Reading in a Foreign Language April 2024, Volume 36, No. 1 ISSN 1539-0578 pp.1–25 Japanese University EFL Learners’ Responses to Lexically Easy Short English Poems Takayuki Nishihara Hiroshima University Japan Abstract This study explored the diverse responses of Japanese university-level learners of English as a foreign language (EFL) to lexically easy short poems. These participants had attained a high level of English proficiency and were able to grasp the literal meaning of the poems. The investigation employed confidence level and reading time as supplementary data points. The study yielded the following findings: (a) the majority of responses were heavily dependent on literal content; (b) learners typically did not substantiate their readings by referencing the poems, but when interpretations deviated from the literal content, they provided more supportive evidence; (c) the learners’ confidence level in creative interpretations was relatively low; (d) considerable variations were observed in meaning construction; and (e) learners spent the same amount of time to produce various interpretations with and without supportive evidence. Keywords: poetry reading, literary interpretation, English language learners and literature, learners’ responses to poetry, learner confidence, reading time, reading literature, teaching literature Poems are among the most important reading materials in English language teaching. Different educational contexts employ varied teaching practices (e.g., Ariyanto, 2021; Billows, 1961; Hanauer, 1997; Lin, 2006; Paran & Robinson, 2015; Rosenkjar, 2006; Scott & Huntington, 2007; Tomlinson, 1986; Widdowson, 1992). Reading poetry involves the construction of both referential and interpretive meanings. Although the construction of the latter is thought to be particularly important, individual learners’ responses at this level of meaning have rarely been empirically investigated. Poetry instruction must also be informed by individual learners’ responses to poetic texts. The current study explores the diverse interpretive responses to lexically easy short poems, which are now widely used in foreign language teaching (Duncan & Paran, 2018; Paran & Robinson, 2015; Rosenkjar, 2006; Tomlinson, 1986). This study focuses on Japanese university-level learners of English as a foreign language (EFL) and utilizes confidence level and reading time as supplementary data points. http://nflrc.hawaii.edu/rfl Nishihara: Japanese university EFL learners’ responses to lexically easy short English poems 2 Study Background Foreign language teaching has included poetry for a long time (Widdowson, 1982). Scholars have pointed out various merits of using poetry in foreign language education, including understanding different cultures, enhancing motivation to learn foreign languages, and facilitating foreign language acquisition (Badran, 2007; Bland, 2015a, 2015b; Hanauer, 2001a, 2010; Lazar, 2016; Lin, 2006; Scott & Huntington, 2007; Tomlinson, 1986; Zyngier & Fialho, 2010). Further, recent research has agentically engaged learners in the meaning-making of poems, intending to foster awareness of interculturality and diversity; promote intellectual, emotional, ethical, and social development; enhance learner autonomy; cultivate critical thinking; expose individuals to creativity; and broaden their understanding of humanity and the world (e.g., Bland, 2015b; Civelekoğlu & Saka, 2017; Delanoy, 2018).1 Although poems, which frequently include linguistically complex structures, deviant grammatical constructions, and low-frequency words (Hanauer, 1997; Paran & Robinson, 2015), have often been identified as challenging materials for learners, researchers have proposed to use linguistically accessible poems (e.g., lexically easy short poems) in teaching practices. This genre of poetry in English is used with the aim of reducing the linguistic burden on learners, allowing them to enjoy the benefits of foreign language learning through English poetry (Duncan & Paran, 2018; Paran & Robinson, 2015; Rosenkjar, 2006; Tomlinson, 1986). However, English language teaching research with literary texts lacks empirical evidence of learners’ responses and engagement with texts (e.g., Duncan & Paran, 2018; Paran, 2008). Literary reading includes solitary and social reading; the former refers to “the solitary interaction between the reader and the original text” and the latter is “the social interaction between readers that reconstitute that text” (Peplow et al., 2016: 33).2 Civelekoğlu and Saka (2017), Hanauer (2001), Scott and Huntington (2007), and Zyzyk and Polio (2008) have made significant contributions in addressing the problem of reporting the responses and behaviors during pair and group social interactions with texts in the case of social reading. In a study by Hanauer (2001), pairs of advanced ESL learners whose first language was Hebrew engaged in the collaborative reading and discussion of a poem to discern its meaning. This study revealed 10 types of learners’ responses to the poem and their patterns of progression. Moreover, Civelekoğlu and Saka (2017) showed that an 8-week poetry reading instruction program with interactive activities improved the intercultural awareness of Turkish EFL university learners whose English language proficiency was estimated to be at the B1 level according to the Common European Framework of Reference for Languages (CEFR) criteria. Solitary Reading However, empirical research on solitary reading is still lacking. In classroom settings, learners typically engage in individual solitary reading before progressing to sharing and discussing the text’s meaning either in pairs or as a group, that is, social reading (Peplow et al., 2016, p. 30). Moreover, if learners are to enjoy poetry outside the classroom, then they will inevitably engage Reading in a Foreign Language 36(1) Nishihara: Japanese university EFL learners’ responses to lexically easy short English poems 3 with the texts individually. Regarding the study of teaching English through literature, Bernhardt (2023) wrote, “interestingly, more studies have appeared that purport to assist second-language readers in their literary reading than purport[ing] to help the field understand what the process of literary reading is for learners” (p. 109). In contrast to research providing guidelines and activities for integrating poetry into the classroom, there is a scarcity of studies focusing on individual learners’ responses to poetry. The paucity of empirical studies on solitary literary reading stands in sharp contrast to studies on nonliterary reading (e.g., Bernhardt, 2023, McCarthy & Goldman, 2019). According to Peskin and Hanauer (2023), the idea that implicit meaning is hidden beneath the surface literal meaning of a text in the reading of literature, especially poetry,3 is widespread among literary critics and general readers alike. For instance, reading between the lines, which is common in poetry reading, is a process of implicit meaning-making beyond the referential meaning explicitly indicated in the text and is based on this assumption. Nishihara’s (2022) work is a rare study empirically investigating the characteristics of solitary reading responses by having upper-intermediate EFL learners comprehend and interpret lexically easy short poems based on this assumption. Some of the findings presented in this study regarding learners include that they read texts of the same lexical level at different speeds and with different confidence levels. Learners spend more time on interpretation than on literal comprehension and tend to experience a reduction in reading speed when encountering texts they find difficult or when they feel less confident. Learners encounter various problems in constructing interpretations, which they are rarely able to solve, and they rarely have high confidence in the interpretations they make. Further, learners’ confidence decreases during interpretation compared with literal comprehension. However, Nishihara (2022) did not shed light on individual learners’ interpretations of lexically easy short poems. As mentioned earlier, English language teaching as a field has attempted to provide learners with linguistically accessible poems to allow them to enjoy the advantages of poetry reading while reducing their linguistic difficulties. To enhance learners’ proficiency in reading such texts, it is necessary to investigate each learner’s interpretive responses, considering the confidence level and effort involved. Such an exploration can also be expected to bring to light some of the problems that individual learners have in the process of interpreting poetry. This knowledge can help teachers to be proactive while supporting learners in their teaching practice. The Present Study In the present study, Nishihara’s (2022) data are reanalyzed; further, learners’ reactions during interpretation are categorized to delineate their variations. Learners’ responses are also characterized in terms of their confidence level and reading effort. Therefore, this study aims to answer the following research questions (RQs) to explore learners’ interpretive responses to lexically easy short poems: Reading in a Foreign Language 36(1) Nishihara: Japanese university EFL learners’ responses to lexically easy short English poems 4 1. What are the responses of upper-intermediate EFL learners when interpreting lexically easy short English poems individually? 