Academic reading as a social practice in higher education

In this study, an undergraduate teacher education course is used to explore whether and how academic reading seminars reflect the theoretical notion of academic literacies and provide a learning environment for developing academic and professional learning and engagement. The data analyzed in this article are transcribed recordings of small group activities where students discuss scientific articles based on a template. First, our empirical analysis shows that the use of the template facilitated dialogical discussions and the development of a cognitive skillset and disciplinary categories when used in a social setting. Second, we found that the most challenging part of designing a reading practice related to the academic literacies tradition was fostering a dialogical environment for discussing the validity of findings across different contexts and provide for discussions encompassing complexity, nuances, and meaning making. We found traces of such discussions in all the transcripts; however, many examples were in a premature stage. The paper concludes with a discussion on, and some suggestions for, further development of the template used in the reading seminars.


Introduction
In contrast to other language and literacy activities, higher education (HE) students' writing has long been foregrounded in political and academic discussions on how HE learning outcomes might be improved (Lillis & Scott, 2007). Academic writing is viewed as a "high-stakes activity" in HE because it often constitutes the main form of assessment. In contrast, this article explores how academic reading as a social practice may develop various aspects of academic literacies among undergraduate students. Our research is theoretically founded in an academic literacies research tradition, meaning that we have expanded the focus from the text as a cognitive and linguistic device to the text as a social practice (Lillis & Scott, 2007;Paxton & Frith, 2014).

3
The context of the present study is the complex picture of requirements for HE programs, beyond the scope of introducing the students to a disciplinary and/or professional field, that have been introduced in Europe over the last 20 years. A call for HE programs to be research based (Brew, 2003) and societally relevant (Afdal, 2016) is strongly present. Global competitiveness has created a need for developing twenty-first-century skills, such as active learning, critical reflection, and problem-solving (Greiff et al., 2014) at all HE levels. Designing HE programs and particularly professional programs is often complex, giving the experience of having to do to integrate contradicting aims such as professional aspects as well as research-based knowledge. This study is part of a longitudinal project exploring student teachers' inquiry-based learning experiences as they progressed through a 60 ECT course (The European Credit Transfer and Accumulation System) per year in Science of Education at the undergraduate level (Afdal & Spernes, 2018;Spernes & Afdal, 2021). An undergraduate teacher education course is used to explore whether and in what ways academic reading seminars accomplish the following: (1) reflect the theoretical notion of academic literacies and (2) provide a learning environment that develops academic and professional learning and engagement.
The data in this article is collected among Norwegian teacher education students in Norway attending the compulsory course "Science of Education." The course covered topics such as learning theories, classroom practices, teacher role and identity, teacher professionalism, and curriculum theory. The teacher education at the time of the study was a 4-year integrated program (BA + 1 year). In the context of this paper, academic reading seminars are a small group activity where students discuss scientific articles based on a template (Appendix 1). The data source in this study is transcripts of audio recordings of the group discussions. We elaborate more on the context of the reading seminars in the methods section. The main arguments for initiating academic reading seminars was that we have seen undergraduate students struggling with academic reading and acquiring knowledge from scientific literature. Baker (2018) identified an "epistemological transition" when students move from compulsory education to undergraduate studies. We also recognized that HE educators expect students to deal with academic texts without training or support.
Below, we introduce earlier research on academic reading from various research traditions before elaborating on the research design and methodology, as well as the theoretical underpinning in the article.

