3.1 ETR German within the architecture of German language variation

According to Coseriu (Reference Coseriu1970, Reference Coseriu1988), language differs with regard to the geographic space (= diatopic differences), the socio-cultural classes within the society (= diastratic differences), and the modalities of expressing content matters (= diaphasic differences). Bredel & Maaß follow the widespread custom of concluding that varieties can be classified into diatopic, diastratic, and diaphasic varieties, and discuss the question of which of the three dimensions is able to provide explanatory power to locate ETR German within the system of language variation. Unsurprisingly, they regard ETR German as related to the diastratic dimension of variation because it is only used by a certain group of people, mainly those with reading disabilities (Reference Bredel and Maaß2016:25). However, ETR is not suitable as a means for group formation (like youth language, the most frequently cited example of a diastratic variety), since it is based on an asymmetry between its producers and its recipients (ibid.). As this asymmetry also holds true for many kinds of scientific texts, Bredel & Maaß do not consider this to be an argument against describing ETR German as a diastratic variety. Apparently, they consider the notion of a non-standard variety according to Coseriu as appropriate due to its restricted range in contrast to Standard German, which Bredel & Maaß classify as neutral in usage (Reference Bredel and Maaß2016:24). On the other hand, Bredel & Maaß also systematically compare ETR German with Standard German (Reference Bredel and Maaß2016:523ff.), which establishes Standard German as a relevant point of reference for describing ETR German.

This brief discussion of the possible status of ETR German as a diastratic variety shows that language variation is much more dynamic than a classification of varieties as diatopic, diastratic, or diaphasic varieties is able to model (see also Section 3.3). Therefore the question of whether ETR German can be classified as a diatopic, diastratic, or diaphasic variety leads to a dead-end. It is a misunderstanding to assume that we use either a diastratic, diatopic, or diaphasic variety whilst speaking. If, for example, I speak to my students in a seminar at my university, my speech is at the same time influenced by the idea of how the social interaction in a seminar works (= diastratic dimension) and by our discussing special content matters (= diaphasic dimension). My accent also shows slight references to my regional origin (= diatopic dimension). Thus the dimensions of variation interact in every act of communication. The idea of chains of varieties (Koch & Oesterreicher Reference Koch, Oesterreicher, Günther and Ludwig1994) makes us aware of systematic correlations between the different dimensions of variation. For example, it is very unlikely that a scientific text carries features clearly marked as diatopically strong (i.e. features of basic dialects) or diastratically low (i.e. colloquial expressions). In contrast, in a conversation with friends from where we come from, we are quite likely to use regional language features and colloquial expressions. What makes ETR languages difficult to integrate into this is that their social and situational settings – i.e. the low diastratic and diaphasic features – would usually correlate with the usage of features of regional languages, but they do not. Rather, in ETR German diatopic language features are a priori excluded, which in the common chains of varieties usually correlates with high diastratic and diaphasic settings (for example written languages for special purposes or scientific languages). But although this might be seen as one further feature of the artificiality of ETR German, we have to note that the general abstraction from the diatopic dimension of variation is also a feature of the Standard German language. This might be the reason why Bredel & Maaß conceptualize Standard German as the basic reference point for ETR German (Reference Bredel and Maaß2016:523ff.) and describe ETR German as a reduction of Standard German. Standard German can generally be characterized by the abstraction from the dimensions of variation, i.e. equally well from the diaphasic and diastratic dimension as from the diatopic dimension of variation. If Standard German is the reference point for ETR German, the conclusion is that ETR German cannot be captured by the above-mentioned dimensions of variation.

