1. Introduction and selection protocol

As in the two previous reviews of dissertation research in Germany published in Language Teaching in 2011 (Behrent, Doff, Marx, & Ziegler, Reference Behrent, Doff, Marx and Ziegler2011) and 2016 (Doff, Königs, Marx, & Schädlich, Reference Doff, Königs, Marx and Schädlich2016), we return to second language acquisition (SLA) research in a language community which is often overlooked by the international SLA research community. This is due to various factors, including that German is not read by a wide range of scholars, despite being the most commonly spoken first language (L1) of Europe (Eurobarometer, 2012, p. 10). Moreover, German dissertations tend to be published in book format instead of online, making them, in today's digital market, less accessible to overseas readers (especially since dissertation publications are often quite costly and run upwards of 100$ in price). Fortunately, many dissertations are summarized – some in English – in easily accessible journal articles.

In common with the two previous reviews, the selection process for German dissertations for this article has required special consideration due to the specific circumstances of German SLA research: Ph.D. programmes per se are still quite rare in Germany, which means that research may not appear in any one comprehensive list, and SLA research itself tends not to take place within applied linguistics departments, but rather within – or in association with – individual language departments. Furthermore, because there is no national catalogue of dissertation research, some relevant dissertations might have been missed in the collection process. In order to counteract this problem, the composition of the authoring team reflects a wide range of disciplines within SLA and languages.

Some specifically German factors did assist in the search process, however. Since all dissertations must be published before a doctoral degree is awarded, gaining access to dissertations was straightforward. Furthermore, Deutsche Gesellschaft für Fremdsprachenforschung (DGFF), ‘German Society for Second Language Research’ keeps an ongoing (albeit incomplete) record of doctoral and post-doctoral work published in Germany in the form of the Klippel-Sauer-Liste,Footnote ¹ ‘Klippel-Sauer-List’. As happened for the previous reviews, this list was supplemented through intensive examination of German publishing catalogues as well as online dissertation listings of national libraries.

It should be noted that this article includes only research published in German since we aim to make this research more accessible to an international community. We thus did not include the small but ever-increasing number of dissertations being published in English, even if they were carried out in Germany by German Ph.D. students studying under German professors. However, we do expect the growing international trend to publish in English will continue to influence publication strategy in Germany. Although it is the central aim of this review to report, but not critically reflect on such trends in dissertation research, we would ask readers to consider how publishing primarily in English reflects on their own attitudes towards multilingualism and the acceptance of English dominating the global linguistic landscape.

This collection of dissertations picks up where the previous review finished, and thus looks at research published between 2014 and 2018. The thematic grouping revealed five ‘hot topics’ in German SLA research during this five-year period, some of which had already been discussed in the 2016 review, which covered the years 2009–2013. Out of 103 dissertations found for this time period, 44 could be sorted into these five categories; the remaining 59 were represented in six much smaller categories and two categories which dealt with very general concerns. As in the previous reviews, in order to best portray the breadth of research, we then chose a smaller, select group of exemplary dissertations from each category for more in-depth discussion, taking care to focus on publications that met sufficiently high scientific standards, and that were judged to be either especially representative of the category or especially innovative. This reduced the total number of dissertations discussed here to 16.

2. Current methods and methodology in dissertation research

Although the authoring team decided to focus on content as opposed to methodological criteria, we have included a short overview of typical research methods and research paradigms that feature prominently in German SLA research. Due to various historical developments in research programmes in the humanities in Germany, research – especially larger projects, and projects completed by new researchers such as doctoral candidates – tends to be qualitative in nature. This is reflected not only in the present and past review articles for Germany (in the dissertations discussed in this article, almost all had a primarily qualitative focus) but also in the difficulties that we encountered in describing certain aspects of research paradigms common in Germany – English translations for common concepts were often not readily available. Since we assume that many readers may not be familiar with the FST and Inhaltsanalyse, two approaches that frequently recur in this article, we first introduce them and their implications, especially with regard to issues of participants, instruments and data analysis.

A large number of current dissertation studies investigate how different participants involved in language learning, language teaching and language use view their learning processes, their languages and their own language use. The information derived from such studies may assist in better understanding how different perspectives within learning affect language learning and use. One very common research paradigm in such projects is the Forschungsprogramm Subjektive Theorien (FST) (‘Research Programme for Subjective Theories’) by Groeben and Scheele (Reference Groeben, Scheele and Mey2010). It takes an explorative–interpretative position in gathering and analysing data and aims for the analysis of ‘rich data’, that is, data that is as extensive as possible; data analysis is generally performed until ‘theoretical saturation’ is achieved. Because of this, elicited data tends to stem from methods which focus on relatively few research participants (‘informants’) who are interviewed and/or observed regarding specific aspects of the learning or teaching situation. This inherently leads to restrictions on research topics, since interview studies are much more difficult to carry out with, for example, younger participants (lower degree of self-reflection of the informants), participants for whom no common language can be found (new immigrants who are beginning learners of German with heterogeneous LI and FL learning experiences), or participants for whom legal restrictions make data collection difficult even in research settings where data is anonymous (in many states in Germany, it is close to impossible to gather research data from students in school settings). Thus, many informants in current qualitative research are acting or pre-service teachers (i.e. university students studying education).

Qualitative research – and the FST in particular – aims to view research informants as autonomous actors in language and social situations, who make decisions and act on them with intentionality, reflection and (potential) rationality, and using a range of communicative skills. Central to the FST is the belief that individuals develop subjective theories in order to explain the world (and their own position within it) to develop informed predictions and make decisions. Since such hypotheses are not easy to validate through non-subjective methods, interviews remain the most common method to gather data.

