So now, we are at the end of our journey through the development of three focal areas – the three most persistently focal areas – of Japanese sociolinguistic inquiry. Tracing the historical development of these areas of inquiry with an eye to the social contexts which pointed both scholars and policymakers toward them in preference to others has helped us understand how we ourselves got to be where we are and sharpened our thinking about where we would like to head from here. We have benefited a great deal from pulling these “three historicizations” together and are delighted to have the opportunity to share them with you.

The centrality of issues concerning regional dialects to sociolinguistic questions rather than the more historically oriented questions of dialectology sets Japanese approaches to sociolinguistics somewhat apart from Western sociolinguistics, which has focused more on social divisions such as class and race or ethnicity. The same can be said for the centrality of honorifics to the expression of politeness to the virtual exclusion of other ways of expressing oneself politely (although see Dunn Reference Dunn2011). Whereas Western models of politeness such as Brown and Levinson’s proposed “universal” model of politeness include consideration of all manner of lexical, syntactic, and discursive constructions as they contribute to politeness, most models of politeness emerging from Japan, including the influential wakimae riron ‘discernment theory’ (Ide Reference Ide1989, Reference Ide2006),Footnote ¹ present a largely honorifics-as-politeness message. Similarly, with gendered language we find that the greatest amount of attention in Japan is given to the lexicon and morphology of gendered expression rather than the discursive properties of the interactional performance of a gendered identity.

We hope that our attempts to offer a sociohistorical contextualization for the choice of topics has helped make sense of why they are what they are and perhaps even why the classic social class versus gender matrix, or the race/ethnicityFootnote ² versus standard language matrix – both central to Western sociolinguistic concerns – did not rise to prominence in the Japanese sociolinguistic field. As a late modernizer, albeit a very early late one, Japan was saddled with being both as good (civilized) as and yet different from the powers with whom they aspired to compete. The selection of this particular set of topics (dialect, honorifics, “women’s” and “men’s” language) all, we have suggested, developed in a profoundly social and political context on an international scale and their selection was a very conscious one at that. By the time a formal field of sociolinguistics appeared in Japan (in 1948 in the form of gengo seikatsu), the topics were well established and well established as social issues, at that. It seems only natural that they would continue to absorb the attention of linguists with specific interests in the social side of language form and use.

And beyond the motivations for these three sociolinguistic phenomena to be the core foci of attention in the context of Japanese modernity, we also see how closely intertwined with each other they are in terms of linguistic forms. Standardization was partly aimed at unification of the citizenry in order to modernize the nation, and the linguistic norms for all three phenomena are based in the standard language as it developed in the early years of the twentieth century. This was a moment when diverse forms of Japanese were being re-evaluated, or resemiotized, in the national context, at the end of which process Standard Japanese became the superordinate language of the nation. Standard language ideology was, then, the variety within which models of “correct” honorifics and “appropriately” feminine or masculine gendered speech were embedded. These complex interrelationships, as we have seen in this study, have had considerable influence on both people’s language attitudes and their everyday practice.

And it was seductively convenient, we began to recognize, that the differences between, for example, a plain way of saying something like he’s coming at nine is different from a polite way of saying that he is, and also different from saying so when the referent “he” is someone deserving of deference (kuji ni kuru versus kuji ni kimasu versus kuji ni irasshaimasu). All the differences are clearly encoded in the surface-segmentable aspects of the three utterances. Equally, the differences between how a “feminine” woman would say “I’m going” and how a “masculine” man would are again to be found in the surface-segmentable forms normatively ascribed to the respective speakers ([atashi] iku wa yo versus [ore] iku zo). These are precisely the kinds of forms that possess the unavoidable referentiality, the continuous segmentability, and the relatively presupposing qualities of linguistic forms that facilitate metapragmatic awareness (Silverstein Reference Silverstein1981). These are the forms implicated in differentiating dialect from Standard, the honorific-using speaker from those hopelessly confused speakers who can’t be polite because they can’t use “correct” keigo, and the women from the men, forms that virtually shout to the analyst, “Here we are.”