2. How are learners’ interpretive responses characterized in terms of their confidence level and efforts to construct meanings? To answer these research questions, this study uses three types of data: learners’ responses to each poem, their confidence level in responding, and reading time. While RQ1 focuses on the first data type only, RQ2 covers all three data types. Presently, there is no framework to categorize the interpretive responses of foreign language learners when they read English poetry. Therefore, this study used the framework of native speakers’ interpretive response categories (Peskin & Wells-Jopling, 2012; Svensson, 1987). Svensson (1987) recruited 72 learners aged 11, 14, and 18 years to read four poems by famous contemporary Swiss poets in first language (L1) and analyzed their responses. The responses were classified into six types ranging from verbatim or paraphrased texts to symbolic responses with analogical inferences: literal descriptive, literal interpretive, mixed literal–thematic, thematic, mixed literal–symbolic, and symbolic (Svensson, 1987, pp. 486–487).4 Peskin and Wells-Jopling (2012) made minor modifications to this categorization framework and classified the responses of 6th-, 9th-, and 12th-graders to poetry into five types, combining the mixed literal–thematic and thematic response types. The present study classifies learners’ interpretive responses deductively and inductively with reference to these models. Additionally, learners’ interpretive responses included responses with evidence to support their interpretations (response with evidence [RWE]) and responses without such evidence (response without evidence [RWOE]). Therefore, differences in confidence level and effort to construct meanings between these two types of responses were also examined. This study also investigates learners’ responses to lexically easy short poems when given a single prompt in their L1— that is, “What do you think the poet is trying to say through that poem?”— while viewing the poem as a whole. Interpretive activities with literary texts are not limited to speculating about the author’s intentions. The text’s field of meaning is richer than the author’s communicative intention, and readers should be encouraged to consider the literary work’s themes, symbolism, and lessons (Burkett & Goldman, 2016; McCarthy & Goldman, 2015). Recently, ways have been sought to move readers away from reading to ascertain the author’s intentions (e.g., Bland, 2018). Traditionally, however, many readers have tended to read literary texts in search of the author’s intentions (the author being a construct created by readers in their mind from the information in the text rather than a flesh-and-blood author; Claassen, 2012; Guy et al., 2018; Stockwell, 2016; Zyngier, 1999). Paran and Wallace (2016) considered this type of reading to be the default position for most readers. The prompts or tasks that require readers to consider the author’s intention or message are frequently used to investigate their L1 literary reading in natural situations (in the absence of educational intervention; Harker, 1994; Levine, 2019; Peskin & Wells-Jopling, 2012; Svensson, 1987). Besides, for the learners in this research Reading in a Foreign Language 36(1) Nishihara: Japanese university EFL learners’ responses to lexically easy short English poems 5 context, inferring the poet’s intention is a familiar literary reading activity for an additional layer of meaning of a poem because of the influence of traditional L1 literature education. Therefore, in this study, the participants were asked to identify the authorial intent as an interpretive activity that they are familiar with (as reported below, the participants were instructed to speculate freely about the author’s intentions). This study’s findings are expected to be used to construct ways to free learners with a traditional view of literature from their default position and transform them into learners who participate actively, affectively, and critically in the text, relating the text to their own experiences, knowledge, the world, and others (e.g., Bland, 2018). Method Participants The study recruited 24 Japanese EFL learners (17 undergraduate and 7 graduate students; male: 12; female: 12; age: 21–26 years) participating in an English teacher training program at a national university in Japan. English language proficiency was estimated to be approximately B1– B2 using the CEFR (their scores on the Test of English for International Communication were 545–925 [mean: 787.61; SD: 106.71; perfect score: 990]). Their English proficiency was developed through the educational curriculum in Japan. They habitually read Japanese for approximately 60 minutes per day and English for approximately 30 minutes per day (for learning and recreational purposes). However, they were less fond of reading poetry in Japanese or English, and the majority of the participants read less than one poem annually. Thus, their knowledge of Japanese and English poets was limited, and they lacked prior knowledge of the authors of the research materials used in this study. No participant received any systematic instruction in English poetry reading strategies before the study. The institutional review board of the university to which the author and participants belong approved the study. The researcher explained the objective and outline of the survey in his class and invited potential participants. Participants voluntarily submitted written informed consent and were compensated. Materials Short poems were selected from poetry anthologies (Brooks & Warren, 1978; DiYanni, 2000; Hoover, 1994; Kacian et al., 2013; Kennedy & Gioia, 2002; Kennedy & Kennedy, 1999; Maley & Moulding, 1985; McRae & Vethamani, 1999; Ramazani et al., 2003; Riddell, 1972). Poems with low-frequency words or grammatically complex constructions, themes unfamiliar to the participants (religious poems and works based on mythology), and sexualized readings were excluded. A total of 19 poems were selected as potential research materials. To incorporate various poems into the shortlist of research materials, they were classified on the basis of the following stylistic features: (A) lines of verse were arranged in a pictorial form Reading in a Foreign Language 36(1) Nishihara: Japanese university EFL learners’ responses to lexically easy short English poems 6 (concrete poems); (B) each line corresponds to the ending of sentences, clauses, or phrases; (C) everyday sentences were presented in a poetic layout; (D) only noun or adverbial phrases were included; and (E) syntactic deviations were considered. Two poems each were selected from Types A to D and one from Type E (only one was classified as Type E). Consequently, a total of nine poems were used as part of this study. Although the selected texts differed in the number of lines, words, and syllables, their lexical difficulty was nearly the same (see Table 1). Based on SVL12000,5 their average word frequency levels ranged from 1.0 to 1.8 without statistically significant differences (Fisher’s exact test; p = .588 with Cramer’s V = .257). Based on the participants’ English proficiency, all works were linguistically (particularly, lexically) accessible. The participants were not made aware of the authors’ names and titles. Table 1. Linguistic traits of shortlisted poems Text Number of Number of Number of Average word lines words syllables frequency level A1 Help by Alan Riddell 1 6 6 1 A2 The Wall by Abdul Ghafar Ibrahim 2 5 6 1.2 B1 October by Richard Aldington 4 22 24 1.1 B2 Island by Langston Hughes 8 30 36 1.3 C1 This is Just to Say by William Carlos 12 28 38 1.2 Williams C2 The Bottom of My Shoes by Jack Kerouac 3 12 13 1.3 D1 City Street by Cor van den Heuvel 3 9 13 1.2 D2 Cold Moon by John Stevenson 