Earlier research on academic reading
Before the 2000s, most studies on academic reading in HE focused on student learning through text and usually explored the relationship between approach, conception, and outcome. Later, in their scoping review, Baker et al. (2019) showed four major strands or conceptions of academic reading in existing literature on academic reading in HE. First, cognitive conceptions emphasize functional skills, such as eliciting meaning from texts, decoding symbols, and applying syntactic and semantic knowledge (e.g., Gorzycki et al., 2016;Wilson, 2016). Other examples are Tomasek (2009), who explored the use of reading prompts to promote active engagement with texts before coming to class. Rhead (2019) found that joint reading retreats (students and faculty) developed practical skills and technical strategies, whereas Francis and Hallam (2000) considered how academic reading influenced the students' knowledge about genres. Bharuthram and Clarence (2015) focused on academic reading as a successful disciplinary knowledge practice in teaching and learning. Second, the compliance conception, including research, seeks to understand why students do or do not complete assigned readings and students' assumptions and values regarding reading practices. Here, reading is understood as a textual practice valued by academia (e.g., Kerr & Frese, 2017;St Clair-Thompson et al., 2018). Third, there is a body of research that uses a contextualized definition, meaning that the focus is on engaging issues of reader identity/ ies, considering not only cognitive processes but also affective and psychological factors (e.g., Lockhart & Soliday, 2016). Finally, more recent studies emphasize the sociopolitical nature of reading, viewing practices as fluid, highly complex, context dependent, and based on the reader's engagement with background knowledge, schematic understandings, and ideological perspectives (e.g., Baker, 2018;Carillo, 2016). In sum, Baker et al. (2019) argued that there is an overall lack of studies on academic reading beyond an Anglophone context; case studies on students situated reading processes and practices, especially from an academic literacies perspective; and studies focusing on students' reading practices as an equity/social justice practice.
Following Baker et al.'s (2019) review, a recent and extensive research contribution drawing mostly on the academic literacies tradition has been published in a special issue in the Journal of University Teaching and Learning Practices (Rhead & Little, 2020). The special issue offers papers in two strands of research. The first strand focuses on the perception of academic reading from students' and academics' perspectives, pointing at how academic reading is valued because it is often not explicitly taught and students are not often engaged in conversations about the nature of scholarship in their disciplines (Maguire et al., 2020;Miller & Merdian, 2020). Kimberley & Thursby (2020) found a lack of confidence in academic reading and that being an adequate reader is the primary barrier to engagement. They recommend increased support for reading through an organized framing of students' encounters with challenging texts. The second strand of research concerns practices and approaches to teaching and learning academic reading, exploring various creative and collaborative activities and practices (Cowley-Haselden, 2020;Nguyen & Henderson, 2020). Studies from both strands show a developmental embodied sequence of learning occurs via collaborative work in social space where dialogic and multimodal interaction with texts is fostered.

Academic reading as a social practice
The first part of our research question relates to the notion of the academic literacies model initiated by Lea (2004), Lea and Street (2006) and Lillis (2003), further developed by Jacobs (2015). We aim to explore whether and how our reading seminars foster a learning environment for developing academic literacies. The academic literacies model is rooted in social and critical linguistics and influenced by socio-cultural perspectives on teaching and learning. The model is developed as a possible frame for developing academic literacies with a focus on pedagogy (Lea & Street, 2006, p. 370). Using the plural form signals the epistemological position that literacy develops through transformative social practices that may take various linguistic forms within and across HE contexts (Lillis & Scott, 2007, p. 16). In academic literacies, research and pedagogy are intertwined; the educational design or activity is not separate from but embedded in the research activity (Lillis & Scott, 2007).
Jacobs (2015, p. 138) expands on this and argues that for lectures to provide for an academic environment with focus on academic literacies, they need to make a conceptual shift 1 3 from a normative toward a transformative pedagogy. She further argues that partnership and collaboration between the lectures, and lectures and students when developing teaching and learning approaches are needed. Taking both a historical (cf. our review section) and empirical point of departure, Jacobs analytically and pedagogically illustrates academic literacies teaching as threefold: academic skills understanding, academic socialization, and academic literacies understanding. See Fig. 1.
The "understandings" represents three different modes of learning. (a) implies that approaching academic texts is primarily an individual cognitive process, and it will be unproblematic to transfer knowledge of how to approach texts from one context to another. It is primarily concerned with written language at the surface level, and learning is seen as transmission of knowledge. In (b), acculturation into discipline-and subject-based discourses and genres is necessary. Discourses and genres are presumed to be relatively stable, and learning is viewed as construction situated in a specific field. (c) builds on (b) but requires an environment where students can approach texts in a complex, nuanced, and situated way. Activities based on the academic literacies model need to engage in meaning making and involve both epistemological and social processes. What counts as knowledge is contextually bounded, and the model emphasizes a theory of learning that foregrounds identity, agency, power, and the role of language. We have used the academic literacies model in designing the academic reading seminars. Following Jacobs, academic literacies practice has the potential for transforming educational settings and activities. However, it requires a back and forth movement on the continuum shown in Fig. 1.