This, we suggest, has led analysts, and continues to lead them, to concentrate on these forms without, in many cases, looking too hard for other ways of orienting to a national identity, of being polite, or of gendering onself. The consequence has been to frame sociolingistic work largely within models wherein the forms in question exist in a kind of presuppositional, hence direct, indexicalityFootnote ³ with some singular social meaning. We ourselves have also focused on these three sets of forms. They are hard to ignore. However, the trajectory in this volume has, we hoped, interrupted that vision of the direct indexicalities of these sets of forms and offered an alternative and more dynamic approach to them.

Our model of polyindexical meanings construed within a fluid indexical field drew on examples, both from our previously published work and new ones, that could plausibly fit within the normative models of one kind of speech or another. The well-known use of first-person “masculine” pronoun ore by Tōhoku dialect-speaking women (Chapter 2) or its perhaps less well-known use by lesbians in Shinjuku 2-chōme bars (Abe Reference Abe, Okamoto and Shibamoto Smith2004) can, certainly, be invoked to suggest something masculine about these ladies. But as we have tried to demonstrate, that kind of direct indexical relationship simply fails to account for what is going on in everyday interactions happening in a myriad of everyday settings in Japan.

We have tried, through both argumentation and exemplification, to show that a more complex – and often creative rather than presupposed – set of potential indexicalities, or “constellation of ideologically related meanings” (Eckert Reference Eckert2008: 453), inheres to all the forms in the socially charged sets we have described in the various parts of this volume. And that these floating potential indexicalities are indeterminate both as used by speakers and as taken up for interpretation by their interlocutors or audiences. We hope our readers are convinced and that the future will bring much more research aimed at teasing out the interactional ebb and flow of utterance and interpretation as speakers negotiate the indexical field of fluid semantic meaningfulness.

One of the aspects of this that excited us the most to find, and to share with our readers, is that the indeterminacy operates not just at the level of speaking and hearing/interpreting, but also at the ideological level. Both in scholarly survey instruments and popular blog posts, individuals offered their reactions to particular uses of Standard Japanese, regional dialects, honorifics, and gendered forms. As we saw, in all of the metapragmatic comments we examined, there is not close agreement about any of the normative prescripts for “correct” use. As visible as this is in survey results, the blogs proved to be an even richer source of information about the range of stancesFootnote ⁴ toward normative constraints on (or suggestions for) use, as bloggers not only express their opinion but offer extended examples of what they do and do not think to be good language along with reasons for their thinking. We have had a great deal of fun listening in to the debates among “the” Japanese public as they share their views about language and its role in making interpersonal relations run smoothly (or not so smoothly), and we have learned a great deal about the slippage between norm and practice from them. It helped us look at the empirical data from naturalistic conversations and emails with new eyes, and helped us to see when stereotypified norms were drawn upon to align with the normative and when they were subverted or simply ignored – by putting a gendered form to a non-gendered use, by using dialect in an email to bring recipients’ attention back to a renewed sense of what they once had had in common (nostalgia), by utilizing honorifics to be pretentious and even rude rather than polite. We came to appreciate the delicacy of speakers’ use of a multiplicity of linguistic resources to forge a particular style in specific contexts.

There is much more to learn from real speakers, we conclude, if we are ready to listen for the normative without assuming alignment with norms and to listen for the structures that stray from the norms without assuming the opposite. And, unsurprisingly perhaps, since we had been helped by surveys and by the more personal posts on language-related blogs to have diversity in mind, we began to hear some other ways in which speakers send messages concerning their negotiations with modern cosmopolitanism (or its rejection), politeness, and “gender.”Footnote ⁵ Ways, that is, outside the stereotypified forms of each of these categories but that interact with them to modulate their meanings (or the meanings of their absence); it is to these important but understudied constituents of verbal presentation that we now turn, in a spirit of hopefulness that we will, in future research, be able to enrich our understanding of how people, inherently socially organized actors as pointed out by Voloshinov (Reference Voloshinov1986/1929) so long ago, enact their immediate, contextualized sociality in and through language. We offer, then, some thoughts about possible ways to go forward with that goal in mind.