3.2 ETR German between immediacy and distance

Whereas Coseriu’s approach is often used to classify varieties due to their major variational features in German linguistics, Koch & Oesterreicher’s model of the language of immediacy and distance is popular for describing various language usage in oral and written communication. Related to Koch & Oesterreicher’s distinction between the language of immediacy and the language of distance, Bredel & Maaß reveal the following paradox: Koch & Oesterreicher model the poles of the immediacy–distance continuum by listing conditions of communication and strategies of verbalization (see Figure 2). Koch & Oesterreicher’s main concern is to show that there is a continuum between the two poles, that is, text types or instances of texts do not have to be either examples of language of immediacy or language of distance, so that there are transitional and hybrid forms as well (for example personal letters, which are written but also show features of immediacy, or lectures, which are spoken but carry features of distance). Surely it should be assumed that the two components of explanation – the conditions of communication and the strategies of verbalization – interact, that is, certain conditions of communication can be expected to lead to certain strategies of verbalization, although Koch & Oesterreicher do not elaborate these relations (for their modelling see Ágel & Hennig Reference Ágel and Hennig2006, Hennig Reference Hennig, Franco and Sieberg2011). Bredel & Maaß show that this does not apply to ETR German: due to the conditions of communication, ETR German can be considered to be a language of distance, because a high degree of planning is demanded and the reception should work without regard to the situation (Reference Bredel and Maaß2016:30f.). On the other hand, ETR German can be characterized by strategies of verbalization such as lower complexity or density of information: see Figure 2.

Figure 2. The paradoxical status of ETR German as a language of distance due to the conditions of communication and a language of immediacy due to the strategies of verbalization.

Note: The figure uses a simplified depiction of the Koch & Oesterreicher model. See Koch & Oesterreicher (1985/Reference Koch and Oesterreicher2012) for the model with all relevant conditions and strategies and the consequences for the location of text types between the poles of immediacy and distance.

Bredel & Maaß conclude that ETR German has neither the structure of the language of immediacy nor the structure of the language of distance (Reference Bredel and Maaß2016:31). To my mind, this conclusion questions the status of ETR German as a variety, since any natural variety ought to systematically interact with the dimension of immediacy and distance as Koch & Oesterreicher have modelled with their notion of ‘chains of varieties’ (Reference Koch, Oesterreicher, Günther and Ludwig1994) (see Section 3.1). Since in ETR German the natural relations between the conditions of language usage and the usage of certain language forms are being suspended, it seems difficult to capture its variational status by linguistic concepts outlined for describing natural languages.

3.3 The notion of ‘variety’ and the dynamics of natural languages

Although the two above-mentioned approaches are very prominent for modelling language variation in German linguistics, and Bredel & Maaß deliver a thorough discussion, I consider it necessary to turn back to the general question of how to define a variety and how to model language variation. In German linguistics, the term variety is commonly used to describe subsystems of the German language. There is broad agreement on describing varieties as subsystems within a natural language that are characterized by certain language features, which are related to extralinguistic conditions (see for example Elspaß Reference Elspaß, Neuland and Schlobinski2018:93). The notion of a variety has mainly been promoted by sociolinguistics. I quote Berruto as a prominent actor in this field, who defines variety as follows:

Berruto himself points out that it is very difficult to quantify the number of features necessary in order to speak of a variety. As far as ETR German is concerned, this should not be the problem: ETR German contains a well-defined number of linguistic features. As far as their functionality is concerned, ETR features are undoubtedly related to the situations of ETR usage. The confusion starts if one considers the correlation with the speaker as an indispensable criterion for calling a certain language usage a variety: ETR features are not fully functional with respect to the producers of ETR texts, but only to their recipients. Producers of ETR texts do of course use the features in order to make them functional for the situations of ETR usage. But the production of a kind of language that falls far below their own language competence and which does not even correlate with their own usage of language in colloquial, non-professional situations, makes them a type of participant that cannot be compared with any kind of participation in language discourse. In other contexts where text producers use linguistic features that are simpler than their usage of language in other situations (the production of textbooks by scientists may be used as an example, or the way teachers speak with their students), the production of utterances that seem adequate for the particular recipients is carried out by the producer’s intuitive understanding of simplicity rather than by the producer’s compliance with a set of predefined rules of simplicity: the producers choose language features because they consider them to be functional for their purposes, not in order to follow the rules by special guidelines. Thus the process of production of more simple utterances in situations like these can be classified as an act of microsynchronization, whereas the orientation towards a norm of simplicity in the production of ETR texts must be seen as macrosynchronization (see below).