Both the FST and other qualitative–explorative paradigms require special methods of data analysis (and validation). In SLA research in Germany, this is presently almost always achieved through Qualitative Inhaltsanalyse (‘qualitative content analysis’), most often associated with Mayring (Reference Mayring2015) and Schreier (Reference Schreier2012). This aims first to analyse both content of subjective information provided by informants and then to group the information according to specific, recurring themes or topics. For this, one of two approaches can be used, although the first, structuring, is the most common. Structuring combines a recursive deductive (theory-based) and an inductive (data-based) approach to data analysis (for example, when certain themes appear within the first interview analyses, these themes are searched for in further analyses). In the summarizing approach, analysis follows a purely inductive (data-based) approach. Considerations of validity become especially relevant and different techniques have been developed in an attempt to achieve intersubjective validity.

While these two approaches to research are often justified, it does seem that in some contexts the decision to invoke for example qualitative content analysis is based less on theoretical perspectives or the central research question, and more on the fact that certain approaches have established themselves as a type of ‘normed methodology arsenal’. It should thus be cautioned that – as in all research projects – a careful consideration of research paradigm, elicited data and data analysis should, first and foremost, derive from the central research question and not primarily from trends within certain communities.

3. Topics of special interest in the German SLA research community

In comparison with previous reviews, the range of topics addressed in dissertation studies appears to have broadened, and it was no small feat to narrow it down to five topics which appear to have special relevance in Germany. Some topics that did not make this list (but certainly could have) were (1) digital media in foreign language (FL) education, (2) grammar teaching and learning, (3) language development and inclusion of new immigrants, (4) autonomous language learning, (5) inclusive FL teaching concepts, and (6) historical research on FL teaching. All these topics were represented by three to six dissertations and some (such as digital media) were identified as particularly relevant in the 2011 and 2016 review articles.

For this review, we chose the topics of (1) learning and teaching about culture in the language classroom (by far the largest category with twelve dissertation studies), (2) literary texts in L2 education (eight dissertations), (3) bilingual models of learning and teaching (nine dissertations), (4) plurilingual teaching and learning (six dissertations), and (5) language teacher education and teacher training (six dissertations). We conclude the discussion with a short section on emerging ‘hot topics’, that is, topics we expect to gain prominence in doctoral publications in the next few years.

4. Learning and teaching about culture in the language classroom

Research into both fostering and measuring intercultural competence continues to be a major focus of doctoral research in Germany (see the last review article for Germany (Doff et al., Reference Doff, Königs, Marx and Schädlich2016)). Characteristic of the area of intercultural competence is its interdisciplinary focus, extending to the fields of history, cultural studies and psychology, as well as political, social and educational sciences. Research in this area is thus quite versatile, with studies focusing on teaching resources, learning environments, intercultural encounters, and perceptions of Germany among international students learning German as an FL (see e.g. Grupp, Reference Grupp2014; Fellmann, Reference Fellmann2015; Sato-Prinz, Reference Sato-Prinz2017). Here, we focus on four studies which critically investigate the concept of intercultural competence. It is noteworthy that all of these studies directly or indirectly address Byram's (Reference Byram, Bredella and Delanoy1997) model of intercultural communicative competence (ICC); however, there seems to be a need to adapt or amend it as all four take a critical stance. While Plikat (Reference Plikat2017) questions and rejects the model, Guttack (Reference Guttack2016) reverts to other theories, and both Brunsmeier (Reference Brunsmeier2016) and Stahlberg (2016) adapt the model to their particular research contexts. Methodologically, most studies follow an empirical approach with several studies, such as Guttack (Reference Guttack2016) and Stahlberg (2016), implementing qualitative content analysis based on Schreier (Reference Schreier2012) and Mayring (Reference Mayring2015). The notable exception to this is Plikat (Reference Plikat2017), who makes a case for theoretical research and highlights the limitations of empirical research in the field of learning and teaching about culture in the language classroom.

In his theoretical study, Plikat (Reference Plikat2017) critically investigates current approaches to intercultural FL education. Drawing on theories from cultural studies, he analyses the general discourse on intercultural learning in the field of Fremdsprachendidaktik (‘foreign language education’) and identifies two major criticisms of ‘interculturality’ (Plikat, Reference Plikat2017, pp. 31–36). First, approaches to FL education are often based on notions of culture which have been rejected by cultural studies. Second, educational approaches often advise a cultural relativist standpoint (Plikat, Reference Plikat2017, p. 193), rather than facilitating a reflective discourse on the dilemma of universalism and cultural relativism. This is particularly relevant in situations where diverging values and norms lead to conflicts for which no consensual solution can be found.

Plikat proceeds to analyse four influential approaches to ICC in the German context. These are Didaktik des Fremdverstehens (approximately: ‘understanding of the self and the other’, Bredella & Christ, Reference Bredella and Christ1995), Kramsch's (Reference Kramsch and Vestergaard1999) notion of Thirdness, Kramsch's (Reference Kramsch2006) notion of Symbolic Competence, and Byram's (Reference Byram1997) model of ICC. His comprehensive theoretical analysis of all four approaches shows that none of them is able to overcome the aforementioned deficits of intercultural FL education. Plikat (Reference Plikat2017, pp. 193–194) concludes by developing the concept of Fremdsprachliche Diskursbewusstheit (‘foreign language discourse awareness’, Plikat, Reference Plikat2017, p. 299) as an alternative FL learning objective to ICC. He argues that this concept allows for the development of an appropriate understanding of culture in the light of cultural theory as well as a reflective discourse on the dilemma of universalism and cultural relativism in the context of FL education. In his concluding remarks, he addresses the challenge of empirical research in the field of intercultural learning, since interculturality is a result of discourse rather than an attempt to objectively capture reality.

