Research questions included the following.How do cultural values and subjective attitudes shape how postcolonial youth in three different contexts read the colonial situation depicted in this poem?What do their responses show us about their understanding of their privileged education, their likely future wealth and opportunities, and their hybrid identities?

Method
Qualitative data in the form of written personal responses was obtained from three participant groups.

Participants
Participants included 109 Cantonese-speakers from Hong Kong, 116 Arabic-speakers from Lebanon, and 86 Bulgarian-speakers from Bulgaria.There are about 1.1 billion Chinese-speakers, and 385 million Arabic-speakers worldwide (OECD, 2012).A south-Slavic variant, Bulgarian is part of the Slavonic language family spoken by about 300 million people in Europe and Asia, of whom about 34 million are south Slavic speakers (Corbett & Comrie, 2003).Having emigrated globally for decades, there are now as many Bulgarian-speakers living abroad as in-country (Smilov & Jileva, 2009).All three language groups have a global diaspora (Ruggles & Silverman, 2009).In all three, language remains an esteemed element of heritage and identity (Benmamoun, Montrul, & Polinsky, 2013).Participants were local native-speakers, proficient enough in English to undertake L2 university courses.All were 18-to 22-year-olds, majoring in humanities, the sciences, business, engineering, computing and tourism, who had registered in a general education class using English-language texts to teach university composition.

Text
The text selected was 19th-century Bulgarian poet Hristo Botev's "Stranger", available in English translation on the internet (www.slovo.bg/old/f/en/botev/strsnger.htm).Botev lived at a moment when a half-millennium of Ottoman rule (1393-1877) was toppled, and replaced by a national government .A teacher, revolutionary, and spokesman for national Bulgarian revival, Botev's poetry focused on ordinary people's lives, in simple language (Silverman, 1983).Composed of 12 couplets, the 313-word poem has a lexical density (content-to-functional ratio) of 63.91, and a (unique-type-to-token ratio) lexical diversity of 0.53, meaning it has relatively low content.This suits proficient adult L2 learners, who generate written texts with lexical densities around 40-70 (Duran, Malvern, Richards, & Chipere, 2004).Lower densities are typically found in narratives, with adult speech averaging about 27 in oral story-telling, rising to 107 when writing (Johansson, 2009)."Stranger" permits participants to focus on meaning rather than L2 vocabulary and grammar.
"Stranger" problematises the position of wealthy, internationally-educated youth, compared to their village origins.Under the Ottoman devşirme system, intelligent provincial youth were taken to Istanbul to be educated at court.Away from their families for many years, they could return home only occasionally, if at all.The poem describes the moment when a Bulgarian youth returns to his village, and his mother cries at seeing him again.It lists many grievances the youth is ignorant of, caused by Turkish authorities, offering ironic consolations.For the loss of his childhood sweet heart to a rival, the diegetic narrator suggests many girls are attracted to wealth, implicitly separating the youth from his own feelings and his people by suggesting wealth is now his main charm.The narrator notes his father's death and his brothers' imprisonment at the hands of the Turks, dismissing these events casually by noting that the youth will soon be married, suggesting he need not notice his family's plight.The narrator scorns the youth's tears as the response of women, the poor and hungry, terms which describe the assembled villagers, countering that wealth means he will never feel their pain.The narrator then advises the young man to marry, either for love or money, and have many children, further impoverishing locals.The final couplet describes this as the life of "a fool of good repute… He never needs to ask himself whether he's human or a brute."The poem presents a trenchant critique of elite education and its impacts on youth and their families, endorsing the local as the only authentic identity.The position of 21st-century English-educated students resembles that of Botev's youth."Stranger" offers participants opportunities to see their colonial situation and hybrid identity as it is seen by those outside the system.