On the other hand, the notion of variety is itself not straightforward. Within their theory of language dynamics, Schmidt & Herrgen (Reference Schmidt and Herrgen2011) criticize the assumption of homogeneity that results from many established notions of modern linguistics such as the distinction of synchrony and diachrony (Reference Schmidt and Herrgen2011:21ff.). Furthermore, Schmidt & Herrgen point out that approaches that take the variability of language into account also tend to end up in the homogeneity trap if they classify varieties as static modes of languages: if varieties are conceptualized as homogeneous subsystems of a language, the inadequate notion of a homogeneous competence is duplicated (Reference Schmidt and Herrgen2011:21). To make this problem more concrete: it has often been pointed out that we cannot speak of the youth language or the language for a special purpose, since in fact language also varies within the respective variational contexts. This is the case because language users do not just reproduce a restricted number of variational patterns. We do not just produce our notion of language for a special purpose if we communicate in a special field and we do not always use the same level of dialect independently of whom we talk to. We adopt our language usage to the particular situations of interaction and to our understanding of what kind of language will be appropriate in addressing our respective communication partner. Schmidt & Herrgen call this process ‘synchronization’, which they define as the adjustment of differences in competence within acts of language performance (Reference Schmidt and Herrgen2011:28). Schmidt & Herrgen differentiate between micro-, meso-, and macrosynchronization. Microsynchronization means the individual synchronization processes in individual interactions (Reference Schmidt and Herrgen2011:29). Schmidt & Herrgen define mesosynchronization as a sequence of aligned acts of microsynchronization (Reference Schmidt and Herrgen2011:31), that is, they speak of mesosynchronization if several similar acts of microsynchronization can be amassed. Macrosynchronization, however, cannot just be seen as a further degree of abstraction; instead Schmidt & Herrgen conceptualize macrosynchronization as acts of synchronization where the participants are guided by a common norm (Reference Schmidt and Herrgen2011:32). Schmidt & Herrgen name the emergence of Standard German and the establishing of an orthographic norm as examples of processes of macrosynchronization.

I refer to Schmidt & Herrgen’s ideas, because they provide a model that exceeds the notion of variety: language usage is far too dynamic to be adequately classified by well-defined varieties. On the other hand, linguistics will never be able to describe every single act of microsynchronization. Scientific approaches depend on abstraction and classification. Thus the notion of variety may yet be helpful and should not be removed from our set of linguistic categories. But we should see variety as a methodologically motivated simplification that helps us speak about different modes of language usage.

Within Schmidt & Herrgen’s model the notion of variety is most closely related to the level of mesosynchronization, which is responsible for the emergence of linguistic conventions specific to certain groups and situations (Reference Schmidt and Herrgen2011:31). It is the dynamics of the mesosynchronization process that Schmidt & Herrgen consider to be a more adequate linguistic view on language variation than defining static varieties with well-defined boundaries. If we follow the assumption of Schmidt & Herrgen that dynamics is always fundamental for language variation, Bredel & Maaß’s attempt to capture the special character of ETR German by calling it a static variety (Reference Bredel and Maaß2016:531) must be seen as a contradiction in itself. The dynamic process of mesosynchronization ought to be relevant to the emergence of any kind of mode of language usage that in other linguistic contexts would be called a variety (or a register, style, practice: see below). Now, what is striking about ETR German is that its emergence cannot be fully explained by the concept of mesosynchronization: see Figure 3.

Figure 3. The special status of ETR German as a result of macrosynchronization.

Surely, the emergence of ETR German can partly be considered a sequence of communication acts in which the producers explore options of making themselves understandable to their recipients, and that certain conventions and patterns of an easier German stabilize in their repeated usage in a sequence of similar situations. But, as mentioned above, ETR German is conceptualized to a high extent by its guidelines. As the guidelines constitute a norm, the usage of ETR German is – at least partly – a matter of macrosynchronization: we can speak of macrosynchronization as long as a speaker or writer reflects the norms relevant for his or her production of utterances. The systematic orientation towards an established set of rules makes ETR German different from the wide range of varieties such as dialects or sociolects. As pointed out by referring to the creation of an orthographic norm as an example of a process of macrosynchronization in German, language change cannot only be seen as a bottom-up phenomenon of stabilizing mesosynchronization acts, but it can also be established by top-down macrosynchronizations. But the correlation of macrosynchronization and the notion of a variety which is restricted to a well-defined group of participants seems rather unusual.

3.4 Is ETR German a ‘variety’?

In the discussion of various approaches towards modelling the notion of variety, the following criteria have been worked out as significant for classifying a mode of language usage as a variety.