In her qualitative study, Brunsmeier (Reference Brunsmeier2016) integrates two major recent research trends – task-based language teaching (Ellis, Reference Ellis2003) and Byram's (Reference Byram, Bredella and Delanoy1997) model of ICC – and considers them in the under-researched context of FL education in German primary schools. The primary focus is on which task characteristics are necessary for successful ICC development among primary language learners (Brunsmeier, Reference Brunsmeier2016, p. 143). To this end, Brunsmeier first develops ‘Can Do’ statements for ICC for primary students and integrates them with task descriptors as well as model tasks for this particular aspect of ICC (Brunsmeier, Reference Brunsmeier2016, pp. 117–141). Second, she investigates the implementation of the model ICC tasks in two participatory action research cycles by analysing data collected through observation, interviews, surveys, and learners’ spoken and written production. In the first cycle, the model tasks were used by four English as a foreign language (EFL) teachers in four primary classes. Based on her analysis of both the teaching process (task-in-process) and the task characteristics (task-as-workplan), Brunsmeier revised the model tasks, which were then re-implemented in one EFL class. Regarding theory development, Brunsmeier extends Byram's model of ICC and suggests the term Reflektierte Interkulturelle Aufmerksamkeit (‘reflective intercultural attention’, Brunsmeier, Reference Brunsmeier2016, pp. 326–327) as an age-appropriate alternative concept for primary language education. It adapts Byram's dimensions of attitude, knowledge and skills to primary education, but discards the dimension of critical cultural awareness for this particular setting. Brunsmeier's study is thus especially interesting in the light of the companion volume to the Common European Framework for Reference (CEFR) (Council of Europe, 2018), specifically regarding its discussion of mediation as well as young learner scales. Finally, it provides evidence of the central role of primary school teachers in facilitating intercultural learning processes among students (see also Brunsmeier, Reference Brunsmeier2017).

In her qualitative interview study, Guttack (Reference Guttack2016) addresses intercultural student exchanges with a special focus on international work placements. Empirical research in this area is rather sparse, probably due to the challenges of accessing it. It is of note that Guttack's study does not use Byram's (Reference Byram, Bredella and Delanoy1997) influential model of ICC, but integrates a set of other models and their adaptations (e.g. Altmayer, Reference Altmayer, Hu and Byram2009; Bechtel, Reference Bechtel, Hu and Byram2009; Eberhardt, Reference Eberhardt2013). Guttack focuses on the individual learning paths and acquisition of ICC of 16 secondary school learners of Spanish as an FL who completed a two-week vocational internship in various Spanish companies between 2007 and 2012. She reports her findings both in detailed case studies as well as a cross-case synopsis (Guttack, Reference Guttack2016, pp. 175–316). In the interviews, learners gave an account of their perceptions, strategies and behaviour during the internship and while staying with Spanish host families. Based on her analysis, Guttack (Reference Guttack2016, pp. 336–345) identifies two different learner types who used the intercultural encounter of the work placement very differently to develop their ICC. While the introvert type used the intercultural situation to focus on the self, the extrovert type concentrated on the context rather than the self.

Guttack's study provides evidence that learners use highly individual paths to deal with and explore intercultural situations and develop their ICC. Guttack considers this aspect to be fundamentally important, especially in the context of intercultural education, as it calls for in-depth comparisons of differences and similarities within a group of learners. Her study consistently makes a special case for individual learner perspectives (e.g. Guttack, Reference Guttack2016, p. 124) and therefore supports the conclusions of the further studies reported in this section that learning about culture and learning in intercultural situations must be understood to be on a continuum between the individual and the collective. Based on her findings, Guttack argues in favour of follow-up activities to such international internships in order to fully exploit this opportunity for developing learners’ ICC.

Finally, in her qualitative study, Stahlberg (2016) aims to reconstruct the ICC of 21 international students at a German university. Again, Byram's (Reference Byram1997) model of ICC is used, but here, it is adapted with reference to further theories of interculturality by adding the dimensions ‘perceptive’ and ‘reflective’ to the domains of attitudes, skills and knowledge. Stahlberg (Reference Stahlberg2016, pp. 162–163) makes a strong case for a qualitative–reconstructive approach to researching ICC rather than quantitative approaches that rate observable behaviour or assess ICC with surveys. Her main argument against the latter is her doubt that ICC can be scaled and measured as suggested, e.g. through the respective scales in the Common European Framework for Reference (CEFR), the companion volume to the CEFR or the Framework of Reference for Pluralistic Approaches (FRePa) (Candelier et al., Reference Candelier, Camilleri-Grima, Castellotti, Pietro, Lörincz, Meißner and Schröder-Sura2012). In particular, she criticises research designs which equate a high ICC with participants fulfilling a multitude of the sub-domains in Byram's model, which she sees as partially contradictory. For example, a high ‘readiness to experience the different stages of adaptation to and interaction with another culture’ (Byram, Reference Byram, Bredella and Delanoy1999, p. 366) could conflict with an individual's own positioning. Thus, Stahlberg's goal is to describe the participants’ ICC rather than evaluate it. She aims to explore how participants perceive intercultural contexts, how they reflect on them, how they act and which cognitive and affective aspects play a role (Stahlberg, Reference Stahlberg2016, p. 23). To this end, she conducted interviews with international students attending German as an FL classes and their teachers at a university in Germany in 2011. Data was analysed both across cases using qualitative content as well as in a case study reconstructing the ICC of one student. Findings show that all five dimensions of ICC – affective, perceptive, cognitive, pragmatic–communicative and reflective – play an important role (Stahlberg, Reference Stahlberg2016, pp. 377–379). Furthermore, the study shows that the reflective dimension is a central component and impacts considerably on the other four. For example, it was only through reflection that participants achieved a deeper and critical understanding of cultures or knowledge about cultures. As a final point, different dimensions of ICC seem to be interdependent. For example, only the recognition of both cultural differences and similarities allowed for appropriate and effective interactions.