Academic reading seminars
According to Paxton and Frith (2014), there are several advantages of academic literacies approach for knowledge making and curriculum design. Specifically, it provides educational designs/activities where knowledge and learning processes are thoroughly developed and made explicit for both educators and students. In addition, the educators' expectations for and students' understanding of semiotic practices are dialogical and interwoven. Following Bakhtin, we understand dialog as an "intense interaction and struggle between one's own and others' words" and as a process in which the actors oppose or dialogically interanimate each other (Bakhtin, 1935(Bakhtin, /1881. Hence, academic literacies approach aims to provide an environment where vocabulary, concepts, and phrases are discussed and misconceptions can be avoided. Various relevant genres and conventions with specific  (Jacobs, 2015, p. 136) fields/disciplines can be discussed and practiced communally. Finally, this approach provides an environment for connecting various types of knowledge, including theoretical, practical, and methodological knowledge.
The initial aims when developing the design of the academic reading seminar were that the activity should.
• Provide an environment where academic reading is a social activity • Initiate critical academic reading for the purpose of understanding, not just remembering • Initiate critical discussions on research methods, quality of academic texts, genre, and trustworthiness in academic texts • Provide the opportunity to connect resources acquired through academic reading with the genre of the discipline • Give opportunities to make sense of, and use theoretical and professional concepts (meaning making) actively in discussions • Initiate discussions where the students might relate research to their teaching practice and vice versa Over time, we have recognized that the students need scaffolding for academic reading, so we decided to develop a template (see Appendix 1) that could facilitate the above aims. The template was developed as a guide that could assist students in approaching scientific articles and covering all aspects of the literacies (a-c). Since academic literacies studies are a socially oriented approach, observation of the practices surrounding literacy practices, as well as the participants' perspectives on textual/linguistic practices, should be studied (Lillis & Scott, 2007).
We focused on empirical articles, and hence, the profile of the template, for three reasons. First, students often encounter theoretical perspectives in the course Science of Education through textbooks; in empirical studies, they see how theoretical perspectives are used to understand and develop educational practices and new theoretical perspectives. Second, students have constant access to new knowledge from their field, and new empirical studies in the format of journal articles are easy to access both as students and as future professionals. Third, the more extensive texts that students are expected to produce during their teacher education (BA and MA theses) are usually empirically based. However, we see no problem in adjusting or developing similar templates for other article genres in various disciplines. The template has been developed over years based on issues raised by the students in the seminars and our reflection around the design of the activity. The text in the template in italics represents questions and words that we have added over the course of the study. We comment further on this in the Results section.
We used approximately 3 h (3 × 45 min) per reading seminar. In the first hour of the class, the students read the article (or they did this at home and came an hour later); they then discussed the article in groups. In our experience, groups of five to seven students are ideal. We organized the group sessions in different ways over the years. However, the students generally worked in one group first and then regrouped and discussed the article a second time. The students either discussed the whole template in both groups or divided the bullet points among the groups, allowing for in-depth discussions on, for example, "method," "theory," or "relevance" (Appendix 1). To create an epistemological egalitarian dynamic in the groups, we deliberately designed the study in such a way that the lecturer was not part of the initial discussion. We ended each seminar with a plenary session, which improved the discussion and added nuance while allowing the students to ask clarifying questions of each other and the lecturer. The students read increasingly advanced literature in the course, and selecting scientific articles independently toward the end.

