The actual use of classroom language is principally limited to the classroom environment. As far as foreign language learning is concerned, the classroom often turns out to be the learner’s only source of exposure to the target language in question. Therefore, this special type of language can be considered as an extremely important factor in foreign language acquisition (Butzkamm, 2007). The first attempts to establish a theory of classroom language can be traced back to the late 1960s and the early 1970s (cf. Loch, 1966; Priesemann, 1971). Although Priesemann’s (1971: 19) claim that classroom language consists of elements of mother tongue and language of science may be regarded as plausible, it is neither of the two. In reality, it is a language in its own right, the one that plays an intermediary role between everyday speech and language of science. Similarly, Leisen (2011, 2018) collocates classroom language somewhere between conversational language and language for specific purposes. The same author attributes to classroom language the function of “a methodical intermediary language” (2018: 13). Gogolin (2003) refers to classroom language as ‘school language’, thus stressing the fact that school communication follows its own regularities, one of them being a strict role division. Specifically, classroom language consists of elements of teacher and student language, with teacher language dominating the classroom discourse. Since classroom language mainly refers to spoken classroom communication, it is rather frequently equated with classroom talk. Accordingly, teacher and student language are referred to as ‘teacher and student talk’. The teacher supervises communication patterns. These patterns were documented for the first time by Sinclair and Coulthard (1975). They introduced the so-called IRF model (initiation - (learner’s) response and feedback) for the analysis of classroom discourse. This model is predominantly used to describe and analyse the interaction between the teacher and the learner(s). Learners are aware of their modest part in this classroom talk and accept the role of passive observer, especially in a foreign language classroom (Holliday, 1994). The reason for this lies in the special role of classroom language in foreign language didactics. In point of fact, it is not only the objective, but also the medium of instruction (cf. Voss, 1986; Walsh, 2006). Leaning on this fact, Walmsley (1986) considers classroom language in a foreign language classroom as “a variety – a kind of Fachsprache in its own right” (1986: 50), thus establishing an immanent relation between classroom language and other languages for specific purposes. In Voss’s view, classroom language is more than an array of classroom phraseology. It is “the language used by teachers and pupils to conduct classroom business” (Voss 1995: 5) and should be actualised almost exclusively in the target language. Consequently, there are many authors who emphasise the need for explicit and systematic teaching of classroom language to students and teachers of foreign languages (cf. Walmsley 1986; Butzkamm, 2007; Voss, 2009; Schröder, 2010). Since teachers play a central role in teaching a foreign language, it is no surprise that teacher talking time, unlike student talking time, is one of the most frequently researched and analysed aspects of classroom language (cf. Legaretta, 1977; Chaudron, 1988; Helmke et al., 2008; Cook, 2013). All results show that teacher talking time amounts to approximately two-thirds of the total in-class talking time. This makes teacher talk one of the most important inputs received by learners during the process of foreign language acquisition. Krashen (1985) stresses the importance of comprehensible input. If input is to be comprehensible, it has to be adapted to the level of learners’ language proficiency, which inevitably implies simplifications at all linguistic levels (Ellis, 2003), such as slower speech speed, overstressed pronunciation, vocabulary and grammar simplification, overusing gestures, more frequent repetition of one’s own statements (Chaudron, 1988). The most important work in the field of classroom language in Croatia can be attributed to Sanja Čurković-Kalebić (2003; 2008) and Yvonne Vrhovac (2001). Apart from comprehensibility, Ehlich and Rehbein (1986) emphasise the importance of fluency of teacher talk. Correspondingly, several surveys of mistakes in teacher talk were carried out (cf. Corder 1967; Breitkreuz & Liedke, 1975; Nünning, 1980). As a result, many pragmatic handbooks and coursebooks on classroom language have been published to help foreign language teachers learn the respective classroom language and acquire communication competences in L2, above all for teaching English as a foreign language (cf. Gressmann & Rich, 1982; Cattliff & Thorne, 1988; Voss 1995; Hughes, 2001). It is worth noting that Croatian research on classroom language appears not to be inferior to international achievements in this field. As early as in 1974, O. Gerčan and A. Menac wrote an article with a list of 100 frequent phrases pertaining to German, English, French and Russian classroom language. The first comprehensive bilingual handbooks of classroom German (Jelaska & Lütze-Miculinić 2018); classroom English (Zergollern-Miletić & Lütze-Miculinić, 2020) and classroom Italian (Mardešić & Lütze-Miculinić, 2021) based on Croatian as L1 were published as part of seven research projects on classroom language conducted at the University of Zagreb – Faculty of Humanities and Social Sciences. However, the question arises as to whether classroom language can be learnt at all. By analysing classroom discourse, Vrhovac (2001) noticed certain ritualised parts in teacher discourse which she called “regulative classroom talk” (2001: 128). She also noticed that these ritualised parts are often missing in the discourse of young and inexperienced teachers. Since classroom discourse is for the most part predictable and prepared in advance and only to a lesser extent it turns out to be unpredictable and unprepared, i.e., similar to natural communication, it can be concluded that teacher talk is mostly learnable, though still dependent on the individual teacher’s language proficiency. The aim of this dissertation was to show the impact of a systematic and explicit teaching of classroom language in the education of German language teachers in the Croatian language and educational system. The pilot study conducted in 2017 by Lütze-Miculinić and Landsman Vinković found that students and novice teachers of German as a foreign language, native speakers of Croatian, showed certain deficiency in acquisition of classroom German, notably at the