Instrument
Qualitative data in the form of written responses was collected from participant groups in the three locations.Teachers encouraged participants to form their own opinions, and discouraged the use of literary-critical, historical and biographical information.The prompt asked participants to give their personal opinion only.These elicit greater subjective content than academic genres such as essays (MacArthur, Graham, & Fitzgerald, 2008).Participation was optional and responses unassessed, to ensure that attitudes realised were authentic.Individual responses were anonymised, and aggregated into three subcorpora by culture.

Content Analysis
Content analysis represents ideas realised in written opinions as frequencies (Smith, 2000).Frequently-realised meanings are aggregated within themes (Roberts, 2000).The clause was the smallest, the single-clause sentence the greatest unit of meaning used to generate frequency scores (Carley, 1990).Themes were constrained by content relevant to the poem selected, rather than the full complement of meaning realised (Feldman, 1994).That is, instances of thematic articulation were recorded when they were clearly identifiable as having "a direct bearing on the question", generating lower frequencies but more reliable variability results between corpora (Bazerman & Prior,p. 17).Clauses were coded for only one theme, with transitivity to distal clauses included (Neuendorf, 2002).As it was likely that thematic comment would normally require more than a single word, n-gram dichotomous variables for word stems were not defined (Hopkins & King, 2010).Analysis of the subcorpora used these frequently-realised themes as a coding frame (Weber, 1985).
A sample of the three corpora was coded to develop a thematic frame.Higher-order themes (Dixon-Woods, Agarwal, Jones, Young, & Sutton, 2005) were identified, in order to evaluate how participants understood the parallels between Botev's youth and themselves.Six higher-order themes were frequently realised: 1) wealth disparities between L2-educated elites and locals 2) opportunity disparities between L2-educated elites and locals 3) political violence 4) evaluations of L2-educated elites 5) perceptions of L2-educated elites 6) statements of the positionality of L2-educated elites Themes 1 through 3 reflect major emphases of Botev's poem, which also impact today's youth.Themes 4 through 6 engage with the position of L2-educated youth, encompassing both the main character in the poem, and the participants who are similarly the beneficiaries of an English-language education.As theme 4, evaluations of L2-educated elites, is completely negative in the poem, any neutral/equivocal or positive evaluations of the youth show participants asserting a different stance.Theme 5, the perceptions of L2-educated elites, is conspicuously absent from Botev's poem.Comments voicing the youth's views, or those of L2-educated elites, may be taken as asserting a different self-evaluation by participants.Theme 6, statements about the position of L2-educated elite youth relative to family and villagers, indicate explicit awareness of how L2 education shapes participants' identities and positions them in various relationships.Themes were further analysed by polarity: positive, neutral or polysemic equivocal, and negative.
Nominal themes reflecting poem content (mother, father, girlfriend, brother) were deselected.Frequently-realised nominal groups reflecting higher-order commentary were observed to function like frames.Frames are interpretive schema which cluster cognitive, emotional and procedural content (Entman, 1993).People use them to organise empirical content within commonly-employed ideological perspectives (Reese, O'Gandy, & Grant, 2001).These schema are routinely transferred from known to new texts and contexts, helping people interpret new content (Shaffer, 2004).Frames are constructed within the context of culture, which is "the primary base to constitute knowledge, meaning and comprehension of the world" (Van Gorp, 2007, p. 61).Content analysis is used to analyse frames, as they organise word choices and reasoning devices such as explanation, illustration, interpretive images and metaphors, justifications, attributions of causes and consequences, and moral evaluations (Brewer, 2002).Three frames were identified: (a) personal development, (b) social cohesion, and (c) ethnicity and conflict.Frequencies were found for the higher-order themes, and the nominal-group clusters, in order to compare the three subcorpora.