3 8 13 1.8 E1 An Unseen Bird Sings by Peter Yovu 3 12 13 1.8 Procedure Each learner participated in a laboratory experiment. All interactions between the researcher and participants were conducted in L1, Japanese. The participants sat in front of a PC screen, where they were briefed on the task. PPT2TTL (WAWON DIGITECH) was installed on the PC to control the presentation of the research materials and to record reading time in milliseconds. The participants engaged in all tasks on the screen. Prior to the experiment, the participants read a list of words used in the research materials and their definitions (to eliminate the influence of new words on task performance). Subsequently, the participants were given practice texts on the screen for them to become familiar with the task types. After this practice session, the participants opened the assignment files and worked on the reading tasks for the nine poems. The poems were presented in a randomized order. The participants were informed that an interview would be conducted on their interpretations of the poems. If they were unable to arrive at an interpretation during the task, then they were allowed to respond with “I don’t know” during the interview. Reading in a Foreign Language 36(1) Nishihara: Japanese university EFL learners’ responses to lexically easy short English poems 7 The participants worked on two tasks for each poem, namely, literal comprehension and interpretation. The order of the tasks is in line with conventional assertions that literal comprehension precedes interpretation in poetry reading (Goldman et al., 2015; Levine, 2019; Rosenkjar, 2006). For the first task, the poems were presented line by line and the participants were instructed to understand the literal meaning of each line at their own pace. They pressed the space key to read the next line (while the previous line remained on the screen). This process was repeated until the participants reached the last line. They then performed the interpretation task (at this point, they could not return to the literal comprehension task). Notably, the specific shape of the concrete poems (Type A poems) was not reproduced, but each line was presented as a simple verse line in the literal comprehension task. In the second task, the participants were given a prompt in their native language—“What do you think the poet is trying to say through the poem?”—while reading the entire poem at their own pace. The participants were told beforehand that there were no correct answers in this task (Earthman, 1992; Sigvardsson, 2020) considering the impossibility of knowing the poet’s exact intentions at the time of writing the poem. They were instructed to respond freely to the poems for this task. Afterward, they pressed the space key if satisfied with their response or if they wanted to pass. The next English poem then appeared on the screen, which once again initiated the literal comprehension task until they reached the last poem. Interviews were then conducted on the nine poems according to their screen task order. The participants were asked about (1) the intended meaning of the poems (their constructed interpretations), (2) their confidence level for their interpretation (percentages), and (3) concerns about the poem during the interpretation task. At the end of the interviews, they were asked (4) whether any of the poems were difficult to interpret (if so, in what order) and (5) whether they had read any of the poems before the study.6 Types of Data Three types of data were used for analysis. The first pertains to the participants’ interpretations of each poem (their inferred intended meanings). The second refers to the students’ confidence level in their interpretation. Lastly, reading time was analyzed. The study used PPT2TTL to automatically record the start of the interpretation task until the space key was pressed to move to the next task. Reading time is commonly used as a measure of the degree of processing effort (Castiglione, 2017) and of difficulty experienced during reading (Sanford & Emmott, 2012). The same is true for the current study. The word counts varied across poems; thus, reading time per syllable was used for analysis. Classification of Responses This procedure followed three steps, namely, 1. preparation of the dataset, 2. creation of categories, and 3. evaluation of the reliability of the categories. Reading in a Foreign Language 36(1) Nishihara: Japanese university EFL learners’ responses to lexically easy short English poems 8 All the responses to interview question 1 were transcribed during data preparation (step #1). However, the interpretation of poems that participants had read before the study were omitted. Therefore, the dataset consisted only of responses to poems read for the first time during the study. Responses such as “I didn’t understand the poet’s intent” were also excluded. Consequently, a total of 174 responses were included for analysis. The study extracted 174 responses and carefully classified them under reading and deductive and inductive understanding using the framework of response types of L1 literary readers (Peskin & Wells-Jopling, 2012; Svensson, 1987). The results yielded four categories that exhibit a high degree of commonality with the L1 classification (step #2). The study assigned names and definitions to each category by slightly modifying those presented in the L1 study based on the responses. Finally, the study evaluated the reliability of the four categories (step #3). Two raters (the author and one assistant rater external to the study) classified the responses using the four categories. They shared the study purpose, the nine English poems with their Japanese translations, and the names of the four categories along with their descriptions and then classified 22 responses (12.6% of the dataset) together. The raters then individually classified the remaining 152 responses (this was a reclassification of data for the author because the author had previously classified the data once during the creation of categories). After this process, the two raters reviewed their results. The degree of agreement with data classification reached an acceptable level (κ = .95). Any disagreements were discussed until a consensus was reached. In summary, the 174 responses were classified under four categories. Additionally, the raters determined whether or not each interpretation provided supporting evidence. Similarly, 22 responses were classified together, after which the remaining data were individually classified. Agreement between the raters reached κ = 1.00. Thus, all responses could be classified under the two categories. Results The materials were lexically accessible; thus, reading time per syllable in the literal comprehension task was shorter than that in the interpretation task. The participants were also more confident about their reading in the literal comprehension task (Nishihara, 2022). The following section presents the findings for the interpretation task, specifically the types of responses presented as interpretations for poems that the participants found easy during the literal comprehension task, confidence level, and reading time. The participants responded to English poems of the same text type in different ways (e.g., they approached A1 and A2 in Table 1 differently). Therefore, data were analyzed by text (A1, A2, B1, B2, C1, C2, D1, D2, and E1 in Table 1) instead of text types A, B, C, D, and E. Reading in a Foreign Language 36(1) Nishihara: Japanese university EFL learners’ responses to lexically easy short English poems 9 Response Type The responses were categorized into four types (Table 2). Table 2. Response type category Category Description Examples Corresponding (Definition) category (Peskin & Wells-Jopling, 2012; Svensson, 1987) Type 1 A response is offered “…I think the poet is saying that the “Literal descriptive” that consists of hesitation the poet felt years ago is like a cold nothing but what is moon” (P3, D2). explicitly expressed “I thought it was an apology, and the poet in the poem. No meant that he was sorry for eating the plum” point is noted. (P11, C1). Type 2 An elaborated “The tree was wilted and lifeless, but your “Literal interpretive” response is presented kisses brought the tree back to life. I don’t that on a literal level know how to describe it, but the leaves at least partly goes became red like autumn time. The tree may beyond what is represent a person” (P5, B1). explicitly expressed “I thought it might be the author’s in the poem. Other experience—that the cold moon reminded parts of the response the author