Corps de l’article

This text emerged out of a series of events to which I was invited. The first occurred in October 2022, when I was invited by the Anthropology section of the New York Academy of Sciences to participate in their annual themed talk series. In these series, representatives of the North American tradition of four-field anthropology are asked to address a common theme; here, I was asked as a linguistic anthropologist to address the theme of “crisis.” That experience stayed with me, given the increasing prominence of crises and feelings and discourses of crisis, even more strongly perhaps with us as I write at the end of 2023. Certainly, it was foremost on my mind when I was awarded an honorary doctorate in February 2023 by the Universitat Autonoma de Barcelona,[1] and asked to address an audience at the ceremony composed of academic administrators and academics from the humanities and education; in many ways, I took my job to be to speak to what scholars of language(s) and scholars who work with language(s) might have to say about the prevalent sense not simply of crisis, but of its manifestation in an unease about how to understand what might count as true, or at least true enough to be taken seriously.

That honorary doctorate emerged out of decades of conversation with Catalan scholars about the links between my own work in francophone Canada and theirs in Catalunya about language, nation and State, and in particular about the role of language in the making of social difference and social inequality, and—perhaps most importantly, about how and why social, economic and political struggles take place on the terrain of language.

In fact, in many ways, I had first visited Barcelona in the late 1970s, when I myself was a graduate student and a stagiaire—an intern—in the sociolinguistic research service of the Office de la langue française in Québec. This agency, now the Office québécois de la langue française, was a hub of activity around minority language policy and planning, with antennae out for similar things happening elsewhere. The links between Catalunya and Québec were evident, and there was a great deal of sharing of ideas and analyses between the two (for example, the concepts of normativizació and normalizació as key dimensions of minority language nation-building show up clearly in Québec language policy and legislation from the 1960s to today). Both Québec and Catalunya were at that time in the midst of political crises connected to the legitimacy of the state of which they were a part, and, I would argue now, more broadly connected to a post-World War II crisis of the state in which the meaning and promises of liberal democracy played a key role.

Indeed, when I first stepped into the streets of Barcelona, I felt I was back in Montréal (albeit with better weather). Every interaction was just as fraught with the question of what language to speak. Tensions between French and English (or, more exactly, minoritized French speakers and dominant English speakers) had their (however inexact) counterpart in tensions between speakers of Catalan and speakers of Castilian. I was impressed, though, with the taxi drivers’ solution of hanging a Lliure/Libre sign out; in Montréal you had to wait until you figured out what radio station the driver was playing before taking a stab at opening your mouth.

It is not surprising then that I found in Barcelona a network of scholars with whom I could get right into the meat of the matter, and with whom I shared a passion for understanding and explaining what we were living, in order to live it in accordance with our values—essentially within a commitment to social justice.[2] Together with other scholars in similar situations, we worked on developing theoretical and methodological tools which, over the years, have crystallized into an approach we think of as critical ethnographic sociolinguistics (Heller et al. 2017; Heller 2023).

At the heart of this way of thinking is the broad recognition of the importance of language in social struggle, and its centrality to the making of social difference and social inequality. The world we live in has, unfortunately, no shortage of those. Indeed, the current moment invites us to reflect on what has become—is becoming—of the crisis of the state and of liberal democracy that brought us together in the first place.

***

Let me begin by walking backwards into the future, as the literary critic Raymond Williams (1989) had it. As my colleague Bonnie McElhinny (2017:2) and I put it, we can take this to mean that imagining a better future requires revisiting the past in new ways. I try here, therefore, to link moments of crisis past and present in a critical ethnographic sociolinguistic frame, both in order to better understand what we are experiencing now, and also to find some ways to lay down tracks for ways of imagining what equalities and solidarities we can build on the terrain of language.

I will start with Germany in the 1920s, a period often compared to ours. Indeed, my mother, a German Jewish refugee from Nazism, said repeatedly at the end of her life during the Trump presidency, “I started my life with Hitler and I am ending it with Trump—not much progress.” I take this as a comment not only on the recurrence of fascist demagogues but also on the false promise of liberal democratic modernity to provide linear improvement of social conditions—one piece of the past which requires understanding differently in order to imagine ourselves otherwise. Maybe linear progress is less objective fact than ideologically embedded chronotope, that is, perhaps it is not an inevitable way to understand time and social change, but rather one which is deeply embedded in the ideologies of modernity, and which serves particular political economic purposes which made sense at the historical juncture of the nineteenth and twentieth century, but may not make so much sense anymore.