Data collection and analytical strategy
Scrutinizing social reading practices offers the opportunity to link texts, language, context, and agency. The empirical data analyzed in this article are recordings of the group discussions in the scientific reading seminars where the lecturers were not present.
We collected recordings from three cohorts of students between 2013 and 2019. See Table 1 for an overview of the data.
A theory-inspired content analysis (Elo & Kyngäs, 2008) was employed for the transcribed recordings. Table 2 shows how we operationalized the theoretical framework. Each theoretical code represents characteristics from the three "understandings" in Fig. 1, and the textual identifications guided the coding of the data material. The codes and their textual identifications were further developed and adjusted while the three authors jointly coded the five transcripts; hence, we allowed the codes to be shaped by the empirical data. The rest of the coding was distributed equally among the authors. The transcripts were coded using Atlas.ti software.
The authors translated and numbered excerpts from the transcripts to allow for discussions across the excerpts; the students are identified as S1, S2, S3, and so on to highlight different voices (not the individual students) across the data. Some words in the excerpts are bolded to highlight arguments in the analysis.

Results
The analysis is presented in a threefold structure, with each section representing the threefold literacies model.

Language socialization and transmission of knowledge
Approaching academic texts involves approaching a new type of vocabulary for HE students. Seminars seem to initiate two types of lexical discussions-considering unfamiliar academic concepts on the one hand and using particular concepts commonly employed in professional settings on the other.
Regarding academic language, the students dwelled a lot on new and unfamiliar concepts in their discussions. It created frustration that the fluidity of their reading was reduced. Unfamiliar concepts often halted the reading process. The character of the concepts was complex, and a lexical translation or dictionary was not sufficient to grasp the deeper meaning and textual coherence. Strikingly, throughout our data, we find that the students in the group sessions accepted relatively vague definitions of unfamiliar academic concepts. Excerpts 1 and 2 show how the students identified challenging concepts-"cultural capital" and "ethnographic study." Excerpt 1: S1: She could have used more space to explain the concept of cultural capital. I'm struggling with what she means by that. S2: The concept is not described clearly. S1: So, you mean she should have argued more clearly Other examples of similar discussions in our data concerned such concepts as tautology, hypothesis, and poststructuralism. We found little evidence of deeper explorations and discussions of new and unfamiliar academic concepts and the necessities for them as the basis for grasping the content and bearings of the main message of the text. An important reason for that might be because the template does not satisfactory facilitate such discussions. Still, we found that the students increasingly acquired and used academic vocabulary while discussing scientific articles, which they recognized toward the end of the course. In the transcripts, the students try out vocabulary they have encountered in the articles, exploring the use of such concepts related to various sections of the articles and matters emphasized in the template. They did this while, for example, identifying the main topic (excerpt 3) or discussing the method section (excerpt 4), findings/ conclusion (excerpt 5), and relevance of the findings (excerpt 6).
Excerpt 3: S1: …Main topic. What is the main topic? S2: …Well, we conclude that there is an asymmetric relationship between the class teacher and the bilingual teacher. It is pervasive throughout the article.
Excerpt 3 is from a seminar where students discussed an article about the professional relationship between class teachers and bilingual teachers. The student was trying out the concept of an "asymmetric relationship" to clarify the main topic of the article. In excerpt 4, students make sense of the main method for collecting data in another article where the author conducted an analysis of science textbooks for lower secondary school:

3
Excerpt 4: Analysis of texts, where she has examined 24 books […] from 1974 to 2007; she also has a survey of scientific researchers within the natural sciences. She has used the hypothesis that Arab scientists are not mentioned in Norwegian science books. The method is discourse analysis, which is a method in sociolin-guistics… It is a method for analyzing oral and written texts. It is a qualitative method.
In excerpt 5, the students try out language and concepts originating from the philosopher Lögstrup as introduced in an article about professional ethics. In excerpt 6, the topic is inclusive education.
Excerpt 5: S1: He [the author] defined power as something to use positively and negatively. It is not the use of it, but one should facilitate to guide and help the pupils. The life of the pupil is in your hands, and the conclusion is that the best way to do it is love and acceptance, but that one should make demands too.
S2: He thinks one should show love when teaching them… S3: Ethics goes deeper than morality and other norms; it lies in the spontaneous expressions… S1: If you respond with life expression and do not allow yourself to be tied up in things, that is the ethical way of teaching. Real, open, and honest with the pupil.
Excerpt 6: What changes can the findings contribute to? … What she [the author] has discovered is that to counteract dualism, one must start early. By showing pupils pictures that are real because this creates a sense of security where they see similarities instead of differences… Excerpts 5 and 6 show how the students combine the act of trying out new concepts with attempting to grasp the concepts' deeper meaning.
A prominent finding across our data is the role of identifying RQs to grasp the article's main message and content. The students use RQs to navigate and search systematically for the article's content knowledge of the article. Excerpt 7 is an example from a discussion on the article discussed in excerpt 5.
Excerpt 7: It is to inform about the use of ethics in teaching contexts and the importance of creating trust between the teacher and students. The problem is: teaching ethics, is there anything to care about? It is the title, and the author specifies what he means by that. The article is difficult. It is advanced, but the theme is there all the time-he follows it, but it is difficult to perceive what is what. He does not answer the [problem] directly or say how important teaching ethics is, but he builds the conclusion with good arguments on why he cares, so he answers but not directly.
The focus on RQs has been emphasized for the students in teaching on research methodology and academic writing, and our data clearly show that using an RQ as a "guiding star" helps the students decipher the text's deeper meaning.
So far, the analysis has shown that the template provides an opportunity for students to make sense of academic texts in a group, systematically approach their reading, and collectively decode the text. They acquire an active and increased academic vocabulary, but they do not display a deeper search for understanding and meaning.

Knowledge about genre and academic socialization
How did we find that our template worked as a guide or device for bringing the students into the academic genre of the discipline and into a process of disciplinary socialization? Beyond focusing on RQs, the template challenged students to consider methodological issues, validity, trustworthiness, research ethics, and textual structures (cf. Table 2).
As we discussed for excerpt 7, the focus on identifying RQ(s) helped the students decipher the main theme of the text; students also became aware of the importance of the relationship between RQs and the texts' structures. They highlighted the importance of RQs for identifying the main findings and conclusions in the article. Across the material, the students particularly engaged with discussing whether RQs were properly formulated and answered, as well as how they were answered. In some cases, they identified clear-cut questions and answers.
Excerpt 8: S1: It does not come out very clear, but she [the author] writes that she wants to shed light on how young people's autonomy and processes of separation from adults are actively constructed in meaningful processes in the interaction between school culture and students' gender cultures. So, it will be the research question even if she does not write directly that this is the research question. It comes as a question, but it also appears in the introduction …. S2: It seems a bit vague. S3: It is a bit woven into the introduction. S2: Without it being a clear question-there is no question mark there.
Like in excerpt 8, in many cases, the students discussed the variation between explicit implicit formulated RQ(s), and they expressed a clear preference for the former. They also identified that the RQ(s) were usually answered by problematizing findings or raising more questions. They seemed to accept this gradually over time.
Methodological issues were discussed in relatively long sequences compared with other parts of the genre discussions. Occasionally, the groups only identified the method and argument listed for the choice, but in most cases, they engaged in longer, rather critical discussion; this is illustrated in excerpt 9, where they discuss an article where the informants were teachers, public health nurses, and social workers, and the topic was interagency work.
Excerpt 9: The whole method section starts with him explaining that the study is based on qualitative interviews. It also says that it is of 58 people, teachers, many agencies, where I thought it was a bit unclear. Are there 58 teachers and an unknown number from other agencies or 58 in total? The interviews, they were open and lasted for 1 hour. But the interviewer does not mention whether it is a group interview or one-on-one. So, that was unclear. How the method is presented is clear, quite straightforward in the article. And sources, there are references to sources along the way-really only two, but it increases credibility […]. She has also interviewed from other agencies. You get a lot of different perspectives. Opinions and experiences are fine. She shows that she has put herself properly into it. I think that in this article, she has chosen a good method. I do not think observation would have been so rewarding. Surveys would have given narrow data. You do not get the experience and things like that, so I think it was a good method here with interviews. But she says WE, then we wonder who are WE? It's one author…?
This excerpt exemplifies common discussions on the fruitfulness of the selected scientific method, number and selection of informants, referencing of methodological literature, possible alternative methods versus trustworthiness of the findings, and the need for transparency and thorough description of methodological issues. Validity and trustworthiness were discussed many times, and the selection of method and number and selection of informants were considered (cf. excerpt 9). Other issues raised for validity were reflected in discussions on the selection of and use of theory in the analysis (excerpt 10), provision of enough context information in the article (excerpt 11), frequency and quality of referencing (excerpt 12), transparency in presenting the analytical strategy (excerpt 13), and how the authors' voices are present in the texts (excerpt 14).
Excerpt 10: […] She's good at using theory that confirms her findings. So, I am not sure, she is good at finding theory that says the same thing as she does...

