The relationship between norms and practice

While the three topics we address in this book – Standard and regional varieties of Japanese; honorifics and politeness; and gendered language – have drawn extensive attention in Japanese sociolinguistics, they have tended to be studied separately. Our goal here is to employ a coherent theoretical perspective to these different phenomena, and examine them not simply as a list of three autonomous topics of interest, but as interrelated phenomena that have been shaped in the historical and ideological context of modern Japan. That is, rather than attempting to simply present a description of how linguistic forms relevant to each of the three phenomena are used in contemporary Japan, we also aim to interrogate how and why such use as well as its interpretation has come about. This consideration will make it evident that we cannot ignore the historical background, particularly the extent of the effect of modernity in the face of cultural “tradition,” or how and to what extent the modern Japanese language ideology constructed by the modern nation-state formed in the Meiji period (1868–1912) affected Japanese speakers’ linguistic life, both in attitudes toward language variation and in situated practice. We believe that this approach helps us understand the seemingly disparate sociolinguistic phenomena more adequately.

This approach, then, requires us to develop a framework that allows us to view these three topics from a broader theoretical perspective in order to account for them in a coherent manner. For this purpose, we employ a combined sociohistorical analysis and an indexical approach based upon an understanding of forms as inherently polyindexical, to consider more fully the fluidity and multiplicity in the relationship between the variation in linguistic forms and its social significance, paying special attention to the role of language ideologies over the span of Japanese modernity in mediating this relationship. The majority of previous studies on the three topics we interrogate in this book have tended to relate linguistic forms directly to social meanings, often independently of their historical and ideological contexts, an approach consistent with earlier research in sociolinguistics in general, which has tended to conceptualize the relationship between form and indexical target as more direct and one that is, moreover, more exclusively speaker-centric in terms of meaning. The underpinnings of this are: structuralist notions of stable connections between signifier and signified, which assumes that a linguistic form has a meaning and hence that the two can be directly linked to each other; a referentialist ideology that privileges the referring function of speech over its indexical – that is, its context-related or performative – functions; and a personalist ideology that privileges speaker intention over interlocutor uptake (Hill Reference Hill2008: 38; Kroskrity Reference Kroskrity and Kroskrity2000: 5; Shibamoto-Smith Reference Shibamoto-Smith2011). In this program of research, which may be termed modernist sociolinguistics (Rampton Reference Rampton1997, Reference Rampton2006), the link between linguistic forms and social meanings has tended to correspond to dominant social norms, or language ideologies, as in claims such as “feminine forms are used by women”; “honorifics are used to express politeness.” In other words, “actual practice” is often confounded with social norms, suggesting that scholars themselves are not free from dominant cultural and linguistic ideologies. Such disregard for the historical and sociopolitical processes by which certain practices become the dominant norms while others are erased or relegated to marginal status (Cameron Reference Cameron, Joseph and Taylor1990; Irvine and Gal Reference Irvine, Gal and Kroskrity2000) effectively occludes our understanding of the potentially diverse social meanings of linguistic variables.

Before proceeding with our argument, we take a moment to elaborate on what we mean by “actual practice.” We use the term “actual practice” to orient the reader to the speaking practices of real Japanese social actors interacting verbally with others in a broad range of real-world interactional contexts (conversations with social intimates, personal emails, business meetings, classroom interactions, and the like); we do not, here, restrict empirically or theoretically interesting “actual practice” to small face-to-face verbal interactions with social intimates, as was the tendency in the early days of Western sociolinguistics, but rather conceptualize it as the speech of real social actors in the multiple communities of practice (or, to use a term preferred by some, communities of performance; M. Tanaka Reference Tanaka2005). Such interactive settings, to be sure, necessarily include practices of making metapragmatic commentary on language, such as seen in the blog entries we analyze here. However, we distinguish our analyses of these blog entries from the use of language for social interactions in real-world situations, such as face-to-face conversations and email, since our purpose here is to examine them not for their form but for their content as it relates to attitudes about the social norms of language use. Those Japanese writers whose fictional work we analyze in this volume as expressive of norms would also be included as engaging in “actual practice.” Again, for the purposes of our analyses here, we distinguish this from the use of language for social interaction in real-world situations, as the representation of language use in fictional worlds is a kind of implicit metapragmatic commentary. For instance, the use of Standard Japanese stereotypical women’s language for heroines, as we have argued previously (Okamoto and Shibamoto-Smith Reference Okamoto and Shibamoto Smith2008) and will revisit in Chapter 5, is, we will argue, an indirect commentary on how this variety of Japanese is evaluated.

We emphasize, however, that the distinction between the expression of metapragmatic stance (such as commentary about language use in online blogs and dialogs in fictional worlds) and simply speaking (or writing) is neither a simple nor clear-cut one, as noted decades ago by Prague School linguist Roman Jakobson (Reference Jakobson and Sebeok1960). Language functions (referring, indexing, waxing poetic) do not occur as neat, mutually exclusive “types” of utterances. For example, the use of language in daily face-to-face conversations also involves metapragmatic stance taking, as we attempt to demonstrate in this volume (Chapters 2, 4, and 6). This is equally true of conversations drawn from televised variety or talk shows involving media celebrities. Although they are taken here as samples of unscripted, interactional (that is, conversational) text, it is certainly the case that they, too, include elements of represented speech or stereotypified speech forms – or a ventriloquistic metapragmatic commentary on how certain kinds of social actors “should” speak – in the interests of capturing audience attention. In short, we make a distinction between metapragmatic commentary and “actual practice” as an analytic heuristic, but stress that we use these terms (“actual practice” and “metapragmatic commentary”) only advisedly, in the spirit of Jakobson, to distinguish speech that is rather more oriented to other communicative functions than the metalinguistic (and metapragmatic; see Silverstein Reference Silverstein and Lucy1993) function, in the first case, than speech that is more consciously and systematically focused on precisely the metalinguistic and metapragmatic.

The present volume also aims to situate our arguments in the context of contemporary trends in sociolinguistics in general. We do this by engaging in a closer dialogue with recent shifts in sociolinguistics away from a focus on quantitative variation in linguistic form to the social impacts of that variation, that is, away from a focus on referential meaning to one on indexical and social meanings (Eckert Reference Eckert2008). In this process, the relationship between linguistic forms and social meanings has come to be seen as multiple, fluid, and mediated by cultural and linguistic ideologies.Footnote ¹

This book is offered as a contribution to inquiry into the three core topics in Japanese sociolinguistics from this perspective by employing an indexical approach that attempts to overcome the earlier static approach discussed above, set against a backdrop of the sociohistorical conditions within which particular directions in sociolinguistic inquiry was encouraged. Building on recent research that has substantially modified earlier descriptions of Japanese language practice based largely on survey reports of centralized tendencies in the use of Standard Japanese, we aim to offer careful and nuanced analyses of the indexicalities that are involved in these three sociolinguistic phenomena, paying special attention to the relationship between sociolinguistic ideologies, or norms, and actual practices in contemporary Japan – or the relationship between macro-level and micro-level social meanings. This allows us not only to illuminate diversity both at the levels of empirical speaking and of interpretation but also to account for such diversity. We examine that diversity within the context of, or in relation to, the dominant as well as competing ideologies of Japanese language use as they pertain to our understandings of each of these three key areas in Japanese sociolinguistics.