As we worked through the three topics that comprise the core of this volume, we at times found ourselves wishing for information that simply was not there. The issue of class, for example, came up in our examination of all three topics, particularly with respect to norm construction, whether it be in the prescripts to use Standard Japanese – or in the potential consequences of not doing so in terms of social prestige,Footnote ⁶ the linkage of honorifics not only to politeness but to the social marking of the “educated” or refined and “classy” speaker, or gendered speech, in particular the association of women’s speech to class status and refinement. In Japanese sociolinguistics, social class has been largely avoided as a factor in language use and in language disadvantage (prejudice), as noted in Chapter 1. Some hints of class can be garnered from surveys that list the occupations of respondents, but not all do. Class differences can be read into studies on honorific language such as the work contrasting the honorific uses of department store salespeople and market vendors in Kyōto and Ōsaka reported in Okamoto (Reference Okamoto1997, 1998) or in the differences in gendered language found in conversations between Shino, the Kyōto geisha, and Takako, the young Ōsaka woman trying to become more “lovable” by imitating her style, from the drama Koisuru Kyōto mentioned in Chapter 3 (and reported on in Okamoto and Shibamoto Smith Reference Okamoto and Shibamoto Smith2008). But the intersection of social class with language norms is rarely tackled directly. But as our study of the three topics suggests, it has importance for Japanese language life, and more focused research is called for.

Here, we should hasten to add, we are not suggesting this in order to bring Japanese sociolinguistics more into line with Western sociolinguistic interests; a more focused look at social class and language advantage or disadvantage would, of course, accomplish that. That is not, however, our aim. What we are rather urging is that sociolinguistics follow the larger shift in thinking about the Japanese social landscape. The narrative of Japan as a homogeneous (and middle-class) society always understated class-based disparities in lifeways and life possibilities. But today that narrative has given way to a large-scale social recognition of Japan as a kakusa shakai ‘disparity society’. Educators write openly about disparities in educational outcomes correlated with the class background of students (see, for example, Kariya Reference Takehiko, Ishida and Slater2010; T. Nakamura Reference Nakamura2003). In our own research and in terms of research on social inequalities in general, we would hope to see more explicit sociolinguistic attention to this potentially powerfully shaping aspect of language practice.

Another arena in which we had predicted there would be a lack of research – and which we found to be at least in part the case – was the arena of ethnicity. Domestic narratives of Japanese identity had, from the post-WWII decades until the late 1990s, systematically occluded, or denied, the archipelago’s multi-ethnic nature. We Japanese, claimed nihonjinron theorists, are a homogeneous group. This discourse has given way to increased recognition of a “multi-ethnic” Japan (see, for example, Graburn et al. Reference Graburn, Ertl and Tierney2008; Lee et al. Reference Lee, Murphy-Shigematsu and Befu2006; Weiner Reference Weiner1997; and Willis and Murphy-Shigematsu Reference Willis and Murphy-Shigematsu2008), but this recognition has not, to date, resulted in much sociolinguistic attention being paid to the everyday interactions of ethnic groups with either an intra- or inter-group focus. We especially miss attention to the latter, where, for example, the worlds of “the homogeneous Japanese” intersect with ethnic difference in such venues as in schools, the workplace, or in other public settings. To be sure, as we have noted earlier, there is a fair amount of sociolinguistic research on Ryūkyūan language speakers or semi-speakers, some with respect to how their Japanese is affected by their Okinawan roots. But we would hope in the future to see (or help produce) more knowledge about how these speakers’ language use impacts their participation in the language life of “the Japan” (an early model for which would be Nakamoto Reference Nakamoto1982) and how or whether it forecloses full participation in that life in other spheres of activity than simply speaking.