Excerpt 11:
What I think is good is that she writes that the students' social background is complex, so there are parents giving good advice, parents with good education, parents with less money and less education, and everything is represented in a school class.

Excerpt 12:
She writes sensibly about the differences between the genders and the schools. It makes a lot of sense, but it would have been fine if she had added some references.
Excerpt 13: In the methods section, they write how they have proceeded with the analysis, of the findings. And that they have grouped and edited the interviews and how they have done it.
Excerpt 14: She appears to be biased…because she does not have the research question in the introduction. She does not seem objective but just wants to justify her case.
Research ethics was the topic least discussed in the group sessions when it came to genre. The one issue students broached the most was anonymity of the informants, as excerpt 15 shows.
Excerpt 15: S1: But did you notice that five small counties are identified by their real names? Was there anyone who reacted to saying which municipalities here, when they also say that it is child welfare employees, and she identifies the informants. Did you think about it in relation to ethics and anonymization? S2: When you say that now, I think yes.

3
Finally, the students discussed textual structures in the seminars. They became familiar with the relatively stable structure of an empirically based article within the field of education and gradually developed specific expectations concerning the content and order of the conventional sections (introduction, earlier research, theory, methodology, findings, discussion, and conclusion). Excerpts 1, 3, and 7 show that students expect accounts and discussions of concepts and RQ, as well as to explore whether and how the RQ are functioning as a golden tread throughout the article. Excerpts 2, 4, and 9 show the students' expectations for the method section of an article, while the discussion in excerpt 16 shows how they identify and make sense of the role of the abstract.
Excerpt 16: S1: You get the whole content of the text in the summary. They describe how they are going to explore-the structure, theories, what methods are used, observations. S2: Do you think that is a good way? S3: Here, I think it's okay. In a way, you get an overview of what is in the article then. When we become teachers, we probably do not read all the articles, but you can read the summary first and consider whether you want to continue; in that way, it is good. S1: Develop your own thoughts too. As long as they do not reveal results, then it's okay. S4: You get to see if it [the summary] arouses your interest. Is there something you want to find out? Instead of reading the whole thing; it might be a waste of time.
Finally, we find that in their search for focal textual structures, students pointed to how some articles are divided into smaller sections with headings and sub-headings and how that makes the text more reader friendly.
Excerpt 17: There were very nice subtitles. That made it easy to follow the structure of the article. It made it easier to return to some of the sections.
Our analysis in this section shows that the template and group sessions provided opportunities for the students to scrutinize various versions of academic text structures and dispositions, decide how to best formulate RQs, discuss a vast range of methodological issues, and assess validity of different research designs and arguments.