In conclusion, it is striking that transculturality has not yet received more attention in empirical research as it has been a major issue in the theoretical discourse on German culture and language education (e.g. Delanoy, Reference Delanoy, Delanoy and Volkmann2006; Doff & Schulze-Engler, Reference Doff and Schulze-Engler2011; Volkmann, Reference Volkmann, Delanoy, Eisenmann and Matz2015). This is probably at least in part due to the difficulty in operationalising some aspects of inter- or transcultural competence (see Plikat, Reference Plikat2017). However, since interest in culture and language education remains strong, and research in the field is necessary both on a theoretical and empirical level as well as in practical classroom teaching, it is to be expected that future studies will tackle these challenges. This is especially important in light of current social and political developments in Germany and beyond, where nationalist, xenophobic and racist stances have been gaining strength. It is to be hoped that research findings from SLA research can contribute to the strengthening of societies based on democratic values while appreciating diversity and heterogeneity.

5. Literary texts in L2 education

The use of literature in L2 education – mostly coming from the tradition of humanities and literary studies – is a research field where Ph.D. dissertations are primarily theoretical and conceptual. However, the recent fundamental shift in educational research towards empiricism can also be observed and was critical in the selection of the studies to be presented in this review. The three dissertations chosen as representative of literary texts in L2 education differ in terms of the questions asked and their methodological approaches yet give insights into current research interests and reveal elements of leitmotif. These include, for example, a pronounced interest in Humboldts's ideal of education (Bildung), as well as personality development and the role of emotions in language learning.

The dissertations deal with the questions of what is learned about (and through) literature in FL classes and how to study this empirically. Here, research approaches can be identified which either attempt to model competencies that are difficult to measure, or that show that such competences defy operationalisation.

Steininger (Reference Steininger2014) aims to operationalise literary competence. His study is innovative in that it focuses on younger learners (aged 10 to 15 years). Steininger investigates the added value of literary reading, compared to more restrictive reading literacy models (reading as a way to achieve certain goals) common to international empirical studies such as the Programme for International Student Assessment (PISA) or the Progress in International Reading Literacy Study (PIRLS). His explorative study aims to provide ‘thick descriptions’ (Geertz, Reference Geertz1973, pp. 3–4), generated using grounded theory methodology (see Strauss, Reference Strauss1994). After conducting a systematic overview of competence models for FL teaching in Germany (using the CEFR and national education standards, Ständige Konferenz, 2003), Steiniger carried out eight case studies of literature lessons in English classes, referring to video data supplemented by teacher questionnaires, analyses of students' products and retrospective teacher interviews. The video data are supplemented by teacher questionnaires, analyses of students’ products and retrospective teacher interviews. By constantly comparing the empirical data with the normative competence models, he offers a basis for fine-tuning reading models. Furthermore, Steiniger argues for more differentiation on the conceptual level than for scaling of literary competence. Since literary competence is, from his perspective, a holistic construct (in contrast to the reductionist reading competence models of CEFR and educational standards), it can be seen as a part of general communicative competence.

Steininger's explorative study thus aims to differentiate between existing competence models. Other Ph.D. theses, such as that published by Bracker (Reference Bracker2015), instead critically question the nexus of literary reading experience and the competence paradigm. Bracker's study aims to reconstruct student conversations about postcolonial literary text in English class. Her research interest is not in modelling or measuring literary–aesthetic competences, but rather in understanding learners' interactions. Such experiences are situated in a ‘third space’ (Bhabha, Reference Bhabha1994, p. 37), where new insights emerge out of negotiation of meaning.

Methodologically anchored in reconstructive social research, Bracker uses the documentary method (Bohnsack, Nentwig-Gesemann, & Nohl, Reference Bohnsack, Nentwig-Gesemann and Nohl2007) to analyse the conversations in small groups of three to four learners from six English classes in upper secondary schools. She is interested in their ‘ways of dealing with the text’. Above all, the observed negotiation processes are about semantic and grammatical uncertainties that produce a common understanding of the text and which are the starting point for meaning-making. Bracker shows in four portraits of student conversations that underlying orientations are strongly convergent, for example in the perception of the text as determined by strangeness. Nevertheless, she also finds strong contrasts in the productivity of the conversations. For example, groups show different degrees of perspective-taking in their negotiations, as well as of reflection regarding both the appropriation and the deconstruction of unreflected stereotypes. The strength of the empirical approach lies in being able to show that the positive impacts of literature teaching claimed in many conceptual studies – such as changes of perspective or critical questioning – cannot always be observed in the recorded interactions.

Our third Ph.D. dissertation applies cultural studies research to FL literature education research. König (Reference König2016) deals with the question of gender as a relevant notion for FL learning. The study first discusses the theoretical and conceptual significance of gender as a category in FL learning and assumes that learning with literary texts makes a special contribution to gender awareness and reflection. König refers to new approaches in culture pedagogy, based on theories of ‘culture as text’ (Bachmann-Medick, Reference Bachmann-Medick2010), and notions such as Symbolic Competence (Kramsch, Reference Kramsch2006) or Discourse Pedagogy (Hallet, Reference Hallet and Legutke2008). Although the scientific strength of König's study lies in its theoretical discussion, it also includes a practical element; a corpus of literary texts is analysed from a gender perspective and discussed with the help of an English (EFL) teacher and his students at a Gymnasium (university-stream secondary school). Aspects of structural inequality, performative constructions of gender (‘doing gender’) and (criticism of) heteronormativity are accentuated in the classroom units using methods that switch perspective, accentuate the performativity of cultural constructions and reflect on power relations in social interaction. The contribution of König's dissertation is, above all, that it provides reference points from cultural and gender studies for the selection of literary texts and their use in FL classrooms. König's conceptual model and analyses of literary texts (including films, pictures and music) provide a sound basis for follow-up empirical work.

In summary, the interest in using literary texts for SLA remains high and this has increasingly become an area of empirical research. However, much conceptual work needs to be done on both the function of literary texts in language acquisition and their operationalisation in empirical approaches, especially considering the development of new scales for ‘creative texts’ in the companion volume to the CEFR.