We also propose to take a step toward addressing the gap between the stereotypified visions of the sociolinguistic phenomena we cover here and the core interests of Western sociolinguistics as well as drawing upon the more fluid and relational understanding of language practice prominent in sociolinguistics today by historicizing its relevance to Japan’s modernity.Footnote ² Accordingly, we begin our discussion of each of the three sociolinguistic phenomena we cover here by addressing the processes of norm construction in relation to Japan’s modernization and post-war democratization, the contexts in which they became critical objects of inquiry. We then examine how – and to what ends – dominant language norms have been constructed in Japanese society, and how – and to what social ends – they have been or continue to be circulated publicly. We ask as well how competing ideologies have manifested in various discursive practices, and how that diversity has (or has not) shifted the dominant discourse. Ideologies are intertwined in empirical practice with diverse actual language choices. A historical examination of what has changed over time and what has not adds a necessary dimension to our understanding of the multiplex relationships between linguistic expressions and their social semiotic effects more fully.

Our work, of course, does not emerge in a vacuum. Other recent studies in Japanese sociolinguistics, including the volume we ourselves co-edited (Okamoto and Shibamoto Smith Reference Okamoto and Shibamoto Smith2004a), have increasingly recognized the importance of distinguishing norms from practices. However, many of these studies have tended to focus either on investigations of the construction of sociolinguistic norms, or ideologies, about how language should be socially structured and used (Heinrich Reference Heinrich2012; M. Inoue Reference Inoue2006; Komori Reference Komori2000; M. Nakamura Reference Nakamura2007a, Reference Nakamura2014a; Yasuda Reference Yasuda1999, Reference Yasuda2007), or on the diverse language practices that are documented in observational studies or sociolinguistic interviews, many of which are simply presented as “facts” without extending the analyses to address broader social theoretical impacts (Ariizumi Reference Ariizumi2007; Ebata Reference Ebata2001; Gendai Nihongo Kenkyūkai Reference Kenkyūkai1999, Reference Kenkyūkai2002; Machi Reference Machi2004; Nakanishi Reference Nakanishi2008; Nishioka Reference Nishioka2009; Okamoto Shin’ichirō Reference Okamoto2001; Ozaki Reference Ozaki2003). These studies have made valuable contributions to our understanding of complex sociolinguistic phenomena in Japan. In particular, recent studies concerning Standard Japanese/kokugo ‘national’ language and those concerning gender and Japanese have considerably enhanced our understanding of the ideological and political underpinnings of these sociolinguistic issues.

While there are studies that do examine how norms and practice are related in specific social contexts, overall there has been a tendency to stop at the level of each analytical context, rather than to build upon individual studies to develop a more comprehensive theory of the range of possible ways in which norms and practices can be related. Most importantly, macro-level and micro-level sociolinguistic issues have not been adequately integrated with each other in a coherent theoretical framework. And to do so is crucial. As emphasized in a number of recent studies (Agha Reference Agha2007; Cameron Reference Cameron, Pichler and Eppler2009; Coupland Reference Coupland, Lacoste, Leimgruber and Breye2014; Mills and Mullany Reference Mills and Mullany2011; Silverstein Reference Silverstein2004), we cannot understand the macro-level enregistration of sociolinguistic patterns of variation as identifiably associated with particular groups without attending to micro-level language use in interaction and vice versa. Throughout the chapters in this volume, we will strive to connect these two levels iteratively and dialectically, and to develop a systematic account of how they are mutually constitutive.

Teasing out the relationship between macro and micro levels of the sociolinguistic issues at the center of our inquiries – Standard and regional Japanese forms, honorific or plain forms, and feminine and masculine forms – requires us to historicize their development, as noted earlier. In addition, we need to consider how these particular sets of variable linguistic forms have come to be associated with certain social meanings, how those meanings gain hegemonic status in the dominant discourses of “good” language practice, and how the use of these forms is negotiated and contested in specific sociohistorical contexts. In examining these processes, we employ an approach we term “polyindexical analysis” – an indexical approach that emphasizes the variability and multiplicity of social meanings of linguistic forms as semiotic signs, or the fluid nature of “indexical fields” (Eckert Reference Eckert2008). This serves as a framework to investigate the social/indexical meanings of linguistic forms in a fluid rather than a fixed form and the relationship between norms and practices. To provide the reader with relevant background to this approach, we introduce several key notions: sociolinguistic categorization; linguistic norm construction; the role of language ideologies; and indexicality as a fluid process.

Sociolinguistic categorization

First of all, investigation of the relationship between linguistic norms and practice as they relate to the three topics that will be covered in this volume requires us to consider the nature (and the historical context) of sociolinguistic categorization in general. We need, that is, to ask the following questions: What are the social categories that sociolinguists and discourse analysts typically draw on cross-linguistically in their analyses of speech practices, and why are certain categories selected for particular investigations and not others? What are the sociolinguistic categories most commonly utilized in sociolinguistic studies of Japan and what are occluded in most studies? Are the chosen categories the ones most relevant to the Japanese linguascape or are they the most relevant for particular representational models of “Japan” to a non-linguistic audience? And finally, why do certain socio-historically “selected” categories appear to be so resistant to challenge in the Japanese case? It is no longer sufficient simply to challenge the old, nihonjinronFootnote ³-based notion of Japan’s population as a homogeneous people living in the homogeneous social world or linguistic community; such views have been challenged repeatedly and effectively (Gottlieb Reference Gottlieb2005; Heinrich Reference Heinrich2012; Heinrich and Galan Reference Heinrich and Galan2011a; Maher and Macdonald Reference Maher and Macdonald1995; Maher and Yashiro Reference Maher and Yashiro1995; Noguchi and Fotos Reference Noguchi and Fotos2001). What we aim for here is less a challenge to the homogeneously construed categories than the historicization of the process by which particular categories came to be the categories of choice and a fuller consideration of the sociopolitical issues of power in language and the struggles that are fought over language which underlie the continued popular circulation of the linguistic “myths” of homogeneity. We try to assess the reasons why the cultural capital of Standard Japanese as opposed to regional dialects, honorifics as an expression of Japanese politeness, or Japanese “women’s language” and its counterpart “men’s language” continue to receive such high valuations both within and outside Japan (Bourdieu Reference Bourdieu1991; for a discussion of Japanese national literature from a similar perspective, see Essertier Reference Essertier and Bourdaghs2010).