Appraisal Analysis
Psycholinguistic analysis reveals subtle cues known as latent content, or subjective stance (Krippendorf, 2004).Regularities in subjective attitudes within the three subcorpora were identified using Appraisal analysis, a form of sentiment analysis based in Systemic Functional Linguistics (Halliday, 1994).When giving personal opinions, individuals choose words to represent their views (Martin, 1995).Appraisal analysis taxonomises these realisations within system networks (Martin & Rose, 2003).The Affect set realises direct emotion, and underlies Judgment and Appreciation, which realise emotion indirectly by attributing it to objects, persons and events (Martin & White, 2005).The Judgment Set includes social esteem and social sanction.Appreciation comprises our reactions to the qualities, impacts, complexity, balance and value of items beyond the self, as in Figure 1.

Figure 1. The attitude system
Appraisal analysis uses opinion-mining methods based in computational linguistics (Scherer, Schoor, & Johnstone, 2001).Machine-tagging softwares are built on compound iterations of supervised classification tasks, which build lexicogrammatical concordances for semantic subcategories.This approach handles the complexities of natural human language via statistical machine-learning techniques (Polanyi & Zaenen, 2006).Appraisal is widely-used in attitude analysis, with computer concordancing an efficient aid in detecting patterns of subjectivity in large corpora (Bednarek, 2009).These methods are "robust" (Taboada, Brooke, Tofiloski, Voll, & Stede, 2011, p. 36), with emotional language the easiest for software to identify.This study used the software Corpus Tool (CT), which integrates the Appraisal system networks (O'Donnell, 2008).Semantic categories and subcategories within these networks "are not arbitrarily posited" (Bednarek 2009, p. 150).Their validity is built on millions of authentic instances, and enhanced by their consistency with psychological models of subjectivity (Oatley, Keltner, & Jenkins, 2006).

Reliability
The three subcorpora were independently content coded and Appraisal tagged by a researcher and a research assistant.Inter-coder and inter-rater reliability was determined using Cohen's κ, including the free-margin, and percent-overall values (Lombard, Snyder-Duch, & Bracken, 2004) (Appendix B).

Data
The Bulgarian subcorpus of 31 002 words contained 1958 sentences (m=15.16words, σ=0.39), with 2793 clauses, of which 133 contained no thematic content (for example, "I am going to give my opinion"), 2402 contained content classifiable within the six themes, and 258 contained thematically-relevant but minimally-realised content units ("it goes to show how everytime we return things change"), or content on other topics ("It reminds me of what my grandmother says about living in the village").The Hong Kong corpus of 19 341 words contained 1853 sentences (m=10.86words, σ=0.62), and 2036 clauses, 572 without thematic content, 1134 with relevant thematic content, and 330 with relevant but minimally-realised or irrelevant content.The Lebanese corpus of 29 923 words contained 2738 sentences (m=14.65,σ=0.74), with 3205 clauses, of which 369 contained no thematic content, 2189 contained content classifiable within the five themes, and 647 contained thematically-relevant but minimally-realised content or content on other topics.