6. Bilingual models of learning and teaching

In the past decade, research on content and language integrated learning (CLIL) has become a stable field of German doctoral research. However, there has been a notable decrease in the number of studies dealing with this topic since the first overview of German doctoral research (Behrent et al., Reference Behrent, Doff, Marx and Ziegler2011). This mirrors the fact that CLIL programmes are now a well-established feature of the German school system, where an FL (mostly English) becomes the language of schooling for one or two subjects in an otherwise German-speaking school environment.

Between 2014 and 2018, all studies on CLIL have followed an empirical paradigm. They study not only the dominant L2, English, but also Italian (Streb, Reference Streb2016) and French (Deutsch, Reference Deutsch2016), and investigate maths (Verriere, Reference Verriere2014), science (Piesche, Reference Piesche2016), and history (Dallinger, Reference Dallinger2015). Two dissertations have emerged out of interdisciplinary projects closely associated with education studies (Dallinger, Reference Dallinger2015; Piesche, Reference Piesche2016).

Given this broad background, it is not surprising that the studies have different theoretical and methodological paradigms, and both small-scale studies with dense descriptions of case studies (Verriere, Reference Verriere2014; Deutsch, Reference Deutsch2016) as well as large-scale studies applying quantitative methodologies (Dallinger, Reference Dallinger2015; Piesche, Reference Piesche2016) are represented.

In the interdisciplinary field of learning and teaching, in terms of language and content, we can identify some ‘hot spots’. The effect of CLIL education on students’ motivation to learn is a consideration that links the studies by Verriere (Reference Verriere2014) and Piesche (Reference Piesche2016).

Verriere's (Reference Verriere2014) is a mixed-method case study and starts with the assumption that students can be grouped into different preference types: those with an interest in learning languages, and those with an interest in science learning. The study poses the questions: how do CLIL modules in maths affect the various learners’ motivation to learn English and maths respectively? If there are any effects, is CLIL a setting which can be instrumental in enhancing motivation for both groups? Students were grouped into three types, according to their self-reported affinity: high affinity for L2 English, high affinity for maths and low affinity for maths. Student questionnaires (N = 410) and qualitative interviews with teachers (N = 18) were then used to gain insights into how students and teachers perceive the CLIL maths setting. The results indicate that students with a low affinity for maths profit from CLIL settings based on reported levels of contentment in the maths classroom; girls especially reported higher motivation through CLIL. The students with higher affinity for maths did not report any loss of motivation through the CLIL settings, while the students with higher affinity for English showed ambivalent results. Thus, the study illustrates that maths is a suitable topic for bilingual modules, and that the use of L2 English as a means of communication does not necessarily increase the motivation for maths learning but (especially for girls) may have positive attitudinal effects.

Piesche (Reference Piesche2016) works on related questions, but using a quasi-experimental, randomised design that not only elicits students’ and teachers’ beliefs but also measures knowledge gain. In her cumulative dissertation (a format which is uncommon in traditional SLA contexts in Germany), comprising three articles in educational studies journals, she attempts to shed light on how CLIL education in science classrooms impacts on the gain of science-related competence, and how previous knowledge, gender and affinity for language or science affects this knowledge gain. Motivation is also investigated as a dependent variable. A battery of different tests is used, among them items validated in the Trends in International Mathematics and Science Study (TIMSS) (Kleickmann, Reference Kleickmann2008), an international project assessing students’ educational achievement in mathematics and science across different educational systems. An important finding in Piesche's study is that male students with a broad knowledge of science are affected negatively by CLIL, in that it reduces self-efficiency and increases anxiety. The same effect does not hold for female students.

Another cumulative dissertation, from the same context as Piesche's, is presented by Dallinger (Reference Dallinger2015). It aimed to shed light on the question of whether the better achievement of history CLIL students really can be regarded as an effect of the learning environment, or whether it can be explained by a positive participant selection in CLIL classes. Because there were no studies that systematically measured the effects of CLIL both in terms of L2 and subject competence development over time, the study employed a longitudinal design.

Dallinger compares the educational family background, cognitive abilities, subject-specific and L2 learning motivation, general L2 English proficiency and listening comprehension, as well as history knowledge of 1,806 CLIL learners (Grade 8, i.e. 13–14 years of age) with non-CLIL learners. A fundamental achievement of this study is that it shows how previous research on CLIL in Germany tends to overestimate the effects of CLIL programmes on the development of competences both in L2 and subject learning. The higher achievement in L2 English by CLIL students can at least partly be explained by learner preconditions, of which motivation is one, before taking up CLIL. Through a multi-layer regression analysis of CLIL and non-CLIL learners’ achievement in L2 and history learning, Dallinger finds that CLIL education enhances receptive L2 skills, but that this effect has in the past been grossly over-estimated due to confounding variables. In measurements of productive skills (a shortened version of a C-test was used), the difference compared with regular L2 education is only marginally better. In the history knowledge test developed by the researcher, CLIL students attained similar scores to non-CLIL students. Further, Dallinger addresses the teachers’ perspective and elicits CLIL teachers’ self-perceived L2 use in their classrooms. She then compares this to their students’ (N = 703) knowledge gain in history and L2 development. The findings are somewhat contradictory; both (teacher-perceived) increased L2 input and L1 German input yield increased learning outcomes, while extended (teacher-perceived) student production has negative effects. The analysis presented here shows that CLIL (English/history) students typically are an elite group with cognitive and socio-economic advantages who are selected because they have a higher L2 proficiency than the students not selected for CLIL classes, and that participation in CLIL increases this difference even more.

