Beyond the binarism: locating past, present and future sociolinguistic research on ideologies of communication

: This opinion piece discusses how research on language ideologies (or, for that matter, ideologies of communication in general) relates to micro/macro distinctions of both the research field and the discipline of sociolinguistics. Starting off from the thesis that both object and strand of research at stake here do not neatly fit this dichotomous distinction, the paper discusses the Fishmanian origins as well as the diverse and diverging definitions of micro-and macro-sociolinguistic s that ensued from Fishman’s proposal, considering the different assumptions that underlie the respective variants of the distinction. The piece discusses why language ideology research locates its object, and itself as a discipline, beyond the micro/macro binar-ism, and how/why it embraces a more scalar, relative and reflexive approach to social and linguistic juxtaposition, ordering and ranking. To this end, the concept of ‘social scales’ and its uptake by language ideology research are introduced and discussed. Finally, the paper provides an opinion on what the future of sociolinguistics should be like – independently from the particular strand at stake here – and about the role of this newly launched journal (with its long history as a yearbook) as the author wishes to see it being taken.


The challenge
On the occasion of the Sociolinguistica relaunch we celebrate with this journal double issue, I was asked to provide an opinion piece on past, present and future developments in 'macro-sociolinguistics', the realm Sociolinguistica dedicates itself to, with specific consideration of my own research area, language ideologies (or ideologies of communication, as I term it more generally; see Section 2).I confess that, at first, this request left me puzzled.Soon, however, after pondering the request more -and why it puzzled me -, I felt challenged.It will take the remainder of this article to fully elaborate the reasons for this; in what follows, I will only briefly sketch why I think the task is a challenge.
To begin with, I would not readily construe what I am doing, and what research on ideologies of communication in general is doing, as 'macro-sociolinguistics' (hence my initial uneasiness with the task).Is it 'micro-sociolinguistics'?No, not that either.What is it, then?As the title of my paper indicates, I think it is an endeavour that does not neatly fit the binary distinction that is presupposed in the task.Research on ideologies of communication, I argue, deals with a phenomenon that is at the intersection of what has traditionally been called the 'micro' and 'macro levels', and therefore cannot restrict itself to either.
What is more, research on ideologies of communication suggests that the social is itself not as neatly dividable into two ready-made realms as the binarism has it.Consequently, as I will expand on in more detail below, some scholars in language ideology research set out to develop a more granular, and relative, approach to their object, which is also more reflexive with regard to social (and disciplinary) scaling.
My elaborations, however, will also reveal that the aim for granularity and to a certain degree also for reflexiveness, and not least the reservations about the separation of sociolinguistic strands alongside the micro/macro dichotomy, are far from novel.In fact, they were already central issues of some classic sociolinguistic approaches and seem to have been sidelined over the course of the ongoing institutionalisation and compartmentalisation of the discipline(s).This is why I end up proposing that the future of sociolinguistics is to be sought in part in its own past.
In order to expand on my argument, I start with a brief reminder of what language ideology research is all about and succinctly introduce the more encompassing and lesser-known concept, ideologies of communication.I then trace the disciplinary roots of the macro-/micro-sociolinguistics distinction and locate research on ideologies of communication in this context.I subsequently elaborate on how and why language ideology research, among other sociolinguistic research strands, departed from the micro/macro distinction of the field, embracing the somewhat shimmering concept of 'social scales' instead.I close the article with some concluding thoughts on 'the future' as I see it.

Language ideologies and ideologies of communication
Language ideology research deals with the social evaluation and construction of language, languages, language varieties, forms of language use, and language users.Established within linguistic anthropology (see Woolard and Schieffelin 1994), this area of research draws on the basic premise that language is being constantly evaluated and assigned indexical meaning, i.e., associated with social values, expectations and assumptions concerning usage and users.Languages (as countable, delimitable entities) and coherent language forms are regarded as social constructs that draw on such evaluations -evaluations which also help to order them hierarchically and hence associate language (use) with social power (see Blommaert 2005;Gal and Irvine 2019).