Within my chosen entry point of the Germany of the 1920s, I will start with Victor Klemperer, a Romance philologist who was named Professor at the Technische Universität Dresden precisely in 1920. He had been keeping a diary since 1899, when he was 16 years old (Aubry and Turpin 2012), and simply kept on doing so. He was in many ways fairly typical of German Jews of that time, heavily influenced by the Enlightenment philosophy of Moses Mendelssohn, oriented towards German culture and identity, and scarcely observant (another reason I focus on him is that he resembles in this way my mother’s family, with her father and his siblings all named after Wagner or one of his heroes or heroines). Indeed, Klemperer converted to Protestantism in 1912. Like my grandfather and great-uncles, he served Prussia in World War I. Unlike my grandfather and his brothers (but exactly like my grandfather’s best friend), he also married an “Aryan,” and her willingness to stick by him, together with his conversion and war service, allowed him to survive, albeit in increasingly oppressive circumstances, up until he was miraculously saved on the eve of his deportation to Auschwitz by the bombing of Dresden. He returned to his post in Dresden after the war and died there in 1960.

Albeit a specialist in eighteenth and nineteenth-century French literature, as a philologist, he was closely attentive to detailed features of language—not only in the French literature he studied, but also in the German of his daily life. In 1933, this “déformation professionnelle” as we might call it, led him to begin systematically noting the ways in which he heard and read German being used in new, unfamiliar, and—given the times—unsettling ways. A selection of his notes was published first in the German Democratic Republic in 1947 as Lingua Tertii Imperii: Notizbuch eines Philologen (Language of the Third Reich: Notebook of a Philologist—note the Latin in the first part of the title, and then the switch to German for the second). He noted for example how some words already in circulation changed valence or frequency (we will discuss, among others, the example of the word “fanatic” in a moment), how practices like adjectival modification associated adjectives and nouns in new ways, or how punctuation in writing (such as the use of the exclamation point) or pitch and volume in speech were associated in new ways to these new practices. However, the West was largely ignorant of this work until the full notebooks covering the years 1933–1945 were published starting in 1991, and subsequently widely translated.

Klemperer recorded a number of types of shifts that disturbed him (see Heller and McElhinny 2017: 153–154), that is, that he saw as signalling a struggle over what would count as discursive authority. He saw in this struggle a fascist repudiation of the regime of truth of liberal democracy, and a recasting of the voice of the state in terms of an interpretative frame that valued patriarchy, hierarchy, authoritarianism and accompanying “structures of feeling” (Williams again; Williams [1976] 1983) centring masculinity, action, and militantism.

The canonical example is perhaps his attention to the word “fanatic,” a major example of many other forms of appropriation and shift in valence of words that were already in regular use, and which, he demonstrated, were being used in a positive way to value strong, focused—and unquestioning—commitment to an idea (contrast our common sense understanding of a fanatic as someone impermeable to reason). Another was to use particularly ideologically charged words to modify other key ones: here, the clearest example is perhaps the word “Volk” to modify nouns like “festival” or “community” or adverbs like “near.” He also noted a rise in the frequency of acronyms (for example, “knif” for “kommt nicht in Frage”—it does not get called into question), as a means of underscoring both the value of discursive efficiency and of in-group belonging.

Klemperer also noted discursive efforts to define and delimit who could engage in such linguistic practice, that is, who had the right to rework how it would be considered proper and valuable to speak and write, and, indeed, who had the right to speak or write at all. This is detailed in the much later work of Christopher Hutton (1999): for example, Hutton (300) notes a key passage in Hitler’s Mein Kampf which distinguished between languages understood as “mother tongues” and which by definition spoke the truth, and others, like the constructed international auxiliary language Esperanto, which were necessarily vehicles for lies. Similarly, people could speak the truth only in their mother tongue, and people who were constructed as having no mother tongue, like Jews, were understood to be unable to speak the truth even in a mother tongue language, like German. As a result, of course, Jews could not really be German either.