of things that happened many consist of what is years ago, and that the author was in a explicitly somewhat sentimental state as they expressed in the remembered the past” (P12, D2). poem. Type 3 The response “It might be a bit of an over-interpretation, “Mixed literal– consists of a but I interpreted it as if I was getting closer thematic” and generalized inference and closer to the afterlife while I was in a “Thematic” (Svensson, concerning a surface state of grief because someone close to me 1987) point. An abstracted had passed away” (P19, B2). “Thematic or mixed overall meaning is “Well, it occurred to me that during winter, literal thematic” proposed, sometimes in a town with no people, what can I say, the (Peskin & Wells- in the form of a poet was feeling lonely” (P23, D1). Jopling, 2012) saying or saying-like wording. Reading in a Foreign Language 36(1) Nishihara: Japanese university EFL learners’ responses to lexically easy short English poems 10 Type 4 The response “It’s a poem about a child who has a hard “Symbolic” consists of an time staying in the house because it’s so abstract, creative cramped. I think the kid was abused. And statement that could then he went outside and got caught in the not be attributed to rain. The rain washed him clean. Anyway, he what is explicitly finally got to go outside the house he hates” expressed in the (P4, C2). poem. “The author’s loved one, the author’s lover or family member or someone, passed away before the person could eat the plum the person had saved for breakfast. I think the text means that the plum is very cold because it was kept for so long. I wondered if the poem was about grieving for a loved one who has gone away” (P20, C1). Note. Participants (e.g., P1) and the poems (e.g., D2) mentioned in each response are enclosed in parentheses. The author translated the participants’ responses from Japanese to English. The study found no responses corresponding to “mixed literal–symbolic” as described by Svensson (1987) and Peskin and Wells-Jopling (2012). Type 1 is a response that repeats or paraphrases the literal meaning of the text, which was also observed among L1 readers with limited literary knowledge and reading experience (e.g., Goldman et al., 2015; Levine & Horton, 2015; McCarthy, 2015; McCarthy et al., 2021; Peskin, 1998, 2010). Type 2 refers to responses in which most parts of the response are repetitions or a paraphrasing of the literal meaning (similar to Type 1) but with added information. For instance, P5 offered a personal idea in the last part: “The tree may represent a person.” P12 added a detail that was not explicitly stated in the work: “The author was in a somewhat sentimental state as they remembered the past.” However, the added information is modest, which is relatively easy to produce given the literal meaning of the text; it illustrates no variation among the participants. Type 2 responses remain largely dependent on semantic contents. Type 3 responses are relatively easy to construct based on the literal comprehension of the text although this study noted several departures and abstractions from the literal content. Although P19’s response included statements such as “I was getting closer and closer to the afterlife” and “someone close to me had passed away,” these ideas are relatively easy to arrive at based on the expressions “Take me there” and “sorrow” in the text. Similarly, P23 pointed to “a town with no people” and “feeling lonely,” which can be easily deduced based on the description of the scene (“darkness” and “snow-covered car”). The responses classified under this category were not diverse. Type 3 responses lacked sufficient development from the literal context. Type 4 responses depart from the literal context and offer novel, creative interpretations. For instance, no clues in the text could have prompted P4’s ideas about a child’s serious situation. Reading in a Foreign Language 36(1) Nishihara: Japanese university EFL learners’ responses to lexically easy short English poems 11 Similarly, P20 constructed a meaning that cannot be derived directly from the information in the text such as “the poem was about grieving for a loved one who has gone away.” The content of this type of interpretation varied widely across the participants as Peskin and Wells-Jopling (2012) showed in their L1 literary reading research. Some were RWE, whereas others were RWOE (Table 3). For instance, P20 justified his response based on the visual shape of the figurative poem. Similarly, P24 built an interpretation for A1 upon discovering that reading the last letters in each line diagonally from bottom left to top right reveals the same message as that of line 1. Table 3. RWE and RWOE Category Examples RWE “When I first took a quick look at it, because ‘I’ is surrounded by ‘walls,’ I felt that the poet couldn’t escape” (P20, A2, Type 1). “Uh, when I saw this text, it was cut in half, but the cross-section also reads the same as the first line , and it made me think that even if friends or people important to me left me, they are still friends” (P24, A1, Type 4). RWOE “I think the poet attempted to express the change of the tree. The lifeless tree came alive, and the leaves turned red” (P1, B1, Type 1). “Well, I thought the author was saying that there are a lot of things scattered all over the place that you don’t understand at first glance” (P7, D1, Type 4). Note. The participant (e.g., P1), the poem (e.g., B1), and the category (e.g., Type 1) of each response are indicated in parentheses. The author translated their responses from Japanese to English. Italicized parts refer to the evidence presented by the participants to support their interpretation. Table 4 provides the number of each response type. The study observed no statistically significant difference in the total number of occurrences of the four types (χ2(3) = 2.10, p = .552, Cramer’s V = .077). Moreover, although the percentage of the four types occurring in each text differed (Fisher’s exact test: p = .001), a substantial difference was not observed (Cramer’s V = .307). Of the responses analyzed, only 50 (28.7%) were RWE. Approximately three-quarters of the responses were RWOE. A chi-square test illustrated that RWE were statistically less in number than RWOE (χ2(1) = 15.60, p = .000, φ = .218). Although the incidence of RWE was higher in A1 than in other texts, no statistical differences were identified between the texts (Fisher’s exact test: p = .337; Cramer’s V = .234). Table 4. Number of participants’ response types by text Category A1 A2 B1 B2 C1 C2 D1 D2 E1 Total Type 1 0 7 3 5 3 6 4 6 1 35 Type 2 3 10 10 4 5 1 2 2 2 39 Type 3 10 1 4 8 3 7 8 1 9 51 Type 4 8 5 3 6 5 7 7 3 5 49 RWE 11 6 6 6 5 3 4 3 6 50 RWOE 10 17 14 17 11 18 17 9 11 124 Note. RWE and RWOE = responses with evidence and responses without evidence, respectively. Table 5 summarizes the occurrence of RWE and RWOE in the four response types. Reading in a Foreign Language 36(1) Nishihara: Japanese university EFL learners’ responses to lexically easy short English poems 12 Table 5. Occurrence of RWE and RWOE in four response types Category Type 1 Type 2 Type 3 Type 4 Total RWE 4 (11.4%) 11 (28.2%) 12 (23.5%) 23 (46.9%) 50 (28.7%) RWOE 31 (85.6%) 28 (71.8%) 39 (76.5%) 26 (53.1%) 124 (71.3%) Note. RWE and RWOE = responses with evidence and responses without evidence, respectively. Supportive evidence was least frequently included in the Type 1 response but provided in approximately half of the responses for Type 4. The occurrence of RWE was significantly different among the four response types according to Fisher’s exact test (p = .003) although the degree of difference was small (Cramer’s V = .281). Response Patterns Table 6 summarizes the responses of the participants. Table 6. Responses presented by each participant Participants Type 1 Type 2 Type 3 Type 4 RWE RWOE P1 4 1 4 0 1 8 P2 0 1 3 1 2 3 P3 6 1 1 0 0 8 P4 0 0 0 8 5 3 P5 1 5 1 0 0 7 P6 2 2 1 2 4 3 P7 2 1 1 2 1 5 P8 0 1 3 3 0 7 P9 1 4 1 0 0 6 P10 1 4 3 0 4 4 P11 3 4 1 0 4 4 P12 0 2 4 1 0 7 P13 2 1 3 0 1 5 P14 1 1 1 5 8 0 P15 0 2 2 1 2 3 P16 0 0 2 6 5 3 P17 1 2 4 0 3 4 P18 3 0 4 1 0 8 P19 1 1 2 2 0 6 P20 1 1 2 2 1 5 P21 2 4 2 1 2 7 P22 2 0 5 1 0 8 P23 0 1 1 7 4 5 P24 2 0 0 6 3 5 Note. RWE and RWOE = responses with evidence and responses without evidence, respectively. Reading in a Foreign Language 36(1) Nishihara: Japanese university EFL learners’ responses to lexically easy short English poems 13 The study noted relatively large differences in response patterns during the interpretation task. Regarding the four response types, P4, P14, P16, P23, and P24 produced more