The field of study sets out to explore how such "sets of beliefs about language" (Silverstein 1979: 193) emerge, how they change, how they are contested and negotiated, and what consequences they have in specific societies.It is presumed that language ideologies are ambivalent concepts which, on the one hand, bear negative consequences (such as social discrimination) for particular social actors, but are, on the other hand, also crucial for social orientation (they are the prerequisite for socialising functions of language; see Silverstein 2003).Furthermore, language ideologies are regarded as the mediating link between language structure and language use, the cultural processor that makes a structural element 'usable' (and 'appropriate' in a given usage) as well as a linguistic use 'structurally' locatable (see Silverstein 1979).
In recent years, the scope of research has been widened from language to a broader range of semiotic means, a widening from language ideologies to ideologies of communication (as I prefer to term it) in a wider sense.Besides language ideologies, ideologies of communication also include sets of ideas about graphics, media, and gesture among other semiotic modes (taking communication as a concept that integrates all these modes).Research into ideologies of communication does not only explore these mode-specific evaluations themselves, but also tackles the multimodal interplay between them (see Spitzmüller 2015;Nakassis 2016).This is, very broadly brushed, the field from which I speak.Before I elaborate how it relates to micro-and macro-sociolinguistics, I will first go back some steps and discuss what these designations are about anyway.We will see that this is not as trivial a question as it might seem.

Micro-and macro-sociolinguistics
As is well known, the distinction between micro-and macro-sociolinguistics was introduced by Joshua Fishman (e.g. Fishman 1971: 37-55).Drawing on the established distinction within his own discipline, micro vs. macro sociology (see Knorr-Cetina 1981), Fishman conceives of micro-sociolinguistics as being concerned with "speech acts and speech events" (Fishman 1971: 42), whereas macro-sociolinguistics is "concerned with social processes and societal organizations" (Fishman 1971: 44).This distinction has since been interpreted in different ways.Many have allocated or even equated Fishman's own approach, the sociology of language, to macro-sociolinguistics.Some have thereby related 'micro' and 'macro' to the number of speakers under investigation (i.e., the quantity and quality of the data), and have consequently conceived of variational sociolinguistics and the sociology of language as being 'macro-', as opposed to ethnographic and interactional approaches, which have sometimes been construed as representing 'micro-sociolinguistics' (see e.g.Trudgill 2004: 1;Wardhaugh and Fuller 2015: 15).Others (e.g.Coulmas 1997;Darquennes 2015) have proposed to associate the two categories with two different social theories rather than with data quantities, aligning micro-sociolinguistics with socio-structural approaches and macro-sociolinguistics with societal treatment approaches (notably leaving out the social-constructive approaches which have become dominant in many new strands of sociolinguistics; see Coupland 2001).Note how ideology-related research is included in the latter: Stated in very general terms, micro-sociolinguistics investigates how social structure influences the way people talk and how language varieties and patterns of use correlate with social attributes such as class, sex, and age.Macro-sociolinguistics, on the other hand, studies what societies do with their languages, that is, attitudes and attachments that account for the functional distribution of speech forms in society, language shift, maintenance, and replacement, the delimitation and interaction of speech communities.(Coulmas 1997: 2) As opposed to the former, this proposal not only considers variational sociolinguistics, which is deeply rooted in a structure-functional social theory (see Coupland 2001: 9-10), as a form of micro-sociolinguistics (despite its orientation towards social structures), it also equates micro-sociolinguistics with 'sociolinguistics proper' and macro-sociolinguistics with 'the rest': […] there are as it were two centers of gravity, known as micro-and macro-sociolinguistics or alternatively sociolinguistics in the narrow sense and sociology of language.These represent different orientations and research agendas, micro-issues being more likely to be investigated by linguists, dialectologists, and others in language-centered fields, whereas macro-issues are more frequently taken up by sociologists and social psychologists.(Coulmas 1997: 2) Unlike these attempts to associate the two categories with specific disciplines or strands, Fishman himself was adamant that any sociolinguistic approach needs to consider both levels of analysis (e.g., see Fishman 1971: 54).As his later re-formulation of the distinction as micro-versus macro-level sociology of language (Fishman 1972c: 29-53) indicates, he did not accept the exclusive allocation of his own approach to the macro level.Also, Fishman located some of his own central concepts precisely at the intersection of ("the boundary area between"; Fishman 1971: 51) micro and macro, most notably the notion of 'domain' (see Fishman 1971Fishman : 51, 1972a)).This is particularly interesting in our context since domains in Fishman's terms -"dominance configurations" (Fishman 1972a: 441) determining language/style preference -are clearly related to language ideologies and social power.Fishman describes domains as "higher-order generalizations from congruent situations" (Fishman 1972a: 444) that "enable us to understand that language choice and topic […] are […] related to socio-cultural norms and expectations" (Fishman 1972a: 441).In other words, domains are normative (ideological) constructs that emerge from (recurrent) local situations and feed back into them as social frames, thus mediating between ('macro') social processes and ('micro') speech events.