Here is the passage (in Hutton’s translation):

Among them is the lie with regard to the language of the Jew. For him it is not a means of expressing his thoughts, but a means for concealing them. When he speaks French, he thinks Jewish, and while he turns out German verses, in his life he only expresses the nature of his nationality. As long as the Jew has not become the master of other peoples, he must speak their language whether he likes it or not, but as soon as they became his slaves, they would all have to learn a universal language (Esperanto, for instance!) so that by this additional means the Jews could more easily dominate them

Hitler [1925/6] 1992, cited in Hutton 1999:300

Dorinda Outram (1987), a British social historian, noted similar phenomena at another time of crisis, the French Revolution. In this case, revolutionaries delegitimated the monarchy and aristocracy as “decadent,” as evidenced by their “flowery language” and their verbosity, both of which were also understood as signs of the undue influence of women. A true revolutionary needed to speak “le langage mâle de la vertu”—the male language of virtue, with “vertu” understood as the opposite of “decadence.” This masculine language, like that of the Nazis, had to be direct, brief and strong, and it could only legitimately be valued as such if uttered or written by a man. Women and the aristocracy could not, by virtue of their embodied social positions, speak in the new way the revolutionaries were using to advance their strategies of violent disruption, and to link their political cause to specific forms of social and moral order.

If I raise these examples, it is to introduce a few reflections on what language has to do with social, political and economic crises. The first is to note the unsettling feeling that taken-for-granted ways of using language are called into question: things do not mean what they used to, or, to use a concept from Pierre Bourdieu, what counts as legitimate language is destabilized (Bourdieu 1977). To take Bourdieu’s formulations further, the same is true of who can count as a legitimate speaker or hearer or what can count as a legitimate discursive space. One example, from Outram: Manon Roland, the intellectually influential wife of the Parisian revolutionary Jean-Marie Roland could only exercise her influence through her husband; she could hold a salon, but she could do no more than listen to the discussion while she knitted. Another, a bit closer to home: my mother recalled frequently that in the Berlin of 1933 it took only three weeks for her schoolmates to go from ridiculing the radio broadcast speeches they were forced to listen to – standing at attention in the gymnasium – to taking them as fully authoritative, and therefore accepting as necessary the exclusion of my mother and her siblings, first from birthday parties, then from the school, then from Germany, and then—for those who had not found a way out as my mother’s family did—from life.

Of course, at the same time, in both these places and in both these periods, there were other discourses circulating and emerging which competed for authority in the service of remaking the world that was coming apart. For Europe in the late nineteenth and early twentieth centuries, we have the backdrop of imperialism and colonialism in tension with the rise and consolidation of the liberal democratic nation-state as the political and economic foundation of the global order, with competing claims to nation-statehood, competition among nation-states, and tensions over inequality between metropole and colony, and both within and across nation-states. These tensions manifested in a number of ways—to reductively gloss them I will evoke a simple schema of various ways of imagining or resisting modernity, from the fascist movements I have mentioned to other forms of conservativism (including the relics of monarchism still alive today), and efforts to construct some kind of egalitarianism, some more oriented to liberal democracy and others to Communism and anarchism.

Language was—and remains—important to all those movements, in a variety of forms. The best-known, of course, is the intensive work done at the time in inventing and inculcating the standard languages and standard language literacy that was key to the making of the centralized market of the capitalist nation-state (Anderson 1983; De Certeau et al. [1975] 2002; Grillo 1989; Hobsbawm 1990). But this move necessarily created communication barriers between nation-states (a hindrance to commerce from a liberal perspective). It also underscored inequalities among states as well as within them, to the extent to which access to the standard language and the making of its rules was necessarily unequally distributed. So we find for example the Austro-Hungarian Empire attempting to handle emergent nationalisms, with their competing claims precisely to nation-state type language standardization in order to gain some control of regional markets that had been subsumed under empire—this is the formation of the so-called minority language movements, triggering endless debates about what is a dialect and what is a language, and more than a century of struggle over how states can (or cannot) manage linguistic diversity.

One response to these problems takes the form of International Auxiliary Languages, the best known of which is certainly the aforementioned Esperanto—these are deliberately constructed languages meant to be easily acquired as an additional (“auxiliary”) language in order to facilitate communication among speakers of different languages. While they have an extremely long history connected to philosophical and theological explorations of language, there was a concentration of work on them in Europe between about 1875 (with the invention of an IAL called Volapük) and 1920 (Okrent 2009), with a resurgence in the 21st century. Esperanto itself was first published in 1887 by a Jewish eye doctor in Tsarist Bialystok. As signalled in the name, its creator, and eventually the many people who took up the language, Esperanto (and other IALs) are often invested with various kinds of hope, for example, hopes of achieving mutual understanding, world peace, and for some, radical redistribution of wealth. Indeed, the Nazis sought to violently suppress Esperanto, they found it so threatening, and eventually, having once looked to Esperanto for international socialism, the USSR suppressed it once it had turned instead to Russification (Lins 2020). Other approaches to the problem included deep attention to translation among national standard languages, or to orthographic and grammatical simplification in order to facilitate access to standard languages (for example, the Basic English movement).