Much of the early research on Japanese language and culture, both in English and in Japanese, has tended to present static characterizations of certain sociolinguistic categories based on the social categories that are considered internally homogeneous. For example, it is often taken for granted that “the Japanese” speak “the Japanese language,” “hyōjungo,” or “Standard Japanese,” and that the residents of “provinces” in Japan speak distinct “regional dialects”; the relationship of those provincial residents to “the Japanese” who speak “the Japanese language” is not, in general, specified. Other common categories are “women,” who, naturally, speak “women’s language,” and “men,” who likewise, naturally speak “men’s language.” Again, the relations of the categories “men” and “women” to the categories “the Japanese” and “regional dialect speaker” – not to mention necessarily co-present categories such as age, social class, gender orientation, and sexuality – are un- or under-specified. Another example is the view that “the Japanese” universally participate in the cultural practice of “Japanese politeness,” which is then linked categorically to the rule-governed use of honorifics (Ide and Yoshida Reference Ide, Yoshida and Tsujimura1999). This approach assumes the internal homogeneity of each social category, which in turn makes it easy to assume equally internal homogeneity in linguistic style.

Yet it is increasingly recognized that this assumption needs to be re-examined (for work on this in other languages as well as Japanese, see Auer et al. Reference Auer, Hinskens and Kerswill2005; Brown Reference Brown2003; Myers-Scotton Reference Myers-Scotton2002; for Japanese, Heinrich and Galan Reference Heinrich and Galan2011a; Maher and Macdonald Reference Maher and Macdonald1995; Okamoto and Shibamoto Smith Reference Okamoto and Shibamoto Smith2004a). Sociolinguistic categories such as women’s language and men’s language presuppose the corresponding social identity categories of women and men. However, as most postmodern feminist linguists would argue, these – and by extension – other social categories cannot be taken as fixed, nor can the assignment of individuals to those categories be assumed as unproblematic or, in any given context, unproblematically relevant (Cameron Reference Cameron2005). Eckert (Reference Eckert, Campbell-Kibler, Podesva, Roberts and Wong2002: 100–101) argues that “[t]he problem of working with identity categories in sociolinguistics in general lies not in attention to categories, but in the way in which the categories are selected and the manner in which they are invoked.” Social categories are not pre-given fixed social realities but rather are emergent constructs. In problematizing the notion of social categories not as natural, but as constructed social norms, recent research on diversity in sociolinguistics has come to focus on the language practice of speakers who do not belong to societies’ pre-fixed or normative categories. However, this has led to a neglect of those normative categories themselves, because “their very supposed ‘naturalness’ insulates them from study as constructed and contingent dimensions of identity” (Bucholtz Reference Bucholtz1999: 443). This is unfortunate, since in any society it is likely that at one time or another many, if not most, social actors will align themselves or find themselves so aligned by others with one or another of the macro-level, pre-fixed social categories used to “define” a population within a state. Our goal in the following chapters is to re-examine the presupposed sociolinguistic categories in the Japanese case in order to offer a more fluid and dynamic account of how Japanese-speaking subjects construct, negotiate, and modify their relationships with the core social categories reified in Japanese sociolinguistics through their speaking choices.

Linguistic norm construction

The second issue concerns linguistic norms, or more specifically, the nature of the norm construction process and its effects. This issue is directly related to the first, as the construction of dominant linguistic norms relies on the ideology of internally homogeneous sociolinguistic categories. In this study we re-examine the social categories relevant to our three core topics, not as given, but as sociohistorical constructs; we interrogate how they have been shaped according to perceived social and political needs in modern Japan and what role they play for those speakers who are seen or who see themselves as members of particular social categories. Irvine and Gal (Reference Irvine, Gal and Kroskrity2000) offered a framework for understanding norm construction processes in the context of building a sociolinguistic modernity, identifying the critical processes that result in certain speakers coming to “stand for” their group (iconicity) while others are rendered invisible (erasure),Footnote ⁴ often with loadings of moral or social value accompanying these iconicized linguistic differences (Shibamoto-Smith and Chand Reference Shibamoto-Smith, Chand, Bayley, Cameron and Lucas2013). More recently, Gal (Reference Gal, Duchêne and Heller2012: 24) reiterates how these processes worked as central aspects of monolingual standardization as a hegemonic sociolinguistic regime in a modernizing Europe, and we show throughout our discussions that these processes have worked powerfully in a modernizing (and modern) Japan as well.

Standard language cultures, including modern Japan, work tirelessly to reach the state where speakers believe that their language exists in a standardized form and that this form is the “correct” or “best” form (Silverstein Reference Silverstein, Brenneis and Macaulay1996); this is accompanied by an implicit sense (a common sense) that if there are two possible forms for saying the same thing, only one of them can be right (J. Milroy Reference Milroy2001). Messages about the “correct” way to speak Japanese abound – as illustrated by the metapragmatic commentaries on what should be Standard, or correct, Japanese (covered in Part I of this volume), or by the voluminous official pronouncements about “correct” honorific use (covered in Part II), or by the similarly voluminous popular literature specifying how women can become more attractive by using “correct” women’s language forms (Part III), all emanating from pronouncements of official institutions and circulating through the popular media. The effects of these messages on both linguists and ordinary speakers are profound, but understudied.

We argue that there is a stronger tendency to take such normative statements more uncritically as “truth” in the “collectivist” Japanese “culture” than in the case of other post-industrial societies, both in the domestic literature but perhaps even more prominently in the globally circulated depictions of Japanese people (for an excellent critique of this tendency with respect to Japanese gender depictions, see Kitamura Reference Kitamura2009, Chapter 3) and their language practices. But these norms, even in the Japanese case, are under increasing pressure to admit new, more various options in late modernity, whether they be to accommodate bi- or multilingualism as ideal states as in the EU, or to accommodate differences of dialect, register, and style as the new “normative” as in Japan. Just as monolingual standardization as the dominant ideal in European language regimes is changing in late modernity, so we recognize it is in Japan as well, albeit without the same prominent shift toward an ideal multilingualism that is seen in the EU. Indeed, our re-examination of the diversity within each sociolinguistic category we examine suggests that many times actual language practice partly conforms to and partly diverges from what has been presented as “the” Japanese language, honorific rules, and women’s or men’s language as social norms. Our aim for each part of this volume is to discuss, on the one hand, how sociolinguistic norms have been constructed and disseminated in society and, on the other, to investigate how and why actual linguistic practices conform to or differ from such language norms.