• – Varieties are part of a natural individual language. As such they underlie the dynamics of language variation and change and emerge via processes of mesosynchronization, i.e. as a sequence of aligned acts of individual synchronization processes in individual interactions.

• – Varieties can be classified by dimensions of variation, such as the diastratic, diaphasic, and diatopic dimensions. Although all the named dimensions influence any individual speech act, individual acts of communication might be classified as instances of diastratic, diaphasic, or diatopic variation if they show particularly salient features of one of the dimensions.

• – Varieties correlate with other varieties within a chain of varieties.

• – Varieties possess a certain number of features characterizing the speaker and/or the situation of language usage. That means that varieties are characterized by systematic form–function correlations.

• – Individual communication acts of any variety of a natural language can be localized between the language of immediacy and the language of distance. That means that they show certain strategies of verbalization which can be derived from certain conditions of communication.

If we consider these characteristics as definitional criteria for a variety, we have to come to the conclusion that it is not adequate to classify ETR German as a variety.

4.1 The notion of ‘register’ by Douglas Biber (and Susan Conrad)

I will now take a closer look at the notion of register by Douglas Biber, since his register approach seems to have an enormous impact on current linguistics, which might be due to its close relation to the currently influential corpus linguistics (see below). Biber’s notion of register is therefore to a high extent adapted to surface characteristics of registers.

Also, in the textbook written together with Susan Conrad, Biber & Conrad (Reference Biber and Conrad2019) classify register as a ‘text variety’, which leads to the assumption that their register approach might be appropriate for discussion of the NJL question of whether ETR languages are ‘just texts that have certain characteristics’. Biber & Conrad use the term text ‘to refer to natural language used for communication, whether it is realized in speech or writing’ and the term variety ‘for a category of texts that share some social or situational characteristic’ (Reference Biber and Conrad2019:5). Starting from this definition, Biber & Conrad differentiate between dialects as ‘varieties that are associated with different groups of speakers’ and registers ‘that occur in particular situations of use’ (which also holds true for genres and styles) (ibid.). Comparing these definitions with the approaches discussed in the previous section, there seems to be an obvious link between Biber & Conrad’s notion of dialect and the diastratic dimension of variation, and their notion of register and the diaphasic dimension of variation. Thus the conclusion above, that the dimensions of variation systematically interact, naturally remains relevant even if we use varying notions while considering language variation. Therefore it seems doubtful to differentiate notions of language variation by strictly separating the groups of speakers and the situations of use: language use always depends on the systematic relations between the situation and the speakers. The notion of register – as understood by Biber & Conrad – might be helpful in order to take a closer look on the situational factors that have an effect on language variation, but it will not provide an overall modelling of language variation. The situational factor is without doubt important for capturing the special character of ETR German. Thus ETR languages are – as the term itself indicates – mainly used as written languages. Although certain attempts to make use of ETR German in oral communication in Germany can be observed – such as radio features in ETR German or the usage of plain language in oral consultations – ETR German is mainly used as a written language in situational contexts with institutional settings. On the other hand, the restriction on situational factors by Biber & Conrad might be considered as problematic in the context of the description of ETR languages, because the social groups unquestionably play an important role in the constitution of such languages.

The question now is what the register approach can offer to locate ETR German within the system of language variation. According to Biber & Conrad, ‘The description of a register covers three major components: the situational context, the linguistic features, and the functional relationships between the first two components’ (Reference Biber and Conrad2019:6). This outline does not differ much from the above-cited definition of variety by Berutto, the only difference being the already mentioned restriction to the situational context. Biber & Conrad describe the functionality as constitutive for registers: ‘When speakers switch between registers, they are doing different things with language – using language for different communicative purposes and producing language under different circumstances’ (Biber & Conrad Reference Biber and Conrad2019:12). They use this defining feature for differentiating registers from dialects as well as from styles and genres. According to Biber & Conrad, ‘register variables are functional, as opposed to dialect variables, which are conventional’ (Reference Biber and Conrad2019:11). They consider variables in dialect studies as being dichotomic whereas variables in register studies are described as scalar: ‘Linguistic variables in dialect studies almost always consist of a choice between two linguistic variants … In contrast, linguistic variables in register studies are usually rates of occurrence for a linguistic feature in a text, and a higher rate of occurrence is interpreted as reflecting a greater need for the functions associated with that feature’ (ibid.). This rather strict distinction seems to originate from a fairly traditional view of dialectology. As far as the current state of research in German linguistics is concerned, modern research on regional languages shows a deep interest in ranges of regional variation (see for example Schmidt & Herrgen Reference Schmidt and Herrgen2011 for the theoretical principles and Kehrein Reference Kehrein2012 for a case study). Thus our discussion of Biber & Conrad’s differentiation between register and dialect leads to the impression that the differentiation depends on how defining criteria are determined and interpreted.