Contextualization, complexity, and meaning making
Moving from academic literacy as a set of cognitive skills and disciplinary categories, the focus in this section of the analysis is on academic reading as a dialogical linguistic practice in context. The questions under the headings "findings" and "relevance" in the template were meant to activate discussion on three such issues: discussions on the validity of findings in different contexts and implications for professional practice, discussions that revealed complexity and nuances for both professional and epistemological meaning making, and finally, discussion on how new knowledge challenges and changes students' conceptions and possibly their future teaching practice and teacher identity. The topics are not mutually exclusive, and the examples we discuss overlap in nature and content.
In excerpt 18, the students are discussing the relevance of the findings in the article about interagency (cf. excerpt 9). They bring up the societal relevance of the findings in the article beyond the professional communities involved (teachers, public health nurses, and social workers) and conclude the discussion by agreeing on the societal relevance of the findings. Parents and media are mentioned as focal audiences.
Excerpt 18: S1: It is about changing the view of children's welfare agency. S2: You have to get rid of the fear of losing your child as soon as they [children's welfare agency] are involved. It is an aid agency. They [the authorities] could create something new, an agency that takes softer cases, because they [children's welfare agency] are scary. One possibility is having a contact person in the agency.
There are several schools that do not know the name of anyone in the agency. At a parent meeting, you can bring someone from the agency and give information about what they do. S3: Yes, give a better view of them. This article highlights the media as well. We hear about terrible cases in the media. There are the worst things then. I believe in dialogue and public enlightenment.
Common misconceptions regarding the role of social workers concerning troubled families and children function as a driver for the various arguments the students broach in this excerpt.
Both professional and epistemological types of meaning making are salient across our dataset but not as prominent as discussions concerning cognitive skills and disciplinary categories. Excerpt 19 is from a discussion on an article describing how pupils experience the transition between lower and upper secondary school.
Excerpt 19: S1: Maybe it's new for many? Yes, I believe that. But if it does not lead to direct changes in the teaching systems, then it can at least lead to further research. Which then can create new… S2: And at least awareness, that one can perhaps understand slightly better why boys and girls do as they do in the transition between primary and secondary school. That it may be a big transition for them, and some struggle more than others... We need more understanding of this. S3: Yes. S1: And be aware that it is actually a change; many people think that the change or transition can be difficult, but others cope well. It is individual, so it is difficult in a way, to facilitate and make everyone happy. But as I said, you can be more aware as a teacher or future teacher.
The notion of the implications of the study expands as the discussion proceeds. It starts with a comment on the relevance and what the findings potentially could be used for (making changes in the educational system). Gender differences in experiences are then broached. Lastly, the variation among the student population and the need for teachers to have knowledge about transitions between school levels are pointed out.
In excerpt 20, the students question the nature and justification of the knowledge that they are left with toward the end of the discussion. The concept of "professional love" is being discussed and related to the article on professional ethics (cf. excerpt 5), where the author argues that teachers need to show genuine love for their students.
Excerpt 20: S1: Can you feel real love for students as a teacher? The article says it cannot just be based on spontaneous utterance […] sometimes one has to think. So, I do not think it was quite apt… S2: It will be a bit like that… Someone needs more love… Can't you have tough love? Not everyone needs that gentle part… S3: It will be very… There is no template for love, you do not have a framework then. Then one must consider all the time. Who will set boundaries then? There are many ways to show it, but it's up to you.
The first student questions the possibility and validity of the author's arguments about the need for teachers to show genuine love for all students. She also questions the argument that love has to be shown through spontaneous utterances (Lögstrup), instead arguing for the need for deeper reflections. The second student expands the discussion by supporting a more adaptive form for professional love based on specific pupils' needs. The third opens up the conception of love even further and argues that professional love must be contextually decided and defined.
The final excerpt discussed in this analysis illuminates how the work with the template in the reading seminars might challenge students' conceptions and practices. The article the students discussed claims that how teachers characterize the students in their everyday practice may influence the deeper conception of the students' abilities and identity. In excerpt 21, the students are discussing a much-used concept among teachers on students performing at the lower level of the (grade) scale-namely, "weak students." Excerpt 21: S1: I think how teachers talk about pupils is what the whole text is about. How they use the language, and how this might be problematic. S2: I think good and bad, one should not call a pupil weak, because then other teachers think that the pupil is weak in their subject as well. S1: That's what's the problem is. I remember Maja, a pupil, saying that everyone said she is so good. It became negative for her even though it was meant positively. If you use it like that, it becomes negative, other teachers also think so. S3: That everyone [teachers] talks about a pupil in the same way, saying he is a weak pupil. S4: Henrik is weak. S3: Then you do not go to class and think that he is a good pupil. S2: I heard such comments in my placement practice and when I was a substitute teacher. They said he was weak, but I found out he could do more than me. I realized he wasn't weak. S4: You do not have to struggle in all subjects either.
They emphasized how concepts and language used in teaching practice (even when not in the proximity of the student) may influence teacher students' view of and actions toward pupils, and ultimately, the pupils' identity and self-image.
The third part of our analysis shows that the "contextual practice" we aimed for was the most challenging for the students, and the issue that the students expended the least time and effort on. However, even though we do not have particularly studied changes over time, we find tendencies for an increased ability to initiate and partake in discussions, bringing up complexity and nuances and contextualization of the findings, over the 3-year course in which students partook in the seminars.