But we can also point to more everyday practices, such as the opening up of discursive spaces (especially to women) for democratic deliberation; dehierarchization of address terms (think “comrade”); and efforts at establishing turn-taking rules that would allow greater numbers of people to participate in a political discussion (the Occupy movement, to which I return later, is a more recent example of efforts in this direction).

Klemperer believed that one could scientifically identify “toxic” language that “poisoned” how people think. I do not agree that things are that simple, partly because I do not think language influences thought in such a direct way, but also because I believe crisis is not just about some people deciding they want to change the rules of the game in order to gain an advantage. Rather it is about the failure of a particular social order to deliver on what it promises—most centrally, at least some kind of hope for access to resources, be they material or symbolic. Under those conditions (and it is in the nature of capitalism to cyclically produce unsustainable gaps in wealth), the legitimating discourses of existing relations of power lose authority or credibility; we no longer know what counts as something to be believed, or who counts as someone to be trusted.

The result is that there are openings for competing discourses, and their concocters and proponents must struggle to establish their authority and credibility. That is a matter of language being used not only institutionally—harnessing, for example, the control over discursive space that religious institutions and states generally have—but also, as Klemperer and Outram pointed out, it is a matter of interaction in everyday life. On the policing side, someone ridicules someone else’s flowery language, or their inability to use the term “fanatic” properly and, on the rewarding side, someone celebrates and disseminates an inventive sharp acronym or delivers a particularly morally correct impassioned speech to those gathered at the salon or in the school gymnasia of the country.

We also need to attend to the link between linguistic forms and practices and the moral values underlying particular regimes of truth. This is what these days we would call language ideologies—the things that shape our recognition, say, that fascism, fanaticism, monologues, loudness, the radio and acronyms go together, and that establish over large populations the authority of particular actors to act in particular ways and to get others to do the same. We need to explain why and how these get linked, and how they get inflected in ways which use all kinds of social categorizations (mainly class, race, gender, and sexuality) in the making of social differences, in turn in the service of making or resisting the inequalities that are at the centre of the crises of the state and of capitalism we have discussed.

I think we can see echoes of these earlier assemblages in contemporary experience. To give just two examples (to which I will return), the United States has recently given us locutions such as “fake news” or “wokeness,” both of which evoke a destabilization of our frames of reference (being “woke” initially signalled a critical awareness of structural inequality; the term has been appropriated to index ideologically-driven—and therefore illegitimate—social critique). We are having a hard time figuring out how to tell who is who, and whom to believe about what is going on, let alone what to do about it. How does this relate to the interrelated key features of the contemporary crisis, notably increasing wealth gaps and climate change?

Let us start with the legitimating discourses of the institutions which have held together the contradiction of capitalism, by which I mean its reliance on inequality versus its promises of universal access to the wealth it generates. A key feature of modernity has been the relative success of the secular state in wresting control of doing this reconciliation (some might call it mystification or masking) from religious and aristocratic authorities, but, obviously, not everywhere, and not always. We have inherited an uneasy alliance (closer in some places than others), between state and religious institutions over what constitutes the moral order and who is responsible for delivering what in the way of access to resources. Uneasy alliances always leave interstices for promoting and resisting alternative discourses.

We can perhaps understand this as a tension between the two main forms of discursive legitimation of the nation-state, which operate in tandem, albeit in tension with each other (see Bauman and Briggs 2003). The first we can think of as Enlightenment modes of understanding, anchored in rationality and technological approaches to living, claiming authority based on universal principles of science; the second can be thought of as Romantic ones, which insist instead on nature and affect as sources of authority because they are held to embody authenticity, and which recognize inherent differences. Together, they introduce into the democratizing promises of liberalism the possibility of hierarchy, and so operate to legitimize the persistence of inequality. They do so by neutralizing the contradiction between promises of equal access to the resources of the modern nation-state and the inequalities of gender, race and class, concentrating power into the hands of a few (white, male, bourgeois metropolitans).

These two discursive formations play out in the key institutions of the state that are designed to make and reproduce citizenship. Education, health care, the legal system and the media play dominant roles in constructing what counts as citizenship and who counts as citizens, with attention to presenting citizenship as universal while practicing social selection. Indeed, as Bourdieu notably has pointed out, the function of these institutions includes masking the social bias of selection built into them (Bourdieu 1982). Someone decides what counts as appropriate ways to act as a citizen, and not everyone is equally well-placed to be able to learn them. Even if they do, the definers of the moral and social order can always change the rules again.