A fourth study (Streb, Reference Streb2016) focuses on a special case of CLIL at primary level. Streb's four-year ethnographic case study of a bilingual German–Italian primary school takes a detailed look at the individual linguistic profiles of both pupils and teachers in a dual immersion context, allowing her to illustrate the large (and differential) impact classroom set-ups have on individual students’ learning opportunities and the development of their linguistic profiles. This setting is different from the CLIL settings investigated by Dallinger and Piesche in that it considers heritage language learners as well as monolingual German children, and therefore reflects heterogeneous learner profiles. The descriptions of individual cases can be lengthy, but they are an important counterbalance to more analytical and nomological approaches. The study illustrates well-known phenomena from bilingualism and heritage language research (e.g. dynamic quality of linguistic competence, dependence on opportunities of use, the pressure heritage languages are under in a dominant language environment, linguistic adaptation to language competences of the addressee). It also follows how former Italian-dominant children become German-dominant over time. An interesting result concerning bilingual education as a content-learning environment is that dual immersion is shown to have the potential to provide additional learning opportunities, e.g. by doubling content-focused phases through the use of Italian and German, respectively.

Finally, Deutsch (Reference Deutsch2016) focuses on the issue of plurilingualism and CLIL. CLIL is generally regarded as an educational tool to promote the development of plurilingual competences. Deutsch's aim was to find out how CLIL education (here, in French in an otherwise German-speaking educational setting) can actually be regarded as a valid tool for promoting plurilingualism by teachers and students, and whether school practice is mirroring plurilingual learning goals. In order to find answers to these questions, Deutsch elicits views on plurilingualism from 21 teachers and 75 students in four different schools. The teachers’ views were collected in group discussions, while the students filled in language portraits and questionnaires. Deutsch's results indicate that the plurilingualism concept advocated in the CEFR and FRePa is neither known to the teachers nor enacted in the classroom context. Thus, CLIL per se does not seem to contribute to the development of an individual plurilingual profile in the sense of European language policy automatically, but instead requires specific pedagogical activities if plurilingual learning goals are to be reached. Deutsch's research makes clear that although plurilingualism is an established concept in European language policy and teacher education, it still has a long way to go before it can be regarded as established in German schools.

7. Plurilingual teaching and learning

Six dissertations comprise our final category, which investigates teaching models involving more than merely the L2 or FL in focus. Five of them are empirical in nature and take a qualitative perspective. Four involve cross-language learning, either in consecutive models (learning a second FL, as well as a majority language and a first FL) or in simultaneous models (learning multiple languages in one language family through intercomprehension programmes). A fifth dissertation looks at the special role of English (almost always learned as the first FL in German schools) for the development of plurilingualism. The sixth dissertation is pedagogical and proposes a model for teaching. Although all these dissertations make a specific and interesting contribution to SLA research in Germany, only two will be discussed in further detail here: Vidgren's (Reference Vidgren2018) empirical project which investigated learners’ subjective theories when learning a tertiary language and which was chosen as highly representative for this category, and Dausend's (Reference Dausend2014) pedagogical project, which represents a more unusual approach.

The five empirical studies focus on learning and teaching in institutional settings (school or university), and are both qualitative and explorative–interpretative with the goal of gathering rich data to explain processes of language learning. All five base at least part of their research on survey data, using either questionnaires or – more commonly – interviews to investigate the primary research focus. As a result, all involve a relatively small number of participants. In three of the five studies, the research paradigm ‘Subjective Theories’ (FST) was chosen as an empirical background. All chose to validate the preliminary results using the Heidelberg structural technique in order to visualise subjective theories and consolidate them for acceptance by the research subjects. Finally, all studies analyse data with a qualitative content analysis approach.

Vidgren's analysis of Finnish high school children's subjective theories on learning multiple languages – especially German in addition to their first two foreign languages, English and Swedish, is representative of these studies. Out of 30 original participants, 16 were chosen for a first analysis and – following the principles of theoretical sampling in grounded theory – a total of eight were included in the final analyses once theoretical satiation was achieved.

The main interest of Vidgren's dissertation revolves around factors influencing the learning of German in addition to English and Swedish, and, especially, whether and how these languages (and the L1, Finnish) interact. Data was gathered using a questionnaire on which basis interviews were conducted. Follow-up structural technique interviews were then carried out two to four weeks after the first interviews in an attempt to validate conclusions drawn about subjective theories.

Vidgren's study is interesting because it sheds light on how high school students view learning multiple languages – their subjective theories on multiple language acquisition collaborate data from adult language learners but also among teachers. While students, for example, see possibilities for interlingual transfer and support, they almost exclusively restrict these perceptions to lexical units and mention very specific directions of interlingual influence. English and Swedish are viewed as a generally good support mechanism when using other FLs, and the third FL, German, may draw upon help from other FLs, both in receptive and productive language use (for example, when students try to guess a word while speaking German). However, they also interfere. The students believe that such interactions are more common when the languages involved are etymologically close to each other, when the language being spoken has not yet reached a high level of proficiency, or when the FL interfering has been learned to a higher level.

Dausend (Reference Dausend2014) takes a different approach to investigating language teaching outside of the constraints of the specific language classroom and learners involved. While her dissertation is not empirical, it provides important insights on language teaching and learning in primary school. She focuses on the conceptual description of different models of FL education, especially on whether and how FL teaching is included in (or seen as a part of) other subjects. The dissertation itself aims to fulfill two goals: first, to provide an exhaustive summary on the implementation of FL education at primary level and on relevant research on the various concepts investigated, and second, to develop a pedagogical concept for transcurricular language education.

The dissertation raises awareness of a number of difficulties in the German primary school system, attempts to explain different possibilities for addressing these problems, and develops guidelines for teaching. In order to address these concerns, it develops – based on research on different schooling models – goals for language teaching in primary grades, principles for skills-based teaching at this level, and guidelines for achieving these goals and principles. This is especially difficult considering that, despite a long tradition of FL education in Germany at primary level (and the fact that since 2004, all children in Germany start FL education at the age), there are no curricular guidelines for teaching FLs at primary schools.