As mentioned above, Biber & Conrad also use the criteria of functionality for differentiating between register, style, and genre. In addition to register, genre and style are introduced as ‘different approaches or perspectives for analyzing text varieties’ (Reference Biber and Conrad2019:15). According to the research question proposed by the Nordic Journal for Linguistics whether ETR is ‘just texts that have certain characteristics’, the differentiation between these three perspectives on textual variation might provide interesting insights into the understanding of ETR languages. I will use Table 1 to start the discussion.

Table 1. Defining characteristics of registers, genres, and styles (Biber & Conrad Reference Biber and Conrad2019:16)

As far as the defining characteristic ‘textual focus’ is concerned, the differentiation between a sample of text excerpts and complete texts indicates a methodological perspective on ‘text’ rather than a defining criterion of textual features. We will come back to that later.

As far as the other defining characteristics are concerned, we can easily conclude that ETR texts can be explained by the notion of register rather than the notion of genre or style as they are understood by Biber & Conrad: according to this outline, genre differs from style and register in the usage of specialized expressions that usually occur only once in a text. This surely holds true for fairy tales with their specialized introductions Once upon a time in English and Es war einmal in German. On the other hand, a feature such as use of the simple past tense (instead of the present perfect) in German fictional prose is frequent and pervasive. Nevertheless, it can be registered that linguistic features in ETR texts are pervasive in general. And they certainly ‘serve important functions in the register’: they are not just preferred aesthetic features. Along with regard to the notion of register, Biber & Conrad provide a fairly restrictive meaning of genre which does not correspond to the use of this concept in sociolinguistics. Thus Spitzmüller points out that the concept of genre in modern sociolinguistics is used for social patterns of action and evaluation that govern the formation of texts and discourses due to parameters such as goals of action and social positions rather than simply categorize them based on surface features (Reference Spitzmüller2013:241). Again we have to conclude that the question of whether or not a certain notion may be appropriate for locating ETR languages within the system of language variation depends very much on how the respective notion is understood. This also holds true for the notion of style: Biber & Conrad’s interpretation of a preference of style due to aesthetic values indicates a rather narrow definition of style. Modern stylistics provides a wider range of more flexible notions of style: ‘Jede Äußerung hat Stil – in Relation zum Textmuster und zu den Umständen ihrer Verwendung [Every utterance has style – in relation to text patterns and to the circumstances of its usage]’ (Sandig Reference Sandig2006:2). Due to this understanding of style, features are indeed functional.

Now how can the notion of register be used to describe the linguistic characteristics and their functions within ETR texts? With regard to the linguistic features, Biber & Conrad distinguish register features and register markers: register features are characteristics that occur ‘more commonly in the target register than in most comparison registers’ (Reference Biber and Conrad2019:54). Register markers on the other hand are ‘distinctive linguistic constructions that do not occur in other registers’ (ibid.). Biber & Conrad conclude that ‘register markers are rare’ and therefore ‘Most registers cannot be identified by the occurrence of a distinctive register marker’ (Reference Biber and Conrad2019:57). This also holds true for ETR German. The differences between ETR German and other modes of German language usage are quantitative rather than systematic, as stated in Section 2.