Constructing state authority over citizenship (understood as legitimate claims to participation in the discursive spaces of the state and access to the resources it helps distribute), is, as I have tried to argue, centrally a communicative matter—first you set up the spaces where regimes of truth are produced and reproduced, then you set up criteria and processes of exclusion and inclusion which are inherently connected to the regimes of truth being constructed, then you double down on teaching people how to perform membership (although, as Bourdieu showed, in many ways you already have to know how to provide this performance in order to have access and to be admitted in the first place; Bourdieu 1982). But these are also spaces of distribution of resources, whether material or symbolic (and of course, these are interchangeable—you can not get a job easily without an education, or hold one down without your health). For things to remain stable, we have to believe that participation in these spaces, acting in the ways these spaces expect us to act, will facilitate our access to wealth generated by whatever economic processes produce it.

In that sense, part of the destabilization we are experiencing is manifested in the lack of trust that this contract can hold, and in refusals to play the game by heretofore routinely accepted rules. The idea of “fake news” attacks the media as an authoritative source of accounts of what is going on, and delegitimates its modes of reporting as constructions of authoritative accounts. The idea of “wokeness” attacks higher education in the same way—as I explained earlier, the term “woke” was originally used to describe people awake to systemic inequalities of class, gender and race, but has been taken over by conservatives to argue that critiques of such forms of inequality have no empirical foundation and are blind to individual difference, individual responsibility and individual potential. Much like the term “politically correct” a few years ago, the concept of “wokeness” has been taken up as a weapon in US culture wars, especially aimed at attacking secondary and post-secondary education as hotbeds of radicalism (not unlike earlier McCarthyite concerns about universities as incubators for Communists).

Fights over the nature of principles of social organization, over the nature or even the existence of social differences and social inequality such as race, gender and class, reveal major cracks in discursive procedures for establishing shared frames of reference. Such shared frames are necessary for the most basic kinds of social and political relations, from agreeing on what it is we are fighting about to finding a way to do something about the conflict that might avert physical violence.

At the same time, the contemporary world is full of efforts to think and do otherwise, many of them, of course, drawing on long-standing discourses from anarchism to science fiction. I referred earlier in passing to one example, the Occupy movement, and its deliberate attention to the vexed matter of turn-taking. Turn-taking—the question of who gets to speak when in a conversation with more than one participant (because otherwise it is a monologue, not a conversation)—has been a focus of work in interactionist sociolinguistics and linguistic anthropology, and in ethnomethodological sociology for at least 50 years. The question is how do we even know how to do this interaction thing, and what difference does it make to do it one way or a different way? From an ethnomethodological perspective, this is one way to understand membership—interaction can only happen if we use the same, or at least similar, rules. That allows us both to signal that we are on the same planet and to together reproduce that planet, build a new version of it, or build a new one altogether. (An experiment you can try to test this out is to refuse to say anything the next time you answer the phone. If you do not speak, you violate the social order of sequential turn-taking, making it impossible to get down to the business of having a substantive conversation.)

So what is the vision? Much work in this field has been devoted to showing that controlling turns at talk allows people with an interest in preserving their position of power to do so; it is a fundamental technique for the exercise of power. If you cannot even say something, there is not much chance of being able to influence the outcome of an interaction. One can see being shouted over as an interactional equivalent of being censored. So one vision is potentially the reproduction of existing relations of inequality. In its attempt to challenge those, Occupy took seriously the problem of voice by trying to ensure that no one could control turn-taking; the movement instituted alternative rules for taking the floor and getting heard that corresponded better to their idea of horizontal democracy—anyone can take the floor, and word gets passed to all the people present literally by word of mouth. I have to listen to you because I have to repeat what you say. This is one way language is being harnessed to produce a world otherwise.

Another example is the renewed interest in problems of global communication, in the forms of promotion of such concepts as “World Englishes,” in renewed interest in Esperanto and other constructed languages, as well as in computer-facilitated translation and in artificial intelligence. In an echo of the late nineteenth and early twentieth centuries, the expansion of capitalism we have called globalization, and the neoliberal policies states have adopted to facilitate it, has led to anxieties about who will control the global communications required to make that process work. In particular, these anxieties focus on the global spread of English, and on the ability of the so-called “native speaker” (a racialized white from the heart of British and US empire) to define what counts as English and who counts as a legitimate speaker of it—and to control and profit from the global English-language teaching industry.