Content Analysis
Thematic frequencies were obtained for the three subcorpora (AppendixA).Inter-coder reliability was calculated for the three subcorpora.Two coders independently obtained content unit frequencies for each corpus.Inter-coder reliability values were not attributable to chance (Appendix B).
The three subcorpora are represented in Figure 2. The Lebanese and Hong Kong subcorpora had more negative (Lebanese 46.23%, Hong Kong 54.32%) than positive (Lebanese 45.50%, Hong Kong 36.86%), with little neutral/equivocal (Lebanese 8.27%, Hong Kong 8.82%) content.The Bulgarian subcorpus had mainly positive (58.74%), moderate negative (31.14%) and little (10.12%) neutral/equivocal content.In the Lebanese and Hong Kong subcorpora, wealth and opportunity disparities were frequently-realised.The Lebanese subcorpus contained 21.37% such content, mainly negative (49.79%) ("It says that having more money than everyone else doesn't make you happy, although he could probably bribe the Turks to get his brothers back").The Hong Kong subcorpus contained 17.72% such content, also negative (59.20%) ("Of course the village person cannot make the good salary like the city, they are trapped in low situation").Only 4.75% of Bulgarian content realised wealth disparities.Opportunity disparities were also frequently-realised in the Lebanese and Hong Kong subcorpora.These comprised 28.41% of Lebanese content, 75.40% negative ("a person from a small village familiar with the powerful can try to receive a prosperous position from them, but only a few will succeed").They comprised 33.16% of Hong Kong content, 57.18% negative ("Maybe in the past you are from one of those wealthy families you can have the higher career, but nowaday this is not the option").Only 7.87% of realisations in the Bulgarian subcorpus realized this theme.The Lebanese and Hong Kong subcorpora infrequently realized evaluations of and the positionality of L2-educated elites, about which Lebanon was positive and Hong Kong equivocal.
Evaluations of L2-educated elites (35.85% of total content, 64.81% positive), the perceptions of L2-educated elites (19.03%, 54.49% positive), and their positionality (27.31%, 59.15% positive) were frequently-realised in the Bulgarian subcorpus: 1.This poem shows a stage in a particular person's life in a very different world.2. The time and the circumstances made him do things that are hard and very painful for a person.3. I feel very sad for the things that he must do so he could feel and become a man.4. It is really hard to leave everything behind and just start over.5.People believe that it's that simple.6.But it is not.7.You cannot just forget your past.8.And apparently think how to reproduce.9. Sadly, this is a major problem in our society now, as well.10.I know what is to feel homesick, to overcome death of someone you loved and I am a potential stranger myself as well as my whole family.11.I plan on finding my place in the world and never come back home.12.So this poem gets me thinking of the things that I will sacrifice and let go for a better life.
Most Bulgarian content realising these themes historicised the Ottoman context ("The poem shows what it was like for Bulgarians to become educated long ago, which was a terrible sacrifice"), contesting the poem's suggestion that elite education means deracination ("In this time, people leave their village every day.But I think you always belong, when you go home").They rejected Botev's dichotomising vision of identity, seeing the L2-educated as eternally connected to a supportive village ("For me home is the place where your loved ones are the people who only want you to be happy no matter what, who support you whatever path you may take.")They acknowledged difficulties, but defended L2-educated youth ("It is hard for his mother that he is absent, but he is a good son because he came back to help them," "When students grow to adults they make their decisions, the same as all the world".) Hong Kong participants saw L2-educated youth as the oppressed (29.82%, 53.55% negative, "In real life, the youth face unfair situation, because even he graduate the highest class, he cannot earn enough for a good level of life"), identifying wealthy Chinese and foreigners as the elite ("It is always like this in Hong Kong when it is dominated by the international circumstances and our efforts are made in vain", "[t]hose powerful billionaires from mainland control the economic").Evaluations of L2-educated elites were infrequent in the Lebanese subcorpus (11.15%, 51.64% positive), mainly seen as wealthy ("they have more money to put on the table"), but suffering separation ("Turkish greed for the children caused them pain to come home.It is the same now").Realisations of the perceptions of the L2-educated were infrequent (8.09%, 57.63% positive), and assumed their alignment with village and family ("he had the courage to stand with his own culture and tradition beliefs", "in contrast to the outside world, where the boy and the Turks might not have to be enemies").Realisations of the perceptions of L2-educated youth were infrequent in the Hong Kong subcorpus (5.56%, 53.97% negative).Realisations of positionality were infrequent in both the Lebanese (4.80%, 56.19% positive) and the Hong Kong (1.68%, 57.89% negative) subcorpora.
Political violence was frequently-realised in the Lebanese subcorpus (26.18%, 67.54% positive), reflecting participants' views of Ottoman rule, or the civil war ("He is a victim of the Turkish empire in the last century stealing his home, and no-one actually wins anything in the end", "This poem has many meanings for the life we live today, especially the reasons why we still have religious wars in Lebanon".)Most thought that hybrid identities led to conflict ("he is in a war even though the cultures do not encourage belligerent behaviors.What he faces now is to battle violently just as both their ancestors fight for years").Most assumed that local identity superseded other ties ("He finds that his father is killed, and the Turks put his brothers into prison.Most of the people by that time have only one thing in their mind, which is to gain honor from the others and the way to achieve that is often violent", "the boy is a combination of both worlds.Regardless of his combination, one dominates the other one in this unusual time, leading him to prove himself").
Political violence was moderately frequently-realised in the Hong Kong corpus (12.80%, 56.93% positive), mainly in defense of family ("Even he must fight to repair the disorder of the family").Most realisations reflected impartially on violence as a response ("If they kill his father, it is a tragic.Although a life for a life is fair", "[t]he society will be much simpler and more violent if he just kill our enemy for the reason that to be killed is easier than using the rest of life to atone"), seeing it as understandable ("In my opinion, taking revenge is a rational behaviour but it should not violate the principle of [not] hurting those innocent").Political violence was infrequently realised by Bulgarians, who saw it as reflecting Botev's biography ("He was one of the few people daring to rise up and fight for what was right in those dark times of oppression").