Concluding discussion
In this article, we explored whether and in what ways academic reading seminars foster a learning environment for developing academic literacies and whether the reading seminars provide a learning environment for developing academic and professional learning and engagement. Academic reading is the basis for almost all literacy activities in HE, and therefore, it deserves particular attention both from educators and researchers.
Our empirical analysis showed that the overall use of our template facilitated dialogical discussions, developing a set of cognitive skills and disciplinary categories rather well when used in a social setting. However, we observed a tendency for students to take shortcuts when scrutinizing unfamiliar concepts. This shows that ideas from earlier research emphasizing functional skills-such as critical reading, eliciting meaning from texts, decoding symbols, and the application of syntactic and semantic knowledge (c.f. Gorzycki et al., 2016;Wilson, 2016), as well as developing knowledge about the genre (c.f. Francis & Hallam, 2000) and disciplinary knowledge (c.f. Bharuthram & Clarence, 2015)-might be transferred to social learning settings.
The most challenging part of designing a reading practice in the academic literacies tradition was fostering a dialogical environment for discussing the validity of findings across different contexts and providing for discussions encompassing complexity and nuances in meaning making. We found traces of such discussions in all the transcripts; however, many of these examples were in a premature stage. Still, we conclude that an increased ability evolved among the students, allowing them to initiate and partake in such discussion over the 3-year program. We must remember that the students in this study were undergraduates, but still, we need to further develop the template, especially when it comes to the common meaning making across contexts and possible changes for future teaching practice. Even more importantly, changes in future teaching practices cannot be studied in full based on the present research design; they can only be glimpsed through passages in the student dialogues leading to professional and epistemological meaning making.
In line with Maguire et al. (2020) and Miller and Merdian (2020), we argued that facilitation of academic reading as a social activity engages students in explicit learning on how to understand and communicate about academic texts. When we provided the students with collaborative work in social space, they exhibited a developmental sequence of learning (c.f. Cowley-Haselden, 2020;Nguyen & Henderson, 2020), as well as extended confidence in academic reading (c.f. Kimberley & Thursby, 2020).
Overall, the academic literacies model (Lillis & Scott, 2007) prescribes a challenging thesis to be developed as an empirical entity for academic reading. Our initial template seems to a greater extent provide for an academic skills and socialization understanding than an academic literacies understanding. We see that in future redesign of the template and the organization of reading seminars, we need to take that into account. Hence, this study has inspired us to further develop the template for reading seminars in several ways (see Appendix 2 for an example). First, we have developed templates accommodating several genres of academic texts, such as book chapters, theoretical texts, and literature reviews. Second, we included a question requiring students to discuss and define unfamiliar concepts. Third, based on the findings showing that the template provides less scaffolding for the dialogical environment to discuss the validity of findings across different contexts and nuances in the texts, we reorganized the template's structure for a stronger focus on the 1 3 contextual nature of knowledge, epistemological meaning making, and academic texts as a source for developing identity and agency. Fourth, following Tomasek (2009), we included some questions or prompts to make connections and applications across courses and practices, approach challenging assumptions, and take a different point of view in the group discussions. Finally, exploring and inventing creative practices concerning scaffolding academic reading in HE is a field with wide potential, and the academic reading seminar we have designed and developed over the years and will continue to develop certainly lies within this scope of scaffolding a creative academic literacies practice.

Conflict of interest The authors declare no competing interests.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.