There have been a number of responses to this issue. One has been to stake a claim for English to what the Corsican sociolinguist Jean-Baptiste Marcellesi called “polynomisme” for Corsican (see Marcellesi et al. 2003), that is, to admit multiple centres of norm-construction (this often goes under the heading of “World Englishes” following terminology introduced by Braj Kachru; see Kachru 1992). Another is to promote and celebrate multilingualism, which occurs in a wide range of ways, ranging from the European Union’s investment in standard language multilingualism to the strengthening of so-called minority language movements, or the revitalization of languages, especially indigenous ones, formerly (and even in some quarters today) thought to be no longer speakable.

Still another has been to return to Esperanto as an International Auxiliary Language, meant to be easy to learn and so more democratically accessible than standard languages associated with specific nation-states. Some adherents orient to Esperanto’s pacifist and anti-nationalist genealogy, connected to its early popularity as a means of countering the political violence connected to the rise of European nationalism in the late nineteenth and early twentieth centuries. Others enter Esperanto through the free software movement, interested in what it can do to facilitate democratic access to and free circulation of information and to the tools to create online discursive spaces out of the reach of corporations; or they enter as self-styled “geeks” interested in the technical capacities of constructed languages and in their ability to facilitate transnational connections (Fians 2021).

Still another response is to ramp up the use of computers to facilitate rapid translation and interpretation, in a tradition unfolding from investments in the Second World War and the Cold War in so-called “machine translation”—here the interest is in ensuring that humans do not mess up in ways that could inadvertently trigger a nuclear war; in facilitating spying by getting around the onerous and expensive process of human language learning; and in avoiding putting too much trust in human multilinguals who can too easily betray masters who do not share their multilingualism (think traduttore traditore). In this view, technology is neutral, although I would say that that conveniently brackets the humans who build and run it. But it helps us understand the general interest in phenomena like artificial intelligence which are forms of imagined ways out of this crisis.

Finally, we need to turn to re-imaginings of this world and imaginings of alternative ones. In recent years we have seen the emergence of post-humanism, an effort to resolve the current crisis by moving past the Anthropocene. In this imagining, sociolinguists and linguistic anthropologists broaden their analyses of interaction to include humans interacting with animals (for example, Cornips 2022), plants (Kohn 2013; Tsing 2021), or machines (for example, Hovens in press), and we open up to understanding non-humans as communicating with each other (see the literature in particular regarding trees, for example, Wohlleben 2015, Simard 2021—not to mention the vast literature on artificial intelligence). Then we have the world of science fiction, a genre which has been quite popular at times of crisis, including this one. Hollywood has produced a number of (utopian and dystopian) science fiction television shows and movies in which fully-formed constructed languages, from Klingon to Dothraki to Na’vi, allow us to enter alternative worlds (see Schreyer 2021). Indeed, there is an entire, mainly online world of language constructing in which participants share tips and sometimes compete to cook up particularly interesting languages, in ways that both overlap with the renewal of interest in Esperanto and which are highly reminiscent of the frenzied production of International Auxiliary Languages between about 1890 and WWI that I mentioned earlier.

By definition, as Fians (2021) points out, this shared activity in the discursive space of the Internet creates community by requiring the negotiation of rules of engagement: what to talk about, how to talk about it, and with whom to talk. However, all these emergent spaces most certainly have their boundaries and their forms of policing, and so, however alternative they may aim to be, often ironically, they reproduce at least some of the hierarchies they are meant to subvert or transform.

In that sense, this example, along with the others I have discussed, highlights the complexities of the struggle for social justice. Movements (and practices) meant to be emancipatory can inadvertently, indeed sometimes perversely, create and recreate differences and inequality at the same time. Language plays an important role, though usually a complex and often ambivalent one. It can be used to foment crisis as often as it is used to construct means of navigating and resolving it. Its polyvalence, multiplicity, and multivocality allow for many, sometimes contradictory things to happen at once, potentially leading to unintended and even perverse consequences. Nonetheless, close attention to its form and practice allows for an understanding of what relations of power are at stake, and what the nature of the complexities and contradictions are.

My central point is that it is in language, broadly understood, that we make and unmake how we understand and organize our world, and so it is a central arena for making both conflict and solidarity when what the world is supposed to do and what it actually delivers get unstuck. What is at stake is nothing less than our understanding of how the world works, or should work, and therefore of who belongs and who does not, of what you have to do in order to belong, and of who—or what—merits life and who—or what—does not.