Language ideologies: neither nor
If we depart from Fishman's work and look at where else language ideologies, or related concepts, have been approached, it becomes even more clear that this research strand cannot be easily allocated to either 'micro-' or 'macro-sociolinguistics'.Besides other sociological approaches (most notably Bernstein 2003Bernstein [1971] ] and Bourdieu 1977), evaluations of language and language use have been a topic in variational socio linguistics (see Labov 1972Labov [1963]]), the social psychology of language (see Giles et al. 1987), and in (perceptual) dialectology (see Niedzielski and Preston 2000) -fields that cover both sides of the micro/macro coin, notwithstanding which of the categorisation proposals we follow.
As sketched out in Section 2, language ideology research proper stems from linguistic anthropology.It is in many ways conceptually aligned with its disciplinary kin, interactional sociolinguistics, which itself meanders between 'micro-level' analyses (of speech events) and 'macro-level' contextualisation (in 'social frames'; see Gumperz' elaborations in Prevignano and Di Luzio 2003).As mentioned above, language ideologies have been construed from the very beginning as being located, and mediating, between 'structure' and 'use' (see Silverstein 1979).Very early on, consequently, the research located itself at the intersection of social structure and local practice (see Gal 1989: 349;Woolard and Schieffelin 1994: 55).However, with further development, language ideology studies concentrated more and more on (transtextual) metapragmatic discourse, thereby developing a bias towards the structural and often neglecting local practices (see Woolard 2008: 437;Spitzmüller et al. 2021).So there was clearly, within a given period, a shift towards the macro-sociolinguistic pole of the continuum (in terms of data quantity and the primary interest in social structures) -although following Coulmas' conception we could also state that language ideology research seemed to morph into micro-sociolinguistics, since language ideologies were increasingly regarded as larger-scale social constructs that frame local behaviour and practice, and it was primarily the impact of the social on the local that was the focus of research.
This shift in focus has been identified as a problem within the field since the 2000s.As an attempt to bring back to focus the other direction of mediation, several concepts have been proposed that attempt to model how local practice feeds into language ideologies, propelling their emergence and transformation.The most seminal among these concepts are iconisation/rhematisation (Gal and Irvine 2019), enregisterment (Silverstein 1993(Silverstein , 2003;;Agha 2007), indexical order (Silverstein 2003), and semiotic chains (Agha 2007;Wortham and Reyes 2020), concepts which all seek to describe how inter-subjective and inter-situative ideologies emerge and thicken from reiterative language use in specific institutionalised (and power-framed) social contexts.The turn towards the question of how ideologies emerge and change in discourse resonates with the aforementioned increasing role of (post-structuralist and phenomenologically informed) social-constructive social theories in sociolinguistics as of the 2000s (see Coupland 2001).Many researchers in the field of study have criticised the structure-functional social theory that dominated sociolinguistics in the linguistic strands of sociolinguistics (i.e.Coulmas' 'micro-sociolinguistics'), which proposes to conceive of language as being dependent on the social ('language reflects society'; e.g.see Cameron 1990).Opposed to this unilateral conception, social-constructive social theories conceive of language and society as multilaterally related to, and deeply entangled with, each other.Informed by these accounts, language ideology research became increasingly interested in how reality (the world as social actors construe it) is being made by human practice (sensu Berger and Luckmann 1966) and built up repetitively and recursively (sensu von Glasersfeld 1995: 118).
In the wake of this re-orientation, and informed by sociological debates from the 1980s along the same lines (e.g.Knorr-Cetina and Cicourel 1981;Giddens 1984), language ideology research -and critical/anthropological sociolinguistics in generalincreasingly stated its qualms with, and consequently challenged, the micro/macro binarism itself.Heller (2001: 212), for instance, notes how [c]onceptualising social life in terms of a dichotomy implies that there are different types of data for each, equally observable (or not, as the case may be), and that, in addition, the linkages should be identifiable.And yet, empirical work fails to identify such types.