Clustered Content Frequencies
Frequencies were obtained for three content clusters which functioned like cultural frames for interpreting the poem.Distinct clusters characterized the three subcorpora.This was most pronounced in the Lebanese and least in the Bulgarian subcorpora, as in Figure 3.The personal development (PD) cluster comprised clauses realising emotional and intellectual growth, changes in personal views, assertions of values found in the poem, and summative statements of meaning gleaned from reading the poem.The Bulgarian subcorpus had most clauses of this kind ("he manages to inspire a whole nation", "at the end, this poem makes me think about life", "I feel like what the author is trying to tell us is that regardless of the situation or problems we are faced with, we need to do our best to try and overcome them and create a brighter future for ourselves").The social cohesion (SC) cluster comprised clauses realising mutual obligations of the individual and the group, including tasks to be undertaken by each, reasonable and unreasonable expectations of authorities and citizens, and the assessment of legitimate versus unreasonable expectations for each.The Hong Kong subcorpus had most clauses of this kind ("everyone should be born with full equality regardless", "[u]nfortunately, his situation are misconstrued by the village altering the reasonable responsibility for an individual", "it is not fair to expect him to solve the many difficult problems easily").The ethnicity and conflict (EC) cluster comprised clauses realising ethnicity, power struggles, claimed obligations associated with identity, and inter-ethnic violence.The Lebanese subcorpus had most clauses of this kind ("this poem delivers a message to readers, that colonial powers are too arrogant, but they naively think that Europeans are more sophisticated", "he fights for the right course, everyone fights for their home") (see Appendix C).
The Lebanese subcorpus was most positive.The Bulgarian and Lebanese subcorpora contained higher levels of affect, the Hong Kong subcorpus highest level of judgment.Of 24 subcategories, realisations in seven subcategories comprised 68.00-75.45% of all realisations, as in Table 1.
The Bulgarian subcorpus contained many positive realisations of impact and quality, most appreciating the poem as literature.These subcategories rework affect as attributes of objects and events outside the self (Martin & White, 2005).For example, "genius", "cool", "great", and "has inspired and still inspires" rework feelings of interest and pleasure as qualities and impacts of the poem).Negative realisations of misery realised emotions felt on behalf of the main character, and also by Bulgarians facing their own difficulties ("He is the most hopeless and distressed person ever", "It made me sad, to remember my family's loss").Judgments of capacity realise the modality of possibility (can, able, capable, ability, flair, know how, skill, expert, proficient, competent, aptitude, knack, and so on) (Panther & Thornburg, 1999).Most realised participant abilities ("what this poem means to me, if I had to put it in one word it would probably be hope.Because in the end the best thing we can do is be positive and hope for a better future".)Appreciations of valuation assess persons and performances for their social worth (Eggins, 1994)."[T]o laugh at the boy's pain is original"and "Botev creates strong and deep feelings" assess Botev's artistic achievement.Bulgarians frequently positively evaluated "Stranger" as worth reading: "The poem made me question if all that my parents, society and me myself deem a decent life path is truly worth".Negative displeasure and disquiet mainly realised responses to the youth's dilemmas ("Botev's provocative assault on many aspects of the way of life I, and perhaps most of us, strive for is bothering", "Stranger…makes me feel not only sick at heart but also angry and disgusted").