The role of language ideologies

Why do speakers conform to or diverge from the dominant norms in their language choices? How are social meanings assigned to linguistic forms? How are norms and practice related to each other? Addressing these questions leads to our third framing issue, namely, the role of language ideologies both in the construction of normative language usages and in the on-the-ground use and interpretation of linguistic expressions. Considering this issue is essential to understanding the indexical process through which linguistic forms are linked to particular social meanings, which then serves as a basis for language choice. In this process, language ideology serves as a mediating link between social structures, or meanings, and forms of talk (Woolard Reference Woolard1992: 235; Silverstein Reference Silverstein, Mertz and Parmentier1985), as linguistic forms cannot be regarded as a direct reflection of social structure. Here we take language ideologies as “‘commonsense’ sets of beliefs about language and their links to ‘reality’ and to social value that speakers can and do use to rationalize their language use and their attitudes toward the language use of others, with consequences for language structure and language change” (Shibamoto-Smith and Chand Reference Shibamoto-Smith, Chand, Bayley, Cameron and Lucas2013: 36). They represent the perception of language structure and use constructed in the interest of a specific social or cultural group (Kroskrity Reference Kroskrity and Kroskrity2000: 8). It is hardly surprising that ideologies constructed by a dominant social group with the resources to promote them, and even to coerce their adoption, would exert a great deal of influence on the formation of linguistic norms that gain hegemony in society (Gal Reference Gal, Schieffelin, Woolard and Kroskrity1998: 321).

However, a dominant ideology, being a reflection of an interested social group, is by definition contested (Gal Reference Gal1992; Irvine and Gal Reference Irvine, Gal and Kroskrity2000; Kroskrity Reference Kroskrity and Kroskrity2000; Schieffelin et al. Reference Schieffelin, Woolard and Kroskrity1998). Indeed, a dominant ideology is itself a product of contestation rather than something that exists autonomously as a given (Briggs Reference Briggs, Schieffelin, Woolard and Kroskrity1998), because a language ideology as part of a cultural system of ideas about social order is always built on some group’s “situated, partial, and interested character of conceptions [about] and uses of language” (Errington Reference Errington2001: 110, emphasis added). In other words, dominant ideologies that arise from a ruling social group are not the only ideologies. Competing ideologies are always juxtaposed, explicitly or implicitly, with dominant ones in any given historical context. Meaning-making is “fundamentally social” and with different ideologies, different interpretations are possible (McConnell-Ginet Reference McConnell-Ginet2011). As we will see, competing ideological formations in the Japanese case, as uncovered in the metapragmatic discourses and actual practices we examine, include an idealized monolingual Standard Japanese ideology versus linguistic pluralism, strict linguistic status marking versus solidarity marking, and the ideology of “Japanese women’s language” versus gender-free language use.

In our analyses, therefore, we pay special attention not only to dominant ideologies but also to competing ideologies, as both are likely to affect the interpretation, or evaluation, and use of particular linguistic forms. In particular, we pay special attention to the possibility of diverse interpretations, an issue that has not been investigated adequately. We do so by examining both metapragmatic comments and the use of language in real interactions. For each of the three topics we examine, we first analyze metapragmatic comments as expressed, directly or indirectly, in various genres of message-disseminating texts, including official documents on language policy, media representations in the fictional world, as well as opinions expressed in online blogs. Needless to say, metapragmatic comments cannot be regarded as accurate descriptions of what speakers actually do. Accordingly, they have not received serious attention in sociolinguistic research,Footnote ⁵ but they have to “come to be part of people’s sociolinguistic world and cannot be dismissed as ‘inauthentic’ or evaluated as only ‘second-best’” (Johnstone Reference Johnstone and Coupland2013: 401). This is because they provide us with important opportunities for understanding speakers’ beliefs about language structure and use, or language norms. While metapragmatic comments may constitute an excellent site for investigating language ideologies, we also examine actual language use closely, as only actual interactional data can afford us direct information about how speakers negotiate norms shaped by the dominant ideology and choose particular linguistic forms, or how and why speakers conform to or diverge from the normative expectations in practice.

Indexicality as a fluid process

The issue of how to account for diverse linguistic practices in a theoretically coherent manner leads us to our fourth and final framing question, namely, the issue of indexicality, or how linguistic forms, language ideology, and possible social meanings are knitted together through linguistic choice as social practice into an indexical system of social meanings.Footnote ⁶ We have previously noted the tendency of many sociolinguistic studies to link certain linguistic forms directly to a particular social variable (or social category), such as nation-state, region, gender, class, or ethnicity. But language forms cannot be considered a direct reflection of a particular social category; rather, language and social category associations are mediated by language ideology. This has profound implications for language use and interpretation, since it concerns our view of language, or language variation, itself. That is, speakers are not robotic constructors of utterances “suitable” to their objectively definable social categories but rather are agents in constituting those broad social categories, and are active participants in constructing – through their own speech and the interpretation of the speech of others – the social meaning(s) of variation (Eckert Reference Eckert2008: 3). Linguistic choice thus should be viewed as social practice. In other words, we take language not as an abstract system with well-formed structures, but “as doing” social acts toward certain interactional goals (Pennycook Reference Pennycook2010: 2). Sociolinguistic variation offers concrete resources to which speakers can attribute particular social meanings and values and in which they make different investments according to the interactional needs in a given situation (see Blommaert Reference Blommaert2010: 28).

But what does a given linguistic choice “mean” in a particular instance of social practice? The answer to this question is that the speaker’s intended meaning is potentially multiple and ever-shifting and the uptake (or meaning for the addressee or audience) is very much up for grabs. Repeated indexical acts at one point in time conventionalize a sociolinguistic sign, enregister it into some “folk model” of indexical value (Agha Reference Agha2005, Reference Agha2007). Once the sign is conventionalized (or stereotypified, or enregistered) in that way, the sign can then be used for additional indexical ends. That is, for example, a sign that is conventionalized as deferent can be offered in a context where deference is socially inappropriate and be meant – and taken – as ironic, sarcastic, or even rude. It is this iterative process of expanding the meaning potential of indexicals in this way which is at the very heart of verbal interactions.

Eckert (Reference Eckert2008) introduces the notion of the indexical field to account for the multiplicity and variability of indexical meanings that emerge in use:

The iterative process of expanding semiotic possibility noted above is possible precisely because linguistic forms and social meanings are not in a fixed and mechanically retrievable direct indexicality but rather emerge within a fluid and ever-changing ideological field.