Type 4 responses, whereas Type 1 was the most common response for P3, Type 2 for P5 and P9, and Type 3 for P22. Meanwhile, P6, P7, P19, and P20 constructed the four types with nearly equal frequencies. The frequency of the occurrence of RWE and RWOE varied widely among individuals. All responses by P14 were RWE, while P1, P3, P5, P7, P8, P9, P12, P13, P18, P19, P20, P21, and P22 mostly did not present supportive evidence for their responses. P2, P6, P11, P15, P17, and P23 presented RWE and RWOE responses with nearly equal frequencies. Confidence Level and Reading Time by Response Type Table 7 summarizes the mean values of confidence level and average reading time per syllable for each response type. Table 7. Confidence level and reading time by response type Category Confidence level Reading time M SD Min Max Mdn M SD Min Max Mdn Type 1 68 24.08 10 100 80 2.42 2.33 0.29 11.42 1.69 Type 2 66.67 21.56 10 100 70 2.69 2.42 0.06 8.98 2.1 Type 3 62.26 21.73 0 100 70 3.96 4.96 0.21 22.5 2.14 Type 4 55.2 18.06 10 80 60 3.25 2.72 0.39 12.56 2.35 RWE 57.9 19.95 10 90 60 3.94 3.92 0.06 17.58 2.65 RWOE 64.23 22.1 0 100 70 2.85 3.2 0.21 22.5 1.85 Note. RWE and RWOE = responses with evidence and responses without evidence, respectively. Figure 1 (see below) suggests that confidence levels tended to decrease sequentially from Types 1 to 4. One-way ANOVA confirmed a statistically significant difference between the four response types (F(3) = 3.21, p = .024) with a nearly medium effect size (η2 = .054). A posthoc test (Tukey’s honestly difference test) indicated a statistically significant difference between Types 1 and 4 (t = 2.72, p = .036). In contrast, one-way ANOVA did not reveal statistically significant differences in reading time among the four types (F(3) = 1.73, p = .163, η2 = .30; see Figure 2). Reading in a Foreign Language 36(1) Nishihara: Japanese university EFL learners’ responses to lexically easy short English poems 14 100 80 60 40 20 0 Type 1 Type 2 Type 3 Type 4 Figure 1. Confidence level for four response types. 5 4.5 4 3.5 3 2.5 2 1.5 1 0.5 0 Type 1 Type 2 Type 3 Type 4 Figure 2. Reading time for the four response types. No statistically significant differences were noted from a t-test between RWE and RWOE although RWOE produced high confidence levels in descriptive statistics (t(172) = −1.76, p = .081, d = .295; Figure 3). Similarly, the t-test did not reveal a statistical difference between the two response types although descriptive statistics displayed a trend toward longer reading times for RWE (t(172) = 1.91, p = .058, d = .320; see Figure 4). Reading in a Foreign Language 36(1) Nishihara: Japanese university EFL learners’ responses to lexically easy short English poems 15 100 90 80 70 60 50 40 30 20 10 0 RWE RWOE Figure 3. Confidence level for RWEs and RWOEs Note. RWE and RWOE = responses with evidence and responses without evidence, respectively. 5 4.5 4 3.5 3 2.5 2 1.5 1 0.5 0 RWE RWOE Figure 4. Reading time for RWEs and RWOEs Note. RWE and RWOE = responses with evidence and responses without evidence, respectively. Table 7 indicates relatively low confidence levels for the four response types, which did not exceed 70%. Discussion Researchers have proposed the usefulness of using linguistically accessible poems as teaching materials for foreign language teaching. However, limited research has been conducted on how EFL learners with high English proficiency and no difficulty in understanding the literal meaning Reading in a Foreign Language 36(1) Nishihara: Japanese university EFL learners’ responses to lexically easy short English poems 16 of such poems interpret them individually during the meaning construction processes. The present study identified several characteristics of learners’ interpretive responses to lexically easy short poems. Upper-intermediate EFL Learners’ Interpretive Responses to Lexically Easy Short English Poems First, when constructing interpretations of English poems, several responses were limited to paraphrasing the literal content of the text while others offered original interpretations. Although the research materials were considered linguistically easy, Type-4 responses comprised only 28.8% (49 responses) of the total. The participants heavily relied on literal content to construct meanings, such that 42.5% of the responses centered on repeating and paraphrasing literal content (Types 1 and 2 included 35 and 39 responses, respectively) and 29.3% (51 responses) were limited to generalized inferences of literal meanings (Type 3). L1 studies have demonstrated that university students without specialized literary training are more likely to engage in paraphrasing in interpretation tasks (Goldman et al., 2015; Levine & Horton, 2015; McCarthy, 2015; McCarthy et al., 2021; Peskin, 1998, 2010).7 The same trend was identified for interpretive responses to lexically easy short English poems by university EFL learners without specialized education in literature and rich poetry reading experience. Constructing original interpretations, which is difficult for L1 readers, is also a challenging task for EFL learners even when poems are linguistically accessible. Second, learners often do not provide evidence to support their interpretations. Out of the 174 responses, only 50 (28.7%) provided supporting evidence. Although two-thirds of the participants provided evidence for at least one response, eight participants did not provide textual evidence for all of the research materials. The prompt did not ask the participants to provide evidence to support their responses to avoid influencing their interpretive process. They may have been less inclined to provide evidence unless specifically instructed to do so (Burkett & Goldman, 2016). Thus, they are not used to supporting their interpretations with explicit references to evidence. L1 studies (Burkett & Goldman, 2016; Lee et al., 2016; McCarthy & Goldman, 2015, 2019; Purves, 1992; Sosa et al., 2016) and EFL studies (Alter & Ratheiser, 2019; Lazar, 2016; McKay, 2001; Zyngier & Fialho, 2010) have widely noted the importance of presenting evidence for literary interpretations. Furthermore, L1 studies have demonstrated that expert readers construct interpretations while referring to various textual evidence during literary reading (Zeitz, 1994), whereas novice readers experience difficulty in presenting evidence (Burkett & Goldman, 2016; Sosa et al., 2016). Perhaps it is unsurprising that EFL learners without literary education and less experience in poetry reading were less accustomed to providing evidence. However, the participants were more likely to provide textual evidence when their responses significantly deviated from the literal content of the text (Table 5). For Type 1 responses, supportive evidence is self-evident because they were merely repetitive or paraphrased literal contents, which discouraged the participants from presenting textual evidence. Conversely, in Type 4 responses, the content of interpretations deviated from the literal content; thus, the Reading in a Foreign Language 36(1) Nishihara: Japanese university EFL learners’ responses to lexically easy short English poems 17 participants may have felt the need to explain, which prompted them to present supporting evidence. Furthermore, the study observed considerable variation patterns in textual engagement among the participants in the interpretation task. Although some were biased toward a particular type of response, others produced the four types in similar degrees. A few of the participants stuck to RWE only or RWOE only, whereas others produced both responses according to the text. The prompt for the interpretive task only instructed them to consider the intended meaning of the authors; thus, the choice of the type of response to the English poems and of offering supporting evidence were entirely theirs. Even if limited, their literary experience and knowledge in L1 and English may have exerted significant individual differences. The varying interpretive responses of diverse readers to the same text were reported for L1 (Burkett & Goldman, 2016) and foreign languages (Hoffstaedter, 1987), which were similar to the results of the current study. Characteristics of Upper-intermediate EFL Learners’ Responses to Lexically Easy Short English Poems Participants’ confidence levels tended to decrease in responses departing from literal meanings. The study noted relative confidence for Type 1 responses, which mainly repeat or paraphrase the text’s literal content (mean confidence level: 68.0%). This level may result from the awareness that the content of responses were fully consistent