Judgements of normality, propriety (negative and positive) and capacity comprised four of seven frequently-realised subcategories in the Hong Kong subcorpus.Judgments voice the support or sanction we attribute to people, their behaviours, and social events (Martin & Rose, 2003).Judgments of normality assess how well something fits with our expectations, through lexis such as different, special, new, unusual and so on.Most judgments were negative ("this is a strange situation", "to face with so many difficulties is out of my experience").Judgments of propriety use lexis associated with law, duty, rules, regulations, rewards and penalties.Positive realisations assessed the youth's family obligations ("It is his duty to arrange the release his brothers," "to fix the family problems should be within an acceptable manner").Negative judgments assessed the youth's past and the village situation ("It is not right to be gone for so long", "The village are unjust and bring a lot of pressure to him," "It is unfair to expect him to marry quickly").Most judgments of capacity focused negatively on handling the problems ("Maybe he cannot rescue his brothers from the authority," "I think it is difficult for the son to be able to fix the problems").Appreciations of complexity realise perceptions of order, through lexis such as complicated, intricate, elaborate, convoluted and so on.Most positive judgments of complexity reflected the grammar of realising an opinion, not a positive view of the youth's situation ("his mother makes his situation more complicated", "he must look in details and try to understand how to improve the situation").Negative appreciations of quality focused on the youth's problems ("It is the gloomy story of a man with problems in the family", "the news brings despondency on him").
The Lebanese corpus was most positive, with confidence voiced on behalf of the youth ("He should not be pessimistic, he should believe he can do it", "they have very close relations, and they praise the work that their fathers have done for them, therefore they can be confident, even facing dying").Most positive feelings of desire articulated hopes on behalf of the youth ("I hope there is another poem, where he has married a person he loves", "we all aspire to lifelong effort if necessary, and perform the behavior on the person we want to revenge on").Most negative Judgments of misery reflected the youth's difficulties, many rhetorically advised how to handle them ("He gets news which is very depressing", "We need to sacrifice our happiness and accept sorrow in the rest of our life until we successfully revenge on that person").Most judgments of veracity were grammatically positive, but assessed the negative qualities of the youth's situation as accurate ("This is exactly what happens, it's depressing", "The poem is true to the old and lasting perception of Turks", "What truly led to the defeat of the Turks was unanimous support from the local people to fight them").Most positive appreciations of impact focused on the literary qualities of poem ("This created a vivid picture for me.I can almost feel the hotness of the sun," "In his life, now, the power struggle is intense").Most positive judgments of normality focused on elements youth's biography, though these often utilised grammatically positive constructions to discuss negative realities ("To have your brothers in jail was pretty common back then", "For one group to dominate the other was usual so they can highlight their heroic deeds").Those negative realities were realised in negative judgments of normality ("He hears that they have shot his father.From now on, he will always have hostile feelings towards the Turks", "The truth is, there is always a retribution for the person who does something bad to others").