The ideological field, moreover, also incorporates indexicalities at several levels. Silverstein (Reference Silverstein2003: 193–194) proposes the notion of indexical order – a concept he regards as “necessary to showing us how to relate the micro-social to the macro-social frames of analysis of any sociolinguistic phenomena.” At the nth order, indexicals are linguistic forms associated with some category of speakers; for example, when we hear a man speaking Japanese with an Ōsaka accent, we recognize him as likely to be from Ōsaka. If we encounter him as a performer in a comedy club and if, over time, we encounter other speakers with Ōsaka accents also performing in comedy clubs, we may – in our next encounter with a man from Ōsaka – have a predisposition to assume that he will be, if not actually a comedian, a little more humorous than his uptight Standard Japanese-speaking counterpart. That is the n+1 order, driven by the tendency to make a contextual entailment (Ōsaka man=funny) from regular exposure to, and engagement with, the nth order indexical. As socially organized actors (Voloshinov Reference Voloshinov1986), we encounter nth order indexicals in contexts of this sort that, over time, allow us to develop what Silverstein (Reference Silverstein2003: 194) refers to as “cultural construals of the nth order usage.” And as he concludes (Reference Silverstein2003: 212), “for any indexical phenomenon at order n, an indexical phenomenon at order n+1 is always immanent, lurking in the potential of an ethno-metapragmatically driven native interpretation of the n-th-order paradigmatic contextual variation that it creates or constitutes as a register phenomenon.” This process of reinterpretation applies iteratively, hence the potential multiplicity and variability of indexical meanings.

It is hardly surprising that, with the potential for any indexical to link to multiple social meanings at so many indexical levels, individuals may take different positions – or stances – toward a given ideological formation regarding the use and interpretation of certain linguistic forms (for example, Standard Japanese, honorifics as the center of Japanese linguistic politeness, and feminine speech); for a recent explication of this latter point and its relevance to language ideology, see Jaffe (Reference Jaffe and Jaffe2012). Furthermore, competing ideologies may lead one to interpret a given linguistic form differently from the meanings offered in a dominant discourse. This has been found to be the case in many historical and social contexts, and we illustrate it in the Japanese case with examples in all three parts of this volume. Differing interpretations are also possible because a dominant ideology itself may vary over time, as is reflected in the changes in government policy on the use of Standard and regional Japanese after World War II (WWII) (as detailed in Part I, Chapter 1) or on the use of honorifics (as detailed in Part II, Chapter 3). Such diachronic variation at the level of state policy implies pressures from synchronic variation among individuals in attitudes and actual practice at different historical moments. There is, then, an ongoing dialectic relationship between ideology and practice.

In examining actual language practice, therefore, we not only investigate the relationship between norms and practices but also attend to what apparent differences in that relationship might mean in social practice, emphasizing the fact that there exist competing ideologies among different groups and individuals within the population of Japanese speakers as well as diverse, more fine-grained (situation-sensitive) attitudes toward language use, even of the most normative and stereotypified forms. In each of the parts to come, our analyses will interrogate how dominant ideologies came into being, and how they have contributed to twentieth and twenty-first century “discourses of self-representation” (Besnier Reference Besnier2009). They will, as well, trace the ways that multiple ideologies may impinge on individual consciousness and shape the multiple stances that any given individual – whether speaker or interlocutor – might take in a given moment (Jaworski and Thurlow Reference Jaworski, Thurlow and Jaffe2009), thus illuminating the roles of dominant and competing language ideologies in the relationship between linguistic forms and social meanings. We show, furthermore, how critical it is to recognize that individual speakers take varied stances both toward what they understand to be the dominant norms and toward alternatives as they make choices in particular interactional settings, and that their interlocutors do so as well, adding to the complexity of teasing out the range of possible meanings of particular verbal offerings.

An approach which keeps considerations of all three key issues in play is vital to our study, in which we regard language use as a historically developed social practice adopted by historically situated social actors to construct identities and interpersonal relationships. It is, indeed, increasingly important as the recognition grows that it is not “an” identity that is constructed through speech but multiple identities, or personae, carried out through a creative manipulation of the broad range of linguistic resources, from the normative to the peripheral, available to the speaking social actor (see, for example, Coupland Reference Coupland2007 and the various contributions in Eckert and Rickford Reference Eckert and Rickford2001) – a view that regards identity not as static and invariably associated with an individual, but as fluid and understood always in relational contexts (Bucholtz and Hall Reference Bucholtz and Hall2004). Linguistic choices thus can be seen as elements of a strategy for constructing identities and relations in specific contexts. While recent research in Japanese sociolinguistics has paid increased attention to this insight, it has yet to incorporate it fully into its investigations of the three classic areas of interest we cover in the volume.

Furthermore, we emphasize that the linguistic construction of identities/personae and interpersonal relationships is not simply a matter of employing linguistic forms, or variants of a variable, independently from each other. Because we regard language choice as social practice, we see individual concrete linguistic forms, or tokens of variants, as constituting a set of resources that are strategically deployed during a given interaction. For example, as we illustrate in the chapters that follow, speakers often do not simply choose one variant (for example, a Standard Japanese form, an honorific form, or a form considered to be feminine) over another in the course of a given interaction. Rather, they often combine forms with distinctly different macro-sociologically dominant indexicalities (Standard and regional forms, honorific and plain forms, and feminine and masculine forms) in the same conversation, combining them in a complex manner. Styles such as feminine and masculine are not constructed simply by employing a limited set of linguistic features (for example, self-reference and address terms or sentence-final particles). Rather, speakers may utilize a wide variety of features (lexical items, intonation, and the like), which may combine both feminine and masculine forms. Contextual contingencies (such as particular speech acts and speech genres) may also trigger the use of certain forms. In other words, speakers attend to both local and overall effects in dealing with variation.

Data and methods of analyses

For this study we examine data drawn from a variety of sources. The first comprises archival materials on official policy concerning the three major topics; the second consists of contemporary metapragmatic commentaries available in a variety of forms; and the third set of data is drawn from samples of actual language practice. For each topic covered we look first at the history behind the development of that topic as one central to Japanese sociolinguistics. The data for the first chapter of each part of the volume are official or semi-official documents developed by government-sponsored policy institutions and disseminated through the national education system. They are accompanied by a briefer look at data exemplifying how normative discourses at this level are circulated through popular media. The second set of data, contemporary metapragmatic commentaries in a variety of forms, appear in the second chapter in each part. They are drawn from a variety of sources: metapragmatic comments on official policy, expressed in surveys conducted by government agencies and researchers, as well as by the authors; and those directly available in such genres as self-help books and online blogs. The second chapter in each part also includes a third set of data, drawn from transcripts of face-to-face conversations and written discourses in the form of emails.