Dausend turns to two successful concepts of teaching language at primary level (a third concept, reform pedagogy, is not included in the final model): CLIL and discourse competency-oriented teaching. An extensive literature review culminates in a transcurricular pedagogical model for teaching FLs at primary school, whereby languages (FLs, majority and heritage languages) and content are co-dependent classroom topics. The classes are determined not by the specific subject, but rather by transcurricular themes; for example, the topic ‘Me and my family’ combines the subjects German, social studies, maths, English and ethics/religion, while a topic such as ‘Living in Australia’ combines social studies, art, music and physical education as well as English. The dissertation thus stands out as a highly practice-oriented critique aiming not to test different concepts itself, but rather to present research, develop guidelines on the basis of this research, and then present a concept for transcurricular language teaching which can be implemented and tested at a later date.

8. Language teacher education and teacher training

As reported in 2016, the professionalisation of FL teachers remains an important research area in Ph.D. dissertations in Germany. This is not surprising, since any topic aimed at changing or improving FL teaching is intimately linked with either implicitly underlying or explicitly elaborated demands concerning teachers’ professional competence.

The professionalisation of teacher education has also received widespread media and political attention in recent years, as can be seen, for example, in a nationwide research programme funded by the Bundesministerium für Bildung und Forschung (BMBF), ‘Federal Ministry of Education and Research’ running since 2015 (BMBF, 2019).

Studies on professionalisation diverge considerably in research questions and methodology. While science, technology, engineering, maths (STEM) studies use validated competency models, linguistics and humanities remain what Blömeke et al. (Reference Blömeke, Bremerich-Vos, Haudeck, Kaiser, Lehmann, Nold and Willenberg2011) define as ‘low structured domains’. In general, research interests are directed towards reflexive competence and practical learning in (especially initial phases of) teacher education. Student teaching, internships and action research have been implemented in many university syllabi for education students. This reflects the belief that linking theory and practice has a positive impact on professionalisation processes. However, there is no compelling evidence supporting this assumption, as Hascher (Reference Hascher2012) was able to show in her literature review on practical phases in teacher education.

In their overview of recent research on FL teacher education, Legutke and Schart (Reference Legutke and Schart2016) show that the field of teaching English (TEFL) is the main interest in research and doctoral dissertations. Other languages often remain marginal, which is also reflected in the studies compiled for this review article: four of the six dissertations are devoted to English teacher education, one to Russian (Kurz, Reference Kurz2015) and one to Spanish (Valadez Vazquez, Reference Valadez Vazquez2014).

For this review, two Ph.D. dissertations were selected as representing a general shift in research methodology within the topic of language teacher professionalisation. While German research has an established tradition of working with retrospective data elicited through questionnaires or interviews and focusing on the emic perspective of teacherś professional identities or their subjective theories, recent studies are increasingly directed towards praxeological methodology helping to understand in-situ data, for example from videography.

In her Ph.D. thesis, Wipperfürth (Reference Wipperfürth2015) examines ‘professional language’ in collaborative teacher networks. She models her construct of professional language using educational and linguistic theories, integrates them with models of teacher knowledge (Shulman, Reference Shulman1986), and shows that teacher knowledge is mostly tacit knowledge going far beyond what teachers are able to verbalise explicitly. The approach of Professional Vision (Goodwin, Reference Goodwin1994) is used to investigate such knowledge in teachers’ practices. During workshops, English teachers collaboratively analysed videotaped lessons of their own teaching. The discussions were audiotaped, transcribed and thematically structured using qualitative content analysis. They were then analysed regarding professional language use and knowledge construction.

The analyses show that professional language ‘mediates’ between scientific and practical aspects of professionalisation because teachers verbalise tacit fragments of knowledge and make them accessible for others. However, it remains unclear whether collaborative analysis can actually lead to new understandings and changed practice or whether collaborative analysis reinforces experiential knowledge that also might refer to problematic beliefs. In this way, Wipperfürth's study offers a basis for follow-up studies.

Knorr (Reference Knorr2015) also attempts to understand the reflective practice of pre-service teachers. She is interested in the social practices of lesson planning, surprisingly a highly neglected area in teacher education research. Knorr worked with university students, observing them planning English lessons for a one-day internship in dyads. To analyse the students’ interactions, Knorr draws on a combination of conversation analysis and qualitative content analysis. This methodological decision has proven to be extremely fruitful: on the one hand, the combination of methods can help to find thematic issues of what is deemed relevant in lesson planning, and, on the other hand, it can show how the interactions work in the discursive process. The study thus answers the question of what pre-service teachers actually do when they plan lessons together.

The analysis shows that the most dominant category is ‘classroom activities’ while the category ‘teaching objectives’ is the least observable. Students therefore do not plan functionally in relation to a specific learning objective, but rather focus on sequences of isolated exercises.

Knorr concludes that collaborative planning helps explicate and expand one's own planning. At the same time, however, the analyses also reveal that the discussions often lead to quick agreement without the intended critical reflection taking place. The study affords a deeper understanding of planning processes for English lessons, while the topics and typical interactions chosen can help to sharpen the notion of ‘planning awareness’, which is an important consideration for academic teacher education.

In summary, research on teacher education in Germany is strongly oriented towards pre-service teacher experiences. Empirical studies focus on practical considerations such as internships, which are investigated using increasingly differentiated but still predominantly qualitative methods. Quantitative methods and a focus on later phases (career entry, continuing professional development) are clearly under-represented, and competency models for the specific area of FL teaching still remain a research desideratum.

9. Emerging hot topics in German SLA research

The five topics addressed in this article are ones where a good deal of research has been carried out by doctoral candidates. However, since we have only discussed dissertations published between 2014 and 2018, it is reasonable to assume that research projects that began in, for example, 2015 or later were not included in this article due to the length of time typically needed to finish a doctoral study in Germany (on average, five years) and the additional time lapse because of the publication process (often another one to two years). In order to counteract this selection problem, we turn to two hot topics that we view to be emergent in dissertation research, and which we expect to attract increased research attention in the coming years.