Discussion
Postcolonial youth in these three cultural contexts read the colonial situation differently.Lebanese and Hong Kong content categories were often negative.As Confucian rhetorical styles avoid direct criticism (Panetta, 2001), the Hong Kong results are notably so.These two groups understood colonialism negatively, in terms of wealth and opportunity disparities, probably reflecting personal expectations.Economic data might suggest Hong Kong youth have positive futures, yet while Hong Kong is the world's 33rd largest economy, locals feel their prospects were foreclosed by the 1997 British Handover, and are now eclipsed by the rise China (Chan & So, 2016).Economic data aligns with Lebanese youth expectations.The world's 82nd largest economy, Lebanon currently faces inflation, unemployment and instability due to the Syrian civil war (Shafik, 2016).For these youth, then, their L2 education does not persuade them of their status as elites with access to the global economy.Further, violence was frequently-realised in these two subcorpora, in the Lebanese case probably reflecting experiences of the 2006 Israeli invasion (Abdelhady, 2008) and in the Hong Kong case the Umbrella Revolution (Fung, 2014).That is, these L2-educated youth perceive themselves as villagers, not overlords.However, economic prospects did not seem to account for the positive polarity or content frequencies in the Bulgarian subcorpus.As the 80th largest economy (World Bank, 2015), with modest growth prospects, and impacted by the 1991-2001 Yugoslav wars, it resembles Lebanon.Yet Bulgarians explored the youth's situation empathetically, voicing his and their own perspectives, and reflecting on their complex identity and positionality.
That cultural dimensions shaped how Lebanese and Hong Kong participants read the poem's colonial situation may be seen in content clusters also.Members of a collectivist culture and obligated to work for family prosperity (Yao, 2000), Hong Kong participants understood the colonial situation in terms of social roles and obligations.Members of a culture where sectarian identities shape all elements of daily life, and wasta ‫ْطة(‬ ‫واس‬ ), or personal connections, define all relationship potentials, Lebanese participants understood the colonial situation in terms of ethnic conflict (Cammet, 2013).That is, again, many contemporary L2-educated youth read themselves not as elites, and seem to employ their local identity when assessing their global prospects.Again, this did not account for Bulgarian participants' more positive reading of the colonial situation in terms of personal growth and development, or in terms of their global ambitions.
Emotion is a crucial area of difference between Lebanese and Hong Kong participants, compared to Bulgarians, and this was evident in how the participant groups responded to the youth, and to the critique of him offered by the diegetic narrator.Participants in all three contexts realised misery on behalf of the youth.Bulgarians also realised misery on behalf of themselves, connecting his pain to their own.Further, they realised displeasure and disquiet, with these three negative emotions comprising more than a quarter of all attitudes realised by Bulgarians, where positive impact, quality and valuation mainly appreciated Botev as poet, and the poem as literary achievement.Hong Kong and Lebanese affect applied more frequently to the youth than to themselves, with little applying to the poem, poet, or diegetic narrator.Lebanese judgments of veracity and normality assessed the youth's troubles as realistic and normal, and positive and negative judgments of normality both voiced negative realities, without also realising emotion on the youth's behalf.About a third of all attitudes realised by Hong Kong participants were negative judgments, censuring the normality, complexity and propriety of the youth's situation and the villagers' actions, both excusing the youth on grounds that the difficulties were too great, and prescribing how he should address his situation.But misery, empathy for or solidarity with the youth were infrequently realised, and indirectly as appreciations of quality.That is, where Lebanese and Hong Kong participants understood the youth's troubled situation, they responded to it indirectly.Bulgarians felt it, and responded emotionally.This suggests that Bulgarians identified more with the Bulgarian youth, hardly surprising as "Stranger" depicts their own history.
Support for the impact of shared national and historic identity on the realisation of emotion came from congruent versus lexicalised constructions in the three subcorpora.Congruent realisations directly connect the writer to an emotion, using I+emotion constructions ("I felt my heart despair when I read this poem"), which "bear a natural relation to the meanings they have evolved to express" (Halliday, 1985, p. xviii).These were more frequent in the Bulgarian (174=6.23%)than the Lebanese (52=0.26%)or Hong Kong (33=1.02%)subcorpora.Nominalised constructions reformulate the experience indirectly.For example, clauses employing shared subjectivity ("we" for "I") inhabit a wider community of discourse, assuming its identity or authority in support of the opinion voiced (Graff & Birkenstein, 2010).The Lebanese subcorpus employed this construction more frequently (431=13.46%)than the Bulgarian (197=7.05%)or Hong Kong (40=1.96%)subcorpora.Typification reframes subjective experience as an instance of a category or class (Brinton, 2008) ("This kind of burden is terrible to suffer").These also were more frequent in the Lebanese (285=8.89%)than the Bulgarian (60=2.15%)or Hong Kong (48=2.36%)subcorpora.
Botev's youth is a proxy for participants in the three groups.Yet realisations of the youth's status and positionality were rare, except in the Bulgarian subcorpus.This result suggests that the shared identity between the literary character and Bulgarian participants enhanced their critical thinking and perceptions.Greater emotional arousal usually accompanies L1 than L2 reading (LeDoux, 2002).Bilinguals experience reduced emotional arousal when processing in their L2 (Doughty & Long, 2002).This is often felt to make second-language acquisition easier (Fussel, 2002).However, this study suggests that authentic literature in translation constitutes a kind of halfway house, and that greater emotional arousal due to identification with literary figures enables complex understandings such as one's own status and positionality.