Our analyses employ archival methods for the sociohistorical aspects of the study, and methods used in discourse analysis and interactional sociolinguistics, with an emphasis on qualitative analyses, for elucidating the relationship between those socio-historically developed norms of language and the realities of language as used in everyday social practice. One of the important issues in analyzing data concerns the interpretation of social meanings. It is common to present researchers’ own interpretations, and that is what we will, of necessity, do here. But we are keenly aware that the potential social meanings of variation are highly variable and multiple, that they depend on the interpreter as well the context. Accordingly, we take pains in our analyses to consider possibilities of diverse meanings, examining a variety of supporting evidence, including the metapragmatic commentary we report for each topic. In this way, while admittedly still dependent on our own interpretations, we try to offer as broad a range of potential interpretations as possible.

Chapter summaries

This book is organized into three major parts that correspond to the three major sociolinguistic phenomena we address. Each part consists of a pair of chapters: the first chapter in each part investigates the historical process of dominant norm construction and dissemination, along with discussion of the role of the developing norms in Japan’s discourses of self-representation. The next chapter explores and attempts to account for the diversity in attitudes toward and actual interactional uses of the linguistic forms related to the particular sociolinguistic phenomenon in question. Although we treat each of the three central topics separately, we stress that none of the three phenomena covered operates in isolation from the others. All three phenomena are viewed as mutually implicated in producing the speaking (and hearing) subject in the historical context of modernizing and modern Japan. As the chapters unfold, we work to develop an account of how they operate mutually to create a stereotypified image of “the” modern Japanese person and how this image has gained hegemony, while at the same time being contested, resisted, or negotiated.

In Part I, we consider the following very basic question: What is Nihongo, or the Japanese language? It is often taken for granted as being “what the Japanese people speak,” but at the same time this is assumed to be Standard Japanese. In school textbooks, Japanese language books, grammar books, and the like, Nihongo is used as a synonym of Standard Japanese. And this is presented as a “natural” phenomenon only to be expected in an ethnically and culturally homogeneous state. But there is diversity within Nihongo that cannot easily be subsumed under the notion of Nihongo-as-Standard Japanese. Such diversity is related variously to speakers’ and hearers’ class, gender, generation, and regional location as well as their beliefs and attitudes toward the imperatives of a standard language culture. The notion of Nihongo then is ambiguous: at the level of dominant ideology, it is regarded as isomorphic with Standard Japanese, while at the level of actual practice, things are more complex, encompassing widely diverse forms of Japanese. Part I addresses the question of how conceptualizations of Nihongo as uniformly standardized are related to empirically observable diversity by focusing on issues concerning the relationship between Standard and regional Japanese (Chapters 1 and 2).

Chapter 1 reviews how Standard Japanese was constructed and promoted as the dominant Japanese language for use in modern Japan. It was neither an easy nor a straightforward task to create such a variety, nor to regiment its use. We review the substantial research on the history of its formation (Komori Reference Komori2000; Lee Reference Lee1996; Sanada Reference Sanada1991; Twine Reference Twine1991; Yasuda Reference Yasuda1999, Reference Yasuda2007). Through this review, we argue that in the case of Japan, the moral panic surrounding the construction of a national language involved issues concerning regional dialects rather than other kinds of language varieties. Consideration of this issue helps us understand how and why Standard Japanese and regional dialects emerged as linguistic categories central to sociolinguistic research. There has also been a vast amount of research on the fate of Japanese dialects under this process. Reviewing this research we offer an outline of some aspects of how regional dialects fared over the course of the twentieth century as well, both in official policy and in popular media circulations of official policy. This chapter as a whole demonstrates how the dominant forces engendered by the government policies (including education policies) and supported by media discourse have brought about a “standard language culture” (J. Milroy Reference Milroy2001) in modern Japan.

Chapter 2 then considers the effects of this standard language culture on real speakers’ language life. That is, here we examine how or to what extent the ideology of Standard Japanese as the language of a uniform public culture has or has not affected people’s attitudes toward different varieties of Japanese as well as their actual language practice. We analyze a variety of data consisting of metapragmatic comments about Standard and regional Japanese discourse samples drawn from situated practice.

Despite the extensive scholarship devoted to the creation of Standard Japanese as well as to the state of regional dialects, not many studies have investigated the uses and interpretations of Standard and regional Japanese in actual discourse in specific social contexts with attention to how they may be related to modern language policy and why. And this is what we attempt to do in this volume. Our analyses show how speakers in situated interaction negotiate the dominant standard language norms and use Standard and regional Japanese forms as interactional resources in an intriguingly complex manner. They also illustrate how the same linguistic forms – here, Standard or regional Japanese variant forms – may be interpreted in diverse ways, indicating that the indexical fields of these forms are potentially much broader and fluid than those based on normative interpretations. Our analyses will demonstrate the intricate and dynamic indexical process involved in the use and interpretation of both Standard and regional Japanese – a process in which the macro-level social force meets the micro-level situated interactional needs.

Part I as a whole demonstrates the complexity of the notion of Nihongo by examining both macro-level sociological issues and micro-level language phenomena and the relationship between the two. It offers a foundation for the following sections of the volume that deal with more specific aspects of the core areas of interest in Japanese sociolinguistics: honorific language and gendered language.

Part II turns to the topic of honorifics and their relation to linguistic and interactional politeness. One of the most studied sociolinguistic characteristics of “the” Japanese language is keigo ‘honorifics’. Keigo has a central place in the Japanese imaginary, both at home and abroad, and a powerful ideological hold on what it means to be a “real” member of the Japanese communitasFootnote ⁷ with a “natural” understanding of how to function within Japanese social structures. This dominant cultural ideology of honorifics is supported not only by official policy and the media, but also by many linguists. Moreover, it has been appropriated by cultural nationalists for use in characterizing “the Japanese,” in the popular nihonjinron discourses of the last quarter of the twentieth century (Befu Reference Befu2001, Yoshino Reference Yoshino1992). When the Japanese are characterized as “innately” oriented toward hierarchical relationships or as being group-centric or “innately” polite,Footnote ⁸ the supporting evidence is often drawn from keigo as indexes of such social meanings. Although the popularity of nihonjinron has waned since the early 1990s, some of these claims have proved to be highly durable and keigo continues to be put forward as evidence of their veracity. On the linguistics front as well, theoretically, honorifics have formed the core forms for thinking about speaking politely in Japan. Ide (Reference Ide2006: 91), for example, provides a succinct summary of this linkage: nihongo de wa, keigo o poraitonesu no chūshin to shite kangaete kita ‘in Japanese keigo has been considered the core of politeness’. Moreover, the complex honorific system in JapaneseFootnote ⁹ has been most prominently featured in the debate between two theoretical approaches to politeness: politeness as a matter of speaker strategy (Brown and Levinson Reference Brown and Levinson1987) and as a matter of cultural discernment of one’s place in the social structure (Ide Reference Ide1989, Reference Ide2006).