with literal meanings. In contrast, although the research materials were lexically easy, the participants were less confident when developing Type 4 responses, so that the mean confidence level decreased to 55.2%. The participants who gave such responses were less able to establish a firm basis for their interpretations compared to when they heavily relied on the literal content. Although a few participants reported high confidence levels (e.g., 80% for P4 in B2), the confidence level tended to be low for Type 4 responses, which showed the lowest mean confidence level and least standard deviation. As previously discussed, approximately half of the Type-4 responses were RWEs. The high rate of provision of evidence with Type 4 responses may be because of low confidence levels. Thus, the lack of confidence may have prompted the provision of supportive evidence to justify the validity of interpretations. Type 1 responses may have included two cases: (1) the participants may have initially considered it appropriate to repeat or paraphrase texts’ literal content or (2) they may have intended to present a Type 4 response but were unsuccessful and, therefore, gave a Type 1 response. For instance, during meaning construction in the interpretation task, P9 reported a 100% confidence level in A2, P18 displayed a 90% confidence level in C1, and P13 and P24 reported 10% and 20% confidence levels for B1 and D2, respectively. L1 studies also reported that novice readers intuit that the text may present a deeper meaning than the literal meaning but do not further their meaning construction (Lee, 1995). Of the four types, Type 1 responses exhibited the largest standard deviation. The mixing of the two cases may have contributed to an average confidence level of only 68.0% for Type 1. Reading in a Foreign Language 36(1) Nishihara: Japanese university EFL learners’ responses to lexically easy short English poems 18 Interestingly, the four response types were constructed within similar durations; thus, the study observed no statistically significant differences in reading time per syllable, which is considered a measure of processing effort. Thus, the study infers that Type 1 (repetitive or paraphrased) and Type 4 (creative) responses were produced with the same amount of effort. The presence or absence of evidence did not influence the participants’ confidence level. The analysis did not reveal differences in confidence levels between RWE and RWOE. Thus, citing evidence did not increase confidence. Similarly, no statistically significant difference was noted for reading time per syllable between RWE and RWOE. However, a possibility exists that a few participants may have considered providing supporting evidence but did not do so because of the lack of instruction. Further in-depth research is required to clearly distinguish between RWE and RWOE and to determine the influence of the discovery of supporting evidence on confidence and processing effort (reading time). Pedagogical Issues for Future Studies This study examined how EFL learners construct interpretive meanings of lexically easy short poems. The findings identified several issues that must be considered for such reading material to benefit foreign language learning. First, strategies that encourage learners’ construction of non- literal or deeper meanings must be formulated. L1 studies have illustrated that even readers with limited literary experience or expertise may improve the quality of their interpretation based on prompts (Burkett & Goldman, 2016; Sosa et al., 2016). Peskin and Wells-Jopling (2012) validated that depending on the instructional approach, learners may be encouraged to construct non-literal meanings regardless of their age. Moreover, McCarthy et al. (2021) have highlighted that constructing non-literal meanings brings literary enjoyment, which many teachers want learners to experience with literary texts. Second, teachers must make learners aware of the importance of presenting not only an interpretation but also supporting evidence when engaging with English poetry. In this study, only half of the Type 4 responses cited evidence. Burkett and Goldman (2016) reported that L1 novice readers gave supporting evidence if instructed in the prompt. Researchers could start by testing whether the same effect exists for EFL learners. Third, supporting learners in engaging in non-literal reading without losing confidence is necessary. Teachers must avoid situations where learners’ low confidence leads to their reluctance to construct deeper meanings. Fourth, it is necessary to analyze the EFL learners’ responses to the poems thoroughly. For instance, as this study showed, the same Type 1 response may have included cases where the participants initially deemed repetitive or paraphrased literal contents acceptable. It could also involve cases where they aimed to construct an interpretation similar to Type 4 but were unsuccessful, resulting in a Type 1 response. Reading in a Foreign Language 36(1) Nishihara: Japanese university EFL learners’ responses to lexically easy short English poems 19 Fifth, examining the characteristics of texts that are likely to engage learners in creative interpretation is necessary. Even when given the same prompt, some texts are easier for L1 novice learners to infer deep meanings (McCarthy & Goldman, 2015). In the current study, A1 was more likely to invite Type 4 interpretations compared with other texts.8 A detailed investigation is required to determine which textual features (i.e., graphemic, lexical, syntactic, semantic, pragmatic, and stylistic) encourage learners to engage in audacious meaning construction. Doing so requires an investigation of learners’ responses to various types of English poems. Unfortunately, the stylistic categories established during the selection of research materials did not provide a solution to this issue. Sixth, there is a need to examine the responses of low-proficiency English learners to lexically easy short poems. For instance, previous research suggesting that English poetry can be beneficial in introductory EFL courses (Alter & Ratheiser, 2019; Bland, 2015a, 2015b; Chesnokova & van Peer, 2016; Scott & Huntington, 2007) must be supported by such empirical evidence. The current participants earned relatively high levels of proficiency in English. They may have presented Type 4 RWE because of their proficiency (Hall, 2015; Hanauer, 2010). Studies must explore the influence of literary experiences on learners’ textual engagement with lexically easy short poems. All the participants in this study reported limited literary experience and knowledge. Thus, comparing how learners with different literary competences construct meanings when reading poetry is necessary. Finally, EFL learners must be classified into different types of poetry readers. Although the participants were homogenous in terms of proficiency in English and literary background, considerable diversity existed in response patterns in the interpretation task. Elucidating the types of EFL poetry readers would advance the understanding of the interaction between reader types and texts, which could help explain individual differences in meaning construction patterns, as identified in the current study. Conclusion This study investigated the responses of upper-intermediate Japanese university-level EFL learners to an interpretation task with lexically easy short English poems, which are considered useful teaching materials, on an individual basis. Learners were not always able to create rich interpretations despite being sufficiently proficient in English to understand the literal meanings of texts. Although they offered audacious non-literal interpretations, they also produced many responses such as repetitive or paraphrased literal contents. This study also found a generally less frequent presentation of responses with supporting evidence. Moreover, the learners displayed low confidence levels in creative non-literal interpretations (Type 4). The results indicated that learners require pedagogical support to engage in poetry despite sufficient English proficiency levels. Thus, researchers and teachers have to develop procedures for guiding learners to present deeper meanings and supportive evidence with confidence, which will not only lead to rich Reading in a Foreign Language 36(1) Nishihara: Japanese university EFL learners’ responses to lexically easy short English poems 20 language learning and literary reading experiences but also enable learners to enjoy constructing meanings from poetry. For this purpose, it is essential to empirically investigate the responses of learners with different English proficiency levels and poetry experience to different types of poetic texts. Researchers must also investigate the types of poems considered useful for learners and the influence of learners’ personal traits on their interpretive responses. Notes 1. These studies conceive of reading as a dialogic practice (e.g., Bland, 2023a) and provide detailed suggestions for instructional procedures and useful resources for learners to share their reading experiences with each other in pairs, groups, and in class to collaboratively make sense of texts (e.g., Albers, 2015; Bland, 2015b, 2023; Scott & Huntington, 2007). 