The same thing could be wished for with respect to other ethnic groups as well. There is a small but growing number of papers on non-native speakers in Japan and the differences between the non-native and native Japanese speaker practices, but there was very little we were able to find on the language attitudes and practices of internal “ethnic” minority groups that are, at this point in time, also native speakers of Japanese. These include zainichi Koreans or Chinese communities, the mixed offspring of international marriages of farming men to foreign brides in places like Furano, and the like. These and others contribute a great deal to the construction of today’s Japan and its language life, but sociolinguists have largely left them out of their investigations.

And finally, we come to the question of gendered language. Unlike the previous two areas, relatively neglected in Japanese sociolinguistics, gendered language has been a persistent focus of research. But we have begun to feel, as we worked through the various issues in this volume, that the focus has been in some ways highly restricted, and has thus limited what contribution can be made to understanding the lifeways of real women and real men making their way through the social world. We have discussed, for example, how deep-rooted heteronormativity is in Japanese language life, although work on lesbian and gay speakers is increasing. We touched here upon the use and interpretation of speakers of alternative sexual orientations, but this is one area that requires much more work, as it will help us to understand how heteronormative gender stereotypes relate – or fail to relate – to individual speakers. We see particular gaps here in studies of the language practice of such speakers in concrete interactions in diverse social situations, where sexual orientation might arguably be less salient than in the bars of Shinjuku Ni-chōme, LGBT activist work, or media performances. Attitudinal studies focused on the language use of speakers of alternative sexual orientations is also a critical component of research, as it allows us glimpses of the ways in which diversity in the ideological frames of interlocutor uptake meet with diversity in speakers’ verbal offerings.

Particularly in terms of “gendered” language and the use of honorifics-as-politeness forms, but also in the analysis of distinctions between Standard Japanese and dialect, we see a future need to go beyond the surface-segmentable markers of keigo, joseigo, and danseigo, or lexico-morphological markers of dialect, to look at how politeness, gender, and locality are encoded by discourse strategies as much as by those surface markers. That there are such discourse strategies to mark politeness was well-demonstrated in a study of employee training courses for businesses, where training in honorifics was coupled with pragmatic and discursive strategies such as speaking clearly and projecting your voice well, rephrasing directives as questions, adding an apology to a negative response, and phrasing that negative utterance with affirmative verb forms, using, for example, wakarikanemasu with the positive verb final -masu instead of the negative verb final -masen in wakarimasen ‘I don’t understand’ (Dunn Reference Dunn2011). We too found similar discursive instructions in the conduct manuals for women (see Chapter 4). We have also noted but did not elaborate on the role of discursive rather than structural (and stereotypified) linguistic forms at the intersection of politeness, gender, and dialect when we pointed out the intersection of politeness (with and without honorifics), gender (with or without Standard Japanese-based joseigo forms), and dialect in the speech of our two elderly Tōhoku women in Chapter 4. We will surely be taking some of these discursive questions up in the future.

In this volume, we examined conversations between young women and between young men who presented us with mixtures of features that included the stereotypified features of joseigo and danseigo, but also some features that have been found in other languages to be associated with femininity or masculinity, such as hedges, facilitative speech acts, and the like. These should certainly be incorporated into an understanding of gendered Japanese along with the highly salient (in part due to their persistent enregisterment and re-enregisterment in the scholarly literature) pronominal series, sentence-final particles, and relative differences in keigo use. Other aspects of language that might be incorporated in future studies are patterns of grammatical particle deletion (Shibamoto Reference Shibamoto, McGloin and Ide1990), choice of evidentials and other modal forms (Shibamoto Smith Reference Shibamoto Smith1992), and the like. These would bring Japanese language and gender research into closer conversation with language and gender research on Western languages, languages that do not have the highly salient surface-segmentable pronominal, sentence-final particle, and keigo forms. At the same time, it would open up an avenue for comparing linguistic resources for the making of social meaning across languages and for understanding the linguistic performance of gender (or politeness or locality) in a more nuanced and richer fashion.

In sum, then, we hope for the future a refreshed inspiration, a new sense of possibility, and a look forward into enhanced understandings of Japanese real diversity, as that real diversity is beginning to slowly to find its way into Japan’s twenty-first century national discourses of self-representation.