Whether it is official policy, nihonjiron, or scholarly or media discourse on honorifics, what is commonly shared is the normative view of honorifics as an essential characteristic of Japanese language and culture. Any uses of keigo that fail to match some conception of social politeness, social orientation toward vertical relationships, or toward social group orientation tend to be ignored, although these are well attested in the empirical record, which suggests a fair number of alternative indexical meanings of honorifics. Although there has been research on the construction of keigo ideology as well as research on the actual use and interpretation of honorifics, the relationship between the two has not been well examined and Part II considers this very relationship.

In order to gain a better understanding of honorifics and linguistic politeness, Part II presents a critical re-examination of honorifics by addressing: how and why honorifics came to play such a central role in Japanese linguistics/sociolinguistics, what the consequences of this centrality are for the (mis-)recognition of specific uses of honorifics (Chapter 3), and how best we can account for the diverse actual uses and interpretations of honorifics that cannot be dealt with by an approach that links honorifics straightforwardly to a politeness that is obligatory and regulated by certain singularly specified social structures (Chapter 4). We believe that addressing these questions will contribute to a better theorization of honorifics in relation to linguistic politeness cross-culturally as well as to the notion of indexicality.

In Chapter 3, we consider the macro-level social meanings of Japanese honorifics, focusing on the historical construction of dominant keigo ideology. We first address the emergence of honorifics as a key “social emblem of the Japanese nation-state” (Koyama Reference Koyama2004: 429) by examining the processes by which it came to be seen as essential that honorifics be a part of the new Standard language. Japan’s modernization was spurred by a keenly felt need to place Japan on an equal footing with the West (Anderson Reference Anderson2009: 9; LeBlanc Reference LeBlanc2012: 862). In the case of language, this required not only a standardized language, but a standardized language that was equal to or even, in some aspects, superior to the languages of the North Atlantic modern states; the existence of the complex honorific system was one answer to this need. After sketching this historical background, we turn to an examination of the ideology of keigo as it developed over the course of the twentieth century and into the present, including the process of constructing norms for “correct” use. In particular, we consider how institutional policy, authoritative pronouncements by linguistic experts, and the popular media all regiment honorific usage. Here, we are concerned to identify what have been the dominant norms, how they have shifted – or have seemed to shift – over time, and to identify moments when the complexities of the micro-interactional indexicalities have had to be worked and reworked to serve the macro-level purpose of maintaining the direct symbolic link between honorifics and a “true” Japanese identity.

Chapter 4 turns our attention to the micro-level social meanings of honorifics, or the situated use and interpretation of honorifics in relation to our findings in Chapter 3. It investigates the diversity of responses to the dominant norms expressed by speakers in their beliefs about honorific use as well as in their actual practice. We reconsider the stereotypical meanings that correspond to the dominant norms of honorific use and examine the meanings of honorifics that go beyond them. That is, we consider broader potential indexical meanings of honorifics through analyses of both native speakers’ metapragmatic discourses about honorific use and their uses of honorifics in real interactions. Through these analyses, we show that the link between honorific forms and specific, singular social meanings based on the dominant ideology of honorifics cannot be assumed as given. Forms considered to be normatively polite, for example, may be interpreted in very different ways by actual speakers. The relationship between linguistic forms and a particular social meaning in a specific context must be seen as a fluid and dynamic process of negotiated offering and interpretation by the speakers and addressees involved.

We emphasize that macro-level normative discourses, such as those described in Chapter 3, cannot account for these dynamic processes – and, indeed, it is not in their interests to do so. Despite Koyama’s assertion that it is misguided to “reduce all social phenomena, especially linguistic structure and its lexical items, including ‘registered’ ones such as honorifics, to micro-social agents” (Koyama Reference Koyama2004: 430), we must recognize that the individual Japanese speaker must, in fact, do so if she or he is to speak. In Chapter 4 we show that individual speakers consciously experience individual agency and anxiety about how to negotiate the complex, dynamically shifting micro-level indexicalities of the honorific register in appropriate and effective ways. That is, for speakers, which honorifics to choose, and when, is the focus of interest. The chapters in Part II provide a vision of how large political projects at the national level with respect to one aspect of Standard Japanese – keigo – intersect or fail to intersect with the smaller projects of everyday life (Hershatter Reference Hershatter2012: 878).

Part III addresses gendered speech, which is also a highly ideological issue. As in the cases of Standard/regional Japanese and honorifics discussed in Parts I and II, Japanese language and gender has been one of the most actively studied topics in Japanese sociolinguistics. Since the early 1990s, in language and gender research in general, there has been a major change in the approach from a static structural approach focusing on gender differences based on the binary distinction of women’s language and men’s language to a more dynamic approach interrogating within-gender language practices and their social meanings, or the discursive construction of diverse gender and sexual identities (Benwell Reference Benwell, Ehrlich, Meyerhoff and Holmes2014; Bucholtz Reference Bucholtz, Ehrlich, Meyerhoff and Holmes2014; Bucholtz and Hall Reference Bucholtz and Hall1995; Cameron Reference Cameron2005; Eckert and McConnell-Ginet Reference Eckert and McConnell-Ginet1992, Reference Eckert and McConnell-Ginet2013; see Milani Reference Milani, Ehrlich, Meyerhoff and Holmes2014, Mills and Mullany Reference Mills and Mullany2011, and Okamoto Reference Okamoto2011a for summary reviews). In line with this direction, Japanese language and gender research has shifted its focus from the “description” of the differences between joseigo, or women’s language, and danseigo, or men’s language (Ide and McGloin Reference Ide and McGloin1990; Ide and Yoshida Reference Ide, Yoshida and Tsujimura1999), to a more critical approach questioning joseigo and danseigo as linguistic norms, or Japanese cultural ideology of language and gender.

In order to go beyond a static structural approach, researchers are now increasingly investigating two main issues: norm construction (see, for example, M. Inoue Reference Inoue2006; M. Nakamura Reference Nakamura2007a, Reference Nakamura2014a) and diverse actual language practices (see, for example, Okamoto and Shibamoto Smith Reference Okamoto and Shibamoto Smith2004a). These are, however, often studied separately, and the relationship between the two has not been adequately investigated. In Part III, we attempt to address this very relationship by examining how norms have been historically constructed (Chapter 5) and how they are or are not related to the interpretation and use of specific gendered forms in specific contexts (Chapter 6).