2. The same subcategories of literary reading are also presented in other studies such as those by Schat et al. (2023) and Schrijvers et al. (2019). 3. Fahnestock and Secor (1991) and Warren (2006) called this assumption the appearance-reality topoi. 4. The results section provides the definitions of each response type. 5. SVL 12000 is a vocabulary list with a total of 12 word frequency levels based on data from the English language use of native speakers. It was developed by ALC PRESS Inc., an ELT material publisher in Japan. 6. During interviews, participants were also asked about their literal comprehension of the text, the lines they considered difficult (and the order of difficulty in the case of two or more lines), their confidence level in their performance, and the texts they deemed difficult (and the order of difficulty in the case of two or more such texts). For the results of this interview, see Nishihara (2022). 7. However, some studies have reported that some novice and young readers without special instruction on literary reading engage in the construction of non-literal meanings of literary texts (Andringa, 1990; de Beaugrande, 1987). Meantime, Harker (1994) revealed that certain readers, even highly capable ones, are limited to interpreting the literal meanings of poems. 8. Moreover, more than half of the responses to A1 (11 out of 21) were RWEs. A1 has very little information and its structure is such that a sentence is broken off in the middle without being completed and repeated with more missing information (no complete sentence is included). The participants provided no information to paraphrase the poetry text and were forced to use imagination to construct meanings, which may have increased the need to justify responses compared with other research materials. Reading in a Foreign Language 36(1) Nishihara: Japanese university EFL learners’ responses to lexically easy short English poems 21 References Albers, C. (2015). Poetry in the intermediate EFL classroom. In W. Delanoy, M. Eisenmann, & F. Matz (Eds.), Learning with literature in the EFL classroom (pp. 103–120). Peter Lang. Alter, G., & Ratheiser, U. (2019). A new model of literary competences and the revised CEFR descriptors. ELT Journal, 73(4), 377–386. https://doi.org/10.1093/elt/ccz024 Andringa, E. (1990). Verbal data on literary understanding: A proposal for protocol analysis on two levels. Poetics, 19(3), 231–257. https://doi.org/10.1016/0304-422X(90)90022-W Ariyanto, S. (2021). Using intercultural rhetoric to teach poetry: An innovative reading task to build students’ critical thinking. TESOL Journal, 12(1), 1–5. https://doi.org/10.1002/tesj.524 Badran, D. (2007). Stylistics and language teaching: Deviant collocation in literature as a tool for vocabulary expansion. In M. Lambrou & P. Stockwell (Eds.), Contemporary stylistics (pp. 180–192). Continuum. Bernhardt, E. (2023). Understanding literary texts in a second language: Instances of advanced and superior comprehension. In M. M. Echevarría (Ed.), Rehumanizing the language curriculum (pp. 101–117). Peter Lang. Billows, F. (1961). The technique of language teaching. Longman. Bland, J. (2015a). Grammar templates for the future with poetry for children. In J. Bland (Ed.), Teaching English to young learners: Critical issues in language teaching with 3–12 year olds (pp. 147–166). Bloomsbury. Bland, J. (2015b). Performing poems in the primary school. In W. Delanoy, M. Eisenmann, & F. Matz (Eds.), Learning with literature in the EFL classroom (pp. 85–100). Peter Lang. Bland, J. (2018). Introduction: The challenge of literature. In J. Bland (Ed.), Using literature in English language education: Challenging reading for 8–18 year olds (pp. 1–22). Bloomsbury. Bland, J. (2023). Compelling stories for English language learners: Creativity, interculturality and critical literacy. Bloomsbury. Brooks, C., & Warren, R. P. (Eds.). (1978). Understanding poetry (4th ed.). Heinle and Heinle. Burkett, C., & Goldman, S. R. (2016). “Getting the point” of literature: Relations between processing and interpretation. Discourse Processes, 53(5–6), 457–487. https://doi.org/10.1080/0163853X.2016.1169969 Castiglione, D. (2017). Difficult poetry processing: Reading times and the narrativity hypothesis. Language and Literature, 26(2), 99–121. https://doi.org/10.1177/0963947017704726 Chesnokova, A. & van Peer, W. (2016). Anyone came to live here some time ago: A cognitive semiotics approach to deviation as a foregrounding device. Versus: Quaderni di studi semiotici, 122, 1–18. https://doi.org/10.14649/84875 Civelekoğlu, N. & Saka, F. Ö. (2017). The use of poetry to raise intercultural awareness of EFL students. The Journal of Language Teaching and Learning, 8(1), 39–52. Claassen, E. (2012). Author representations in literary reading. John Benjamins. de Beaugrande, R. (1987). The naive reader: Anarchy or self-reliance? Empirical Studies of the Arts, 5(2), 145–170. https://doi.org/10.2190/PBCQ-AUJ0-765C-26NR Reading in a Foreign Language 36(1) Nishihara: Japanese university EFL learners’ responses to lexically easy short English poems 22 Delanoy, W. (2018). Literature in language education: Challenges for theory building. In J. Bland (Ed.), Using literature in English language education: Challenging reading for 8–18 year olds (pp. 141–157). Bloomsbury. DiYanni, R. (Ed). (2000). Poetry: An introduction. McGraw-Hill. Duncan, S., & Paran, A. (2018). Negotiating the challenges of reading literature: Teachers reporting on their practice. In J. Bland (Ed.), Using literature in English language education: Challenging reading for 8–18 year olds (pp. 243–259). Bloomsbury. Earthman, E. A. (1992). Creating the virtual work: Reader’s processes in understanding literary texts. Research in the Teaching of English, 26(4), 351–384. Fahnestock, J., & Secor, M. (1991). The rhetoric of literary criticism. In C. Bazerman & J. Paradis (Eds.), Textual dynamics of the professions: Historical and contemporary studies of writing in professional communities (pp. 77–96). University of Wisconsin Press. Goldman, S. R., McCarthy, K. S., & Burkett, C. (2015). Interpretive inferences in literature. In E. J. O’Brien, A. E. Cook, & R. F. Lorch, Jr. (Eds.), Inferences during reading (pp. 386– 415). Cambridge University Press. Guy, J. M.., Conklin, K., & Sanchez-Davies, J. (2018). Literary stylistics, authorial intention and the scientific study of literature: A critical overview. Language and Literature, 27(3), 196–217. https://doi.org/10.1177/0963947018788518 Hall, G. (2015). Literature in language education (2nd ed). Palgrave Macmillan. Hanauer, D. I. (1997). Poetry reading in the second language classroom. Language Awareness, 6(1), 2–16. https://doi.org/10.1080/09658416.1997.9959912 Hanauer, D. I. (2001). The task of poetry reading and second language learning. Applied Linguistics, 22(3), 295–323. https://doi.org/10.1093/applin/22.3.295 Hanauer, D. I. (2010). Poetry as research: Exploring second language poetry writing. John Benjamins. Harker, W. J. (1994). “Plain sense” and “poetic significance”: Tenth-grade readers reading two poems. Poetics, 22(3), 199–218. https://doi.org/10.1016/0304-422x(94)90005-1 Hoffstaedter, P. (1987). Poetic text processing and its empirical investigation. Poetics, 16(1), 75– 91. https://doi.org/10.1016/0304-422X(87)90037-4 Hoover, P. (Ed.). (1994) Postmodern American poetry. W. W. Norton. Kacian, J., Rowland, P., & Burns, A. (Eds.). (2013). Haiku in English: The first hundred years. W. W. Norton. Kennedy, X. J., & Gioia, D. (Eds). (2002). An introduction to poetry (3rd ed.). Longman. Kennedy, X. J., & Kennedy, D. M. (Eds). (1999). Knock at a star: A child’s introduction to poetry. Little, Brown and Company. Lazar, G. (2016). Literature and language teaching. In R. H. Jones (Ed.), The Routledge handbook of language and creativity (pp. 468–482). Routledge. Lee, C. D. (1995). A culturally based cognitive apprenticeship: Teaching African American high school