And what are functions in a register and how can they be described? Biber & Conrad admit that this step of the overall analysis is open to interpretation (Reference Biber and Conrad2019:69) and thus has to be carried out inductively: ‘once the situational and linguistic analyses are completed, the functional analysis involves matching up characteristics of the two’ (Reference Biber and Conrad2019:74). But even if this step is open to interpretation, we need an overall model for functions relevant to distinguishing various modes of communication. Biber & Conrad provide a list of ‘Major functions that distinguish among registers’ and ‘Selected linguistic features associated with each function’ (Reference Biber and Conrad2019:73). The list consists of general functions such as interactivity or referring to the time and place of communication or shared personal knowledge; general communicative purposes such as narrative, description, directive, or procedural; presentation of information such as elaboration or condensation; and finally production circumstances such as real time and careful production and revision. Examples of associated features are pronouns for interactivity, past tense verbs for narrative, prepositional phrases for condensing information, and complex noun phrases and sentences for careful production and revision. Biber & Conrad describe their list as a ‘starting point for undertaking the functional interpretations’ (Reference Biber and Conrad2019:74). This quite modest classification of their own approach offered for the analysis of functions seems adequate, since the list does not provide a systematic approach to relations between functions and language variation. To describe and model functions is a much more difficult task for linguistics than to describe features visible on the surface of language usage. The modelling of functionality might be more promising if the model were restricted to one dimension, as is shown by Koch & Oesterreicher’s (1985/Reference Koch and Oesterreicher2012) model of the language of immediacy and distance, which is restricted to the variation between orality and literacy. Nevertheless, in a way it is disappointing that an approach that declares functionality to be the major key to understanding the modelled notion does not provide a more consistent model of linguistic functions. Furthermore, the ‘starting point’ list does not provide any relevant examples or ideas that give us a better understanding of how ETR languages can be explained.

Since functions of language phenomena are less feasible than the description of linguistic forms, the register approach pays more attention to the methods of describing linguistic features. Biber & Conrad advocate a comparative quantitative approach to the analysis of register (Reference Biber and Conrad2019:53ff.): assuming that register markers are rare and thus registers can be distinguished by register features as defined by the notion of frequency, it is obvious that a quantitative approach is the means of choice. Note that one of the central projects of the Collaborative Research Centre ‘Register’, currently being carried out in Berlin, is named ‘Data management and statistical analysis’ and, for instance, one of the projects on ‘register and grammar’, which deals with syntactic register variation in German, aims at developing Biber’s approach further by ‘(i) using a probabilistic approach to registers, (ii) using a more sophisticated method for automatically inferring registers from distributions of features in texts’. Unsurprisingly, the register approach is also closely linked to corpus linguistics (Biber Reference Biber2012). To avoid misunderstanding, I want to make it clear that I do not point out the relations of the register approach to corpus linguistics and quantitative linguistics because I doubt that these relations are useful or because I want to criticize the tendencies towards these kinds of linguistic approaches. On the contrary, I assume that these relations are one reason for the apparent success of the notion of register in current linguistics.

4.2 Is ETR German a ‘register’?

Whereas the discussion of the applicability of the notion of variety to ETR German leads to the conclusion that ETR German cannot really be classified as a variety because it does not fulfil the criteria compiled from various approaches, we are confronted with a different picture with regard to the notion of register: register seems to be a term for a wide range of notions which are too different to amount to a coherent picture of one clearly focused notion of language variation. Therefore the question of whether the notion is applicable to ETR German can only be discussed with regard to individual approaches. I chose Biber & Conrad’s approach due to its current popularity and to their focus on text variation.

Admittedly, the last section provided a critical discussion of Biber & Conrad’s notion of register rather than a thorough application to ETR German. As a theoretical approach, Biber & Conrad could not convince me due to their attempt to make clear distinctions between concepts such as dialect, register, style, and genre by arranging relevant parameters such as function and situation in a distinctive model. Such a clear-cut model is surely attractive in a way, but it neglects the existing research traditions.

As a methodological approach to the analysis of larger quantities of texts, the register approach could serve as a reference point for research on ETR languages. The methods applied in the register approach could support access to ETR languages that concentrates on the mesosynchronization processes, that is, processes of conventionalizing language phenomena proven to meet the needs of easy-to-read texts. As far as the research on ETR German is concerned, it is highly influenced by the strong impact of the ETR guidelines and has therefore promoted empirical research on phenomena sanctioned by the guidelines by the members of the target groups. Corpus studies on the usage of these phenomena in ETR texts have also been carried out (see for example Lange Reference Lange2019 on the genitive case or Rocco Reference Rocco2021 on clause linking). For a more neutral view on how simplicity is displayed in texts that are supposed to be easy to read (as for example carried out by Lange Reference Lange2018), it might be helpful to make use of the methods proposed by the register approach.