The focus of previous research has remained largely on women’s speech norms and practices and on heternormatively oriented women at that,Footnote ¹⁰ while heterosexuality has been taken for granted in most studies. It is only recently that studies on the language of men and of non-heteronormative speakers have begun to emerge (see, for men’s speech, SturtzSreetharan 2004a, b and for the language practices of non-heteronormative speakers, Abe Reference Abe2010; Maree Reference Maree2013). In Part III, we consider why women’s speech has historically been the focus of attention and then explore the relationship between speech norms and practice concerning women and men by attempting to move beyond the simple binary of man versus woman underpinned by an assumed universally heterosexual organization of social roles to reconsider this relationship for (heterosexual) female and male speakers in this light as well as to consider the implications of heteronormativity for speakers of both normative and alternative sexualities and genders.

In Chapter 5, we discuss what the dominant gendered linguistic norms are and how they have been and continue to be constructed. In order to provide a historical background to our review of contemporary linguistic gender norms, we first review the scholarship on linguistic femininity in premodern Japan, early modern Japan, and the Meiji period, drawing attention to how language and gender norms were constructed, and to whom the norms were held to apply. We do so by reviewing dominant narratives produced by scholars, government agencies, and social commentators focused on women’s language, which has historically been the main target of regulatory control. This review will help us understand how the particular identity category of gender came to be a central object of Japanese sociolinguistic analyses, and how this has focused sociolinguistic research on particular sets of language forms, specifically, a limited set of forms found in Standard Japanese, at the expense of others.

In this review, we take up the issue of joseigo as a product of the modernization project of the Meiji government (for example, M. Inoue Reference Inoue2003, Reference Inoue2006; M. Nakamura Reference Nakamura2007a, Reference Nakamura2014a). While this research has made invaluable contributions in historicizing Japanese women’s language, we suggest that the situation is more complicated than a simple dichotomy between claims for a joseigo that, as an ideology, is a newly minted modern construct versus a vision of a joseigo that is fully continuous with past traditions of womanly behavioral norms. We argue, rather, that the joseigo of today is the result of complex processes in which the modern ideology of standard language has incorporated premodern femininity regimes, which emphasized such traits as politeness, gentleness, and refinement, thereby giving the effect of a Japan that has always been, at least with regard to femininity, civilized (Anderson Reference Anderson2009) while at the same time introducing novel language forms to index alignment with that femininity regime.

We then turn to the construction, reconstruction, dissemination, and speaker uptake of gender norms in contemporary Japan. We examine media representations of normative women’s speech as well as normative men’s speech, drawing on two data sources: the dialogue of characters in the fictional worlds of print and televisual texts, and instructions in self-help books aimed at women that link attractiveness to linguistic femininity. Another data source that is richly informative about how everyday speakers respond to dominant gender norms for language use are metapragmatic comments on femininity/masculinity and feminine/masculine speech expressed in online blogs. Through our analyses of these multiple data sources, we are able to consider the indexical processes in which the normative meanings of femininity and masculinity are attributed, as though they were given or natural, to particular linguistic forms by way of the dominant gender and language ideology.

In Chapter 6, we investigate the use and interpretation of gendered speech. While Japanese linguistic gender norms may appear to be firmly established in society, closer examination reveals highly diverse views about and uses of gendered speech, suggesting that normative meanings are contested and that the indexical fields of gendered speech forms are potentially broader and variable. We first examine the diversity in explicitly expressed attitudes about linguistic femininity and masculinity. After reviewing previous studies, we examine metapragmatic comments about linguistic femininity and masculinity posted in online blogs, contained in dialogue in print fiction, and in the results of a survey we conducted on reader responses to stereotypically feminine language in media texts. Our examination indicates that the perception or evaluation of linguistic femininity and masculinity cannot be made without reference to specific social contexts and that it is potentially variable and multiple.

The second part of Chapter 6 examines data drawn from actual speech to see how speakers deal with linguistic gender norms in specific social contexts. For this purpose, we analyze the use of a variety of linguistic forms that are stereotypically and non-stereotypically gendered and consider their possible indexical meanings. Our data here include several audio-recorded naturalistic conversations, segments of television variety shows, and several televised interviews from a popular televised talk show. Expanding the scope of “gendered” speakers in this way broadens our understanding of how the dominant norms for gendered language work to constrain and/or “aid” speakers in making linguistic choices, as they individually construct their sometimes complicated gender and sexual identities through language (see Bucholtz and Hall Reference Bucholtz and Hall2004, Reference Bucholtz and Hall2005 for a discussion of discursive construction of identity in relation to gender and sexuality).

In other words, our analyses help us to better trace the various strands of the complex relationship between Japanese language and gender in situated practice, wherein speakers negotiate what they believe to be linguistic gender norms (which may not necessarily agree with the dominant norms but be more situation-specific) with what they want to say. As noted by Cameron (2014: 294), sociolinguists in recent years have tended to emphasize speaker agency in using linguistic resources to construct identities, but it is important to remember that speakers are not entirely free agents, but are rather constrained, to a greater or lesser degree, by cultural norms and expectations. Our data demonstrate that while speakers are in fact highly affected by linguistic gender norms, there is also wide inter- and intra-speaker variation in the way these norms are negotiated, which leads them to use gendered linguistic forms in a creatively intricate manner. That is, we attempt to unlock real Japanese women and men from the constraints of linguistic norms and expectations in order to tease out the relations between stereotypical characterizations and real people trying to make a particular gendered impression or one that gestures toward a speaker’s sexuality.

In each of the three parts, focusing on the relationship between the ideologies underlying dominant societal linguistic norms and diverse language attitudes and real practices in specific social contexts, we emphasize how together these create a portrait of variation as an indexical system of social meanings (Eckert Reference Eckert2008). The overall theme that binds the separate chapters of the book is not the issue of dominant versus subversive language use or simply a restatement of the difference between language at the ideological level versus language at the level of real interaction. Rather, we see our contribution as centering on questions of the interplay of history and sociopolitical interest, ideologies, stances, and actual language uses as manifested in the linguistic practices of individual speakers. We argue that in order to fully understand the sociolinguistic life of Japanese, we need to disentangle the various strands of cultural subjectivity and individual agency that form the complex relationship between language ideologies and practices. Our goal here is not to offer a comprehensive description of the language practices of speakers in all segments of Japanese society, but to illustrate our approach to the relationship between norms and practice, drawing on an indexical view of language that is explicitly founded in an understanding of the sociocultural “surround” of everyday interaction.

Japanese sociolinguistic practices have before now stood somewhat on the sidelines of many key sociolinguistic debates, as interesting but somehow still “exceptional” cases among modern/postmodern societies and the sociolinguistic theories that have developed to account for language practices within them. Throughout the volume, we stress the parallels and disjunctures with the concerns of Western sociolinguistics. By clarifying some hitherto understudied aspects of Japanese sociolinguistic practice, we also hope to contribute to reducing the gap between Japanese and Western sociolinguistic scholarship and to incorporate Japan more fully into contemporary sociolinguistic debates.