Two decades earlier, in the context of the sociological debate on the topic, Knorr-Cetina (1981: 16) suggested that the associations drawn to the two levels are probably more a function of the researcher's stance than of the objects and data that are commonly being subsumed to them: […] many of the dimensions in terms of which the micro-macro problem has traditionally been couched will have to be reconsidered, if not abandoned.I have in mind distinctions such as that between the individual and the collectivity, between action and structure, between smallscale uniformity and large-scale complexity, or between the association of the micro level with neutrality or powerlessness and of the macro-level with power.The above remarks suggest that the connection made between the various poles of such dimensions and the micro-versus macro-level may be a function of the observer's distance to the respective field of study rather than being inherent in the problem itself.
As an alternative to the now problematised dichotomy, a new concept has been brought to the fore and eagerly embraced: social scales.

Embracing 'scales'
The notion of 'scales' is not just an attempt to replace the dichotomous macro/micro opposition with a more granular "continuum of layered scales, with the strictly local (micro) and the global (macro) as extremes, and with several intermediary scales […] in between" (Blommaert 2007: 1).The concept involves a more fundamental shift, as implied in the quote by Knorr-Cetina above, from a-priori (albeit heuristic) categorisations of one's own research field to the process and the consequences of categorising, or scaling, objects and processes within that field.This requires making scales or scaling an object, rather than a prerequisite, of research.By doing so, researchers try to address, as Latour (2005: 183)  As Carr and Lempert (2016: 4) argue: The task of the analyst […] is to leave behind a priori scalar distinctions and instead empirically track how social actors carve and cleave -or scale -their worlds.
From the perspective of language ideology research, Gal and Irvine (2019: 218-219) furthermore note that: To assume, as is so often done in invoking a micro-macro binary, that scalable dimensions must coincide -or that there is only one kind of scale, or one way of scaling -is an ideological artifact, perhaps emerging from the institutions that regiment people's lives (including our own) and align scales according to some social project.Scaling is a relational practice that relies on situated juxtapositions and comparisons among events, persons, things, and activities; that is, among sites of attention […].Moreover, like any other ideological project, scaling implies positioning, hence point of view: a perspective from which scales, as modes of comparison, are constructed.It is from such a perspective, a line of sight as it were, that aspects of the world are noticed, evaluated and compared.
In other words, scaling is regarded as an ideological process, and a scale as an ideological product (often fostered by powerful actors and institutions).At the same time, scales are also means of ideological work.Like other ideological constructs, they are conceived of as powerful devices that help to form social order and positions (see Carr and Lempert 2016: 3).In this sense, distinctions between 'micro-' and 'macro-sociolinguistics' or 'narrow-sense' and 'broader-sense sociolinguistics' discussed above can be regarded as being more than rational attempts to divide the broad realm of sociolinguistic research into two sensible halves.They are the results of attempts to scale (and ideologically rate!) academic disciplines, attempts that are conducted from particular academic points of view and that express specific disciplinary stances.
With regard to language ideology and ideologies of communication themselves, it has been pointed out that scaling is a central part of the evaluation of language, language use, language user and communication in general.As Blommaert neatly demonstrates by example of a mini-dialogue between a PhD student (S) and their tutor (T) -which is actually quite reminiscent of Fishman's take on social domains -, scaling communicative practices is a common way to (de-)authenticate and (de-)legitimise forms of communication: S: I'll start my dissertation with a chapter reporting on my fieldwork.T: We start our dissertations with a literature review chapter here.The tutor performs a scale-jump here, in which s/he moves from the local and situated to the translocal and general, invoking practices that have validity beyond the here-and-now -normative validity.This "upscaling" is articulated through a change from personal to impersonalcompare S's use of "I" and "my" with T's "we" and "our," and T's invocation of "here": a community larger than just the student and the tutor.(Blommaert 2007: 6) As Blommaert notes, the normative account of the tutor ('we do it like that here') points the student to genre expectations which the student is supposed to meetacademic ideologies of communication (usually called 'genre conventions' or 'academic style') which include notions about 'usual' or 'appropriate' text structuring, style, format (see Spitzmüller 2021) and so on.These notions are located on the larger (translocal) scale of the 'we' that rules out the local 'I' here.