morphosyntactic, semantic and pragmatic levels. Following the results of the study, a module dedicated to classroom German was created with a view to facilitating the acquisition of classroom German by students of German who specialise in teaching German as a foreign language. The research in this dissertation was conducted among students of German Language Teacher Education before (S1a) and after the process of explicit learning of classroom German (S1b) as well as among novice teachers of German who had not been exposed to this input (S2). Since the number of students enrolled in the courses of German as a foreign language in one generation is relatively small, the sample comprised two generations of students (N=32). The respondents belonging to novice teachers of German (N=24) served as the control group in the research. Beside some characteristics typical only of this type of language, classroom language consists of general language elements. For this reason, respondents’ linguistic competence was assessed through a placement test conducted according to the Common European Framework of Reference for Languages. The expected level of German language proficiency of the respondents was C1-C2 (proficient user). The following hypotheses were put forward: S1a and S2 should achieve similar results in the test on both general and specific classroom German, whereas S1b should achieve better results only in specific classroom German. Respondents within S1a and S2 with a higher level of German language proficiency would better adjust their classroom language to the real classroom situation than those with a lower level of German language proficiency, while this would be the case within S1b only when it comes to general classroom German. The first research instrument was a combination of an anonymous questionnaire and a test on classroom German written specifically for the purpose of this research. It contained elements of the already developed module on classroom German, focusing on areas which were found problematic in the diagnostic test conducted by Lütze-Miculinić and Landsman Vinković in 2017. Special attention was given to compensation and avoidance strategies as special types of communication strategies. The classification of errors and strategies used in the research lean on the classification proposed by Kleppin (1998). The questionnaire contained general questions on respondents, whereas the test consisted of 14 Croatian to German translation tasks as well as of 6 fill-in-the-gaps tasks. Students of German Language Teacher Education took the test before (T1) and after (T2) attending the module on classroom German, whereas novice teachers took the test (T1) during their Preparation Training for Teacher Bar Examination organised by the Croatian Education and Teacher Training Agency. Three ideal independent respondents, native speakers of German, well-acquainted with the Croatian school system, took the test for instrument validation. Their answers served as the model for assessing the answers of the respondents on the scale from 1 to 3, where 1 was used for an unacceptable answer, 2 for an acceptable with minor grammar and stylistic errors and 3 for the answer identical to the answer provided by native speakers of German. The test results of S1a and S2, as well as of S1a and S1b were compared based on the grade point average on a 3.0-scale for every item and respondent. The SPSS Programme Package for Windows 20.0 was used for the quantitative analysis and description of the results (correlation between the variables, graphs and tables). In the second part of the research, short interviews in L1 were conducted with the respondents on the difficulties they had encountered while doing the test on classroom German. A qualitative analysis of the results was carried out in order to gain a deeper insight into and a better understanding of problems the respondents meet while using classroom German in conducting lessons in L2. The third part of the research consisted of audio-recorded microteaching lessons of S1 before and after taking the module on classroom German. The aim of this part of the research was to confirm the positive effect of explicit teaching of classroom German on the respondents. Since classroom language refers predominantly to spoken communication conducted in the classroom, this part of the research provided an insight into potential pronunciation and intonation deviations, i.e., phonological errors. The qualitative-quantitative analysis was conducted on the corpus of transcripts of audio-recorded microteaching lessons. Parts of teacher bar exams in which candidates’ inclass performance was assessed were partially observed and partially audio-recorded with the aim to detect problems in usage of classroom language in a real classroom situation. The results proved that S1a and S2 achieved similar results in the test on both general and specific classroom German, whereas those exposed to this input (S1b) achieved better results not only in specific classroom German, but in general classroom German as well. This proved the efficiency of explicit and systematic teaching of classroom language. Respondents within S1a and S2 with a higher level of language proficiency in German adjusted their classroom language to the real classroom situation better than those with a lower level of language proficiency in German. The analysis of the audio-recorded in-class performance of students of German Language Teacher Education provided an insight into diverse aspects of teacher talk, such as repetition, use of pauses, code-switching and code-choice, turn-taking, responding to learners in a foreign language classroom. Furthermore, it crystallised certain problematic areas which were beyond the scope of the first part of the research which analysed the written version of respondents’ knowledge of classroom German. The research contributed to a better understanding of the process of acquisition of classroom language and communication competences of teachers of German as a foreign language. It is expected that the obtained results will provide a theoretical basis for the implementation of explicit teaching of classroom language into study programmes of higher education institutions in Croatia. Furthermore, the above-mentioned results and insights into various aspects of teacher talk may be employed as a starting point for future research into classroom language for other foreign languages taught in Croatia. Finally, the research indicated that there is a necessity to provide professionals already employed in Croatia’s system of education with lifelong teacher training, notably with regard to this aspect of teaching German as a foreign language.