Conclusion
This study has limitations.First, using tertiary participants offers greater accuracy in depicting how those entering the global knowledge economy in coming decades see themselves.However, it remains less representative of national populations overall.Second, because literary texts evoke subjective content, studies using larger samples might produce different attitudinal profiles, though it is likely that similar results would occur.Third, it is possible that the negative content of the poem led Lebanese and Hong Kong youth to employ their local identity in assessing their global prospects.Further study using a literary vehicle with positive content would clarify this.
How do cultural values and subjective attitudes shape how postcolonial youth in three different contexts read the colonial situation depicted in this poem?This study has reached three main conclusions.First, English-educated youth no longer seem to view English-language education as conferring elite status, or functioning as an effective ticket-of-entry into the global economy.One research question in this study had been to explore contemporary youth's views of their elite second-language education, and the opportunities it offered them.While English proficiency does in fact privilege them as better able to participate in the global economy, their self-understanding is not as elites.Instead, they seem to be most aware of the ways in which local economies and global realities shape their futures.Probably correctly, they view these as uncertain.These views of self and context condition how they read L2 literature.
Second, faced with significant difficulties, many bilingual youth do not construct themselves as elites, but rely on their local identity.One research objective for this study had been to explore how contemporary youth understand their hybrid identites.This result reflects inevitable realities, but in the Lebanese and Hong Kong subcorpora, the lack of realisations applying to the poet, poem and positionality of the youth show their need to become more aware of the multiple lenses through which others, including future colleagues and customers, may view them.Being aware of one's own assumptions about self, in local and global contexts, are essential to effective functioning in both contexts.While the global economy remains uncertain, hybrid L2 identities are now found worldwide.This suggests that the optics of status and positionality need to be taught explicitly to tertiary youth.Further, while a retreat to the comfort of home territory is understandable in response to difficulties, for bilingual youth to construct themselves as underdog may not serve them well when they enter employment.This result has implications for tertiary administrative units helping prepare youth for the job market.
Third, in contradistinction to those trends in how bilinguals understand their global status and identity, the Bulgarian data suggests that greater emotional identification with literary characters, for example in authentic national and historical texts, may induce a more powerful emotional response and support greater critical understanding of status and positionality.One research objective in this study had been to explore how contemporary youth in post-colonial contexts understand the colonial situation.This study suggests that authentic texts may support some kinds of perceptions or critical thinking tasks.Further study is needed to identify how authentic texts in English translation support learning among advanced bilinguals.This has implications for curriculum, as it seems that English-language translations of authentic content elicit greater emotion from readers.Bulgarians realised more emotion, across a greater range of affect subcategories, than Lebanese or Hong Kong participants.

Figure 2 .
Figure 2. Thematic frequencies by percent, for the three participant groups

Figure 3 .
Figure 3. Nominal-group thematic clusters for three subcorpora by percent