students skills in literary interpretation. Reading Research Quarterly, 30(4), 608– 630. https://doi.org/10.2307/748192 Lee, C. D., Goldman, S. R., Levine, S., & Magliano, J. (2016). Epistemic cognition in literary reasoning. In J. A. Greene, W. A. Sandoval, & I. Bråten (Eds.), Handbook of epistemic cognition (pp. 165–183). Routledge. Reading in a Foreign Language 36(1) Nishihara: Japanese university EFL learners’ responses to lexically easy short English poems 23 Levine, S. (2019). Using everyday language to support students in constructing thematic interpretations. Journal of the Learning Sciences, 28(1), 1–31. https://doi.org/10.1080/10508406.2018.1485023 Levine, S., & Horton, W. (2015). Helping high school students read like experts: Affective evaluation, salience, and literary interpretation. Cognition and Instruction, 33(2), 125– 153. https://doi.org/10.1080/07370008.2015.1029609 Lin, B. (2006). Exploring the literary text through grammar and the (re-)integration of literature and language teaching. In A. Paran (Ed.), Literature in language teaching and learning (pp. 101–116). TESOL. Maley, A., & Moulding, S. (1985). Poem into poem: Reading and writing poems with students of English. Cambridge University Press. McCarthy, K. S. (2015). Reading beyond the lines: A critical review of cognitive approaches to literary interpretation and comprehension. Scientific Study of Literature, 5(1), 99–128. https://doi.org/10.1075/ssol.5.1.05mcc McCarthy, K. S., & Goldman, S. R. (2015). Comprehension of short stories: Effects of task instructions on literary interpretation. Discourse Processes, 52(7), 585–608. https://doi.org/10.1080/0163853X.2014.967610 McCarthy, K. S., & Goldman, S. R. (2019). Constructing interpretive inferences about literary text: The role of domain-specific knowledge. Learning and Instruction, 60(1), 245–251. https://doi.org/10.1016/j.learninstruc.2017.12.004 McCarthy, K. S., Magliano, J. P., Levine, S. R., Elfenbein, A., & Horton, W. S. (2021). Constructing mental models in literary reading: The role of interpretive inferences. In D. Kuiken & A. M. Jacobs (Eds.), Handbook of empirical literary studies (pp. 85–118). Walter de Gruyter. McKay, S. L. (2001). Literature as content for ESL/EFL. In M. Celce-Murcia (Ed.), Teaching English as a second or foreign language (3rd ed., pp. 319–332). Heinle & Heinle. McRae, J., & Vethamani, M. E. (Eds). (1999). Now read on: A course in multicultural reading. Routledge. Nishihara, T. (2022). EFL learners’ poetry reading traits for lexically easy short poetry. Cogent Education, 9(1), Article 2150010. https://doi.org/10.1080/2331186X.2022.2150010 Paran, A. (2008). The role of literature in instructed foreign language learning and teaching: An evidence-based survey. Language Teaching, 41(4), 465–496. https://doi.org/10.1017/S026144480800520X Paran, A., & Robinson, P. (2015). Literature. Oxford University Press. Paran, A., & Wallace, C. (2016). Teaching literacy. In G. Hall (Ed.), The Routledge handbook of English language teaching (pp. 441–455). Routledge. Peplow, D., Swann, J., Trimarco, P., & Whiteley, S. (2015). The discourse of reading groups: Integrating cognitive and sociocultural perspectives. Routledge. Peskin, J. (1998). Constructing meaning when reading poetry: An expert–novice study. Cognition and Instruction, 16(3), 235–263. https://doi.org/10.1207/s1532690xci1603_1 Peskin, J. (2010). The development of poetic literacy during the school years. Discourse Processes, 47(2), 77–103. https://doi.org/10.1080/01638530902959653 Reading in a Foreign Language 36(1) Nishihara: Japanese university EFL learners’ responses to lexically easy short English poems 24 Peskin, J., & Hanauer, D. I. (2023). A life with poetry: The development of poetic literacy. John Benjamins. Peskin, J., & Wells-Jopling, R. (2012). Fostering symbolic interpretation during adolescence. Journal of Applied Developmental Psychology, 33(1), 13–23. https://doi.org/10.1016/j.appdev.2011.08.002 Purves, A. C. (1992). Testing literature. In J. A. Langer (Ed.), Literature instruction: A focus on student response (pp. 19–34). NCTE. Ramazani, J., Ellmann, R., & O’Clair, R. (Eds). (2003). The Norton anthology of modern and contemporary poetry (3rd ed). W. W. Norton. Riddell, A. (1972). Eclipse: Concrete poems. Calder and Boyars. Rosenkjar, P. (2006). Learning and teaching how a poem means: Literary stylistics for EFL undergraduates and language teachers in Japan. In A. Paran (Ed.), Literature in language teaching and learning (pp. 117–131). TESOL. Sanford, A. J., & Emmott, C. (2012). Mind, brain and narrative. Cambridge University Press. Schat, E., van der Knaap, E., & de Graaff, R. (2023). Key principles for an integrated intercultural literary pedagogy: An educational design research project on arts integration for intercultural competence. Language Teaching Research, 27(2), 332–358. https://doi.org/10.1177/13621688211045012 Scott, V. M., & Huntington, J. A. (2007). Literature, the interpretive mode, and novice learners. The Modern Language Journal, 91(1), 3–14. https://doi.org/10.1111/j.1540- 4781.2007.00506.x Schrijvers, M., Janssen, T., Fialho, O., De Maeyer, S., & Rijlaarsdam, G. (2019). Transformative dialogic literature teaching fosters adolescents’ insight into human nature and motivation. Learning and Instruction, 63, Article 101216. https://doi.org/10.1016/j.learninstruc.2019.101216 Sigvardsson, A. (2020). Don’t fear poetry! Secondary teachers’ key strategies for engaging pupils with poetic texts. Scandinavian Journal of Educational Research, 64(6), 953–966. https://doi.org/10.1080/00313831.2019.1650823 Sosa, T., Hall, A. H., Goldman, S. R., & Lee, C. D. (2016). Developing symbolic interpretation through literary argumentation. Journal of the Learning Sciences, 25(1), 93–132. https://doi.org/10.1080/10508406.2015.1124040 Stockwell, P. (2016). Cognitive stylistics. In R. H. Jones (Ed.), The Routledge handbook of language and creativity (pp. 218–230). Routledge. Svensson, C. (1987). The construction of poetic meaning: A developmental study of symbolic and non-symbolic strategies in the interpretation of contemporary poetry. Poetics, 16(6), 471–503. https://doi.org/10.1016/0304-422X(87)90014-3 Tomlinson, B. (1986). Using poetry with mixed ability language classes. ELT Journal, 40(1), 33– 41. https://doi.org/10.1093/elt/40.1.33 Warren, J. E. (2006). Literary scholars processing poetry and constructing arguments. Written Communication, 23(2), 202–226. https://doi.org/10.1177/074108830628686 Widdowson, H. G. (1982). The use of literature. In M. Hines & W. Rutherford (Eds.), On TESOL ’81: Selected papers from the Fifteenth Annual Conference of Teachers of Reading in a Foreign Language 36(1) Nishihara: Japanese university EFL learners’ responses to lexically easy short English poems 25 English to Speakers of Other Languages, Detroit, Michigan, March 3-8, 1981 (pp. 203– 214). TESOL. Widdowson, H. G. (1992). Practical stylistics. Oxford University Press. Zeitz, C. M. (1994). Expert-novice differences in memory, abstraction, and reasoning in the domain of literature. Cognition and Instruction, 12(4), 277–312. https://doi.org/10.1207/s1532690xci1204_1 Zyngier, S. (1999). The unheard voices: A reader model for students. Language Awareness, 8(1), 30–37. https://doi.org/10.1080/09658419908667115 Zyngier, S., & Fialho, O. (2010). Pedagogical stylistics, literary awareness and empowerment: A critical perspective. Language and Literature, 19(1), 13–33. https://doi.org/10.1177/0963947009356717 Zyzik, E., & Polio, C. (2008). Incidental focus on form in university Spanish literature courses. The Modern Language Journal, 92(1), 53-70. https://doi.org/10.1111/j.1540- 4781.2008.00686.x About the Author Takayuki Nishihara is Associate Professor of Applied Linguistics at Hiroshima University, Japan. His research interests include literary reading processes in a foreign language, EFL literary reading difficulties, teaching and testing procedures for literary texts in EFL, and the effects of literary reading on foreign language learning. He is the author of “EFL Learners’ Poetry Reading Traits for Lexically Easy Short Poetry” (Cogent Education, 9, 2022), “Achievement Tests for Literary Reading in General EFL Reading Courses” (M. Teranishi, Y. Saito, & K. Wales (eds.), Teaching Literature and Language in the EFL Classroom, Palgrave Macmillan, 2015), and “Of Learning and Poetics: Exploring Strategies Used by L2 Japanese English Learners” (M. Burke, O. Fialho & S. Zyngier (eds.), Scientific Approaches to Literature in Learning Environments, John Benjamins, 2016). E-mail: ntakayk@hiroshima-u.ac.jp Reading in a Foreign Language 36(1)