On a more general level (yes, I am scaling myself here), conceptions of 'correctness', 'appropriateness', 'standardness' (see Milroy and Milroy 1999), 'normality' or 'authenticity' (see Bucholtz 2003) provide firm anchor points on translocal scales with regard to which forms of communication and communicators are being calibrated (e.g. as to the question whether and to what extend they are 'typical' or 'normal') but which might, reversely, also be calibrated ('scaled down') to the local (e.g. in terms of local specificities and creative [re-]contextualisation). Within such processes, practices are made to appear more 'local' or more 'global' for specific interpreting actors, and according to the proponents of the scalar approach, it is the challenge of sociolinguistics to reveal how such scaling processes work and what they effect, rather than pre-categorise and pre-allocate them according to the discipline's own scales.As we have seen, this also requires a reflexive (or critical) perspective on one's own ordering processes (see Heller et al. 2018: 10).

Sociolinguistics: back to the (indistinct) future?
Now where do we go from here?Are scales the new master key of sociolinguistics or just a shiny new concept?Neither.Scales are heuristic tools that help to understand how important it is to look at how the world is (communicatively) configured, how phenomena (including linguistic and semiotic phenomena) are juxtaposed and ranked, also within academic research, and how this relates to social positioning.The concept, however, does not relieve us of the essential epistemic task of categorisation, ordering and perspectivation (and the concept of 'scales' is arguably itself a specific and contestable way of doing that).As far as 'micro' and 'macro' are concerned, the concept of 'scales' at least helps to remind us that these are (rather rough) constructs that have been established within academia in order to ease categorisation, ordering and perspectivation -constructs that are, as we have seen with the example of the evaluative rankings of micro-versus macro-sociolinguistics (and as can be seen more generally with regard to the quantitative/qualitative debate; see Dörnyei 2007: 24-47), all but devoid of ideology.However, this does not necessarily mean that we need to drop them completely.If we look back, we can see that these scholars have been very aware of the issues discussed here and have desperately tried to prevent the heuristic dichotomy from being essentialised and hard-linked to disciplines.
One reason for this awareness might have been the initial disciplinary fuzziness of sociolinguistics, the ambitious interdisciplinary endeavour that took off six decades ago, heading towards a completely open, and avowedly not exclusively linguistic, future.As we know, with the increasing institutionalisation, interdisciplinarity got lost and the field became scaled into zones of 'proper sociolinguistics' (= 'linguistics') and -as Fishman (1972b: 216) remarked resignedly while he retreated from sociolinguistics (the interdisciplinary endeavour) to the sociology of language (a branch of sociology) -"'not really sociolinguistics' or […] only 'so-called sociolinguistics'".This is more than a pity.Despite all that specialisation has brought us, this sort of discipline-keeping has come at the cost of epistemic diversity and friction.As we have seen, it took two decades for the concept of 'scales' to make its way from sociology to sociolinguistics.On a more general level (yes, I am scaling again), sociological debates about what society and 'the social' actually are, are rarely being taken up in sociolinguistics (cf. the notable exception of Coupland 2001) -this is considered a problem sociology needs to take care of.And these are just two cases from two of the former sibling disciplines that constituted the joint endeavour incidentally called sociolinguistics.
Given this, I think part of the future of sociolinguistics is to be found in its (often overlooked) diverse and interdisciplinary past, a past where the future was all but clear-cut.I hope that Sociolinguistica, itself deeply rooted in that past, will provide an arena that will help us to discover new indistinct futures -an arena that is open not only to 'macro-sociolinguists', not only to 'proper sociolinguists', and ultimately not only to linguists.
As far as research on ideologies of communication is concerned, the development sketched here -the extension of the research scope from language (proper) to other modes of communication as well as the consideration of scaling -is definitely worth pursuing.In order to grasp the complexity of evaluation in communication, the more critical and reflexive approach to social scales, the abandoning of the micro/macro binarism and the embracing of multimodality are indispensable.However, the turn towards (even more) complexity has raised an abundance of theoretical and methodical questions that are far from resolved.After all, it is easy to state, but much harder to tackle, that language and society are in a complex co-constructive relationship, and that ideologies shape and emerge from local practice.Even if some promising concepts are at our disposal already, research on language ideology has embarked on a long and tough journey.Here as well, (inter-)disciplinary companions will facilitate the travel.
puts it,[t]he problem[…]that social scientists use scale as one of the many variables they need to set up before doing the study, whereas scale is what actors achieve by scaling, spacing, and contextualizing each other.