The past few years have seen a growing interest in the academic register for teaching German as an L1/L2. This is also reflected in a number of recent dissertation studies. Although we decided not to include these studies in this article – as they often focus less on language learning than on students growing up bilingually with heritage languages and educational success in immigration contexts – we would like to take the opportunity to mention a few studies concerning academic German.

While there is a tradition of research on academic language development in English-speaking areas, research is still rather new in the German context. Further, such research in Germany is presently dominated by non-SLA studies from the fields of German, education or the subject pedagogies, which led to the decision not to include it in the present overview. However, due to its importance for foreign language pedagogy and teacher education, especially regarding linguistically and culturally heterogeneous student groups, we believe SLA research on academic language will expand considerably in the future. Interesting issues range from conceptual ties to CLIL concepts, language awareness and multiple language curricula. Two exemplary studies in this area are Olson (Reference Olson2017) and Springob (Reference Springob2017).

Olson investigates L2 English reading in heterogeneous learner groups from a socio-cognitive perspective and looks into peer scaffolding techniques in cooperative reading of demanding L2 texts. In a mixed-method study, she re-analyses six cases of videographed student–student interactions during joint text comprehension from the ADEQUA study (Finkbeiner, Knieriem, Smasal, & Ludwig, Reference Finkbeiner, Knieriem, Smasal and Ludwig2012). Both qualitative content analyses and conversation analysis are carried out to gain insights into the joint construction processes and strategies. The analysis shows that collaborative reading leads to scaffolding strategy use and is also positively related to learning. A clear tendency emerges: scaffolding which is initiated by both students respectively leads to considerably more scaffolding instances than unidirectional scaffolding with one dominant peer. Interestingly, scaffolding activity does not correlate with proficiency level – the more L2-proficient peer is not necessarily the one who initiates learning opportunities for the less proficient one. Furthermore, learner dyads with high degrees of equality and mutuality use peer scaffolding most often. The study confirms that it is possible to teach peer strategies to enhance reading processes, and suggests increasing peer work opportunities in balanced dyads. Especially in terms of linguistically heterogeneous classrooms, this opens up a promising area for SLA classroom pedagogy and heterogeneous learner groups.

Heterogeneity in SLA teaching is also addressed by Springob (Reference Springob2017). He looks at inclusive learning contexts which involve students with a broad range of cognitive abilities, especially those with special needs. Inclusion of such students is an emerging hot topic for the German educational system as a whole, since legal decisions regarding inclusion now have to be enacted in the educational context. The study's contribution lies both in the thorough theoretical reflection of insights from general pedagogy, SLA and special needs education and in the integration of these insights into a framework for inclusive English teaching, which is then implemented in an action research setting. A learner group including students with special needs is followed over the period of 2.5 years, and individual cases are their individual development. The dissertation discusses the conditions in which inclusive learning is possible, and what is necessary for successful inclusive SLA settings. This is a valuable starting point for further research in inclusive L2 learning and teaching.

Another area which we would expect to emerge as a major topic in SLA doctoral research is German as a second language (GSL), Deutsch als Zweitsprache (DaZ), for recent immigrants. Similar to other European countries, Germany has experienced a large influx of refugees in recent years, peaking in 2015 and 2016. The media attention has led to a steep increase in the development of teaching and learning material, professional development programmes for teachers, and research funding. Three dissertations approach DaZ from different perspectives.

Decker-Ernst's (Reference Decker-Ernst2017) study is an eminent reference document with a focus on Vorbereitungsklassen (‘preparatory classes’), whose aim is to provide newly arrived immigrant students with the basic German skills to participate in regular school education. The study first provides a thorough overview of recent immigration to Germany and sets this in relation to educational policy decisions, including pedagogical frameworks of Vorbereitungsklassen. Second, a corpus of legal documents and curricula is systematically analysed in order to present the structural conditions in which this area of education operates within the German context. Third, in an empirical investigation, Decker-Ernst presents a survey study of 140 teachers of Vorbereitungsklassen in the German state of Baden-Württemberg. Interestingly, her results show that the actual situation differs considerably from the official numbers presented by the regional authorities; for example, Baden-Württemberg offers many fewer Vorbereitungsklassen than is officially reported. Fourth, the study involves a qualitative case analysis of nine young immigrants and their language learning situations over several months. Thus, a thorough, multi-faceted picture of the present situation emerges, which illustrates the interdependence of educational, political, local and individual variables.

Outside of the school system, Integrationskurse (‘integration classes’) are obligatory courses that immigrant adults are required to successfully complete in order to attain a residence permit. In addition to a German language class, they consist of a culture course with the intent of familiarising newcomers with the historical, cultural and legal cornerstones of German society. Zabel's (Reference Zabel2016) study is set within this context and provides detailed insights into the actual discourses that occur in this type of classroom, and learning materials, focusing on religious pluralism. Zabel concludes that power relations in the Integrationskurse aim for assimilative behaviour regarding dominant societal norms, which runs contrary to pedagogical principles of participation and autonomy.

Finally, Rokitzki (Reference Rokitzki2016) investigated L2 learners of German in adult literacy courses. The study took a classical comparison approach to teaching methods starting with the fact that Montessori's pedagogical approach to literacy development, although apparently suitable for non-alphabetised German L2 learners, had not yet been used for this target group. Using a pre-post design, skills development in a Montessori group were compared with that of two control groups taught through more traditional methods using a rotation design. The results were then triangulated with data from student and teacher interviews in which participants were asked to comment on each method. The results show that the Montessori method seems suitable for learners with no previous literacy competencies, especially elderly learners, and appears to have no disadvantages compared to other teaching methods. However, the investigation also provides evidence that certain learners experience problems with open learning formats.