Katy Highet takes the page 99 test

I wasn’t expecting my p. 99 – right in the middle of my methodology – to be particularly revealing. But as luck (or indeed Ford Madox Ford) would have it, on p. 99 we land midway through a description of my participants that speaks to the crux of my thesis:

Most of the students came from the same sub-caste, one that represents the vast majority of the population of the area around the branch. I have chosen not to disclose the name of the caste, as they are heavily concentrated in one particular area of Delhi – where my branch was located – and revealing this would make the branch potentially identifiable to those familiar with Delhi. I understand that this has scholarly implications, as my writing on this caste cannot be cross-referenced if it is kept anonymous, but this is a sacrifice that I am willing to make in the interest of protecting my participants’ anonymity. With regard to the positioning of the caste, there is great disagreement over which varna (the four main caste subdivisions) they belong to, with some claims that they descend from the Kshatriyas (second ‘highest’ caste) and other arguments that they are from the Vaishyas (the ‘third’). While their ancestral descent may be disputed, in Delhi this caste group have been labelled as OBC (Other Backward Classes) by the government as the caste is deemed socially, economically and educationally ‘backward’. Interestingly, this classification is not ubiquitous across India. In some states, this caste has been allocated ‘Scheduled Caste’ (SC) status by the government, a status reserved for the most oppressed peoples, usually those from the Dalit communities. Parallel to this, in 2008, there was civil unrest in another state of Northern India as this caste group rallied to be re-classified from OBC to SC, which would grant them further quotas for government jobs. While they may not be the most socially oppressed caste, there are widespread negative stereotypes in circulation of them as thuggish, uneducated and uncouth. There was a rumour circulating in the branch while I was there that Dominos no longer delivered pizza to that area as the delivery boys were too afraid. Although a facilitator informed me that this was not the case (Dominos were allegedly refusing to deliver as the road names were not marked clearly and they struggled to find addresses and deliver in the 30-minute timeframe that was promised by the pizza chain), the rumour was a clear indication of the types of assumptions made about this caste group (both by members and non-members). A Google search of the caste brings up links to a question posted in a forum asking why this caste is so feared in Delhi. Most of the responses chastised the caste members for being ‘unintelligent’ and only capable of responding to issues with violence, but one member of this caste took this as an opportunity to attempt to quell such stereotypes…

The aforementioned branch was one of several hundred branches of a non-profit English and employability training programme in North India and was the primary site for my ethnographic exploration of social mobility and English. Specifically, I was interested in the material, discursive and affective consequences of English speakerhood as a deeply embodied phenomenon enmeshed with wider historical processes and structures of inequality. As p. 99 suggests, caste – and its intersection with class, gender, race, colonialism – became the bedrock of my analysis.

Listening to the experiences of the students such as those described on p. 99 unearthed ambivalent feelings about their investments in English. On the one hand, they were deeply committed to pursuing a language that they had come to understand as their key to not only jobs but prestige, pride, (upward) marriage, and modernity (see also Highet 2022); on the other, there were recurrent moments of despair as they wondered whether this language framed as a resource could really live up to its promises; whether their transformation into an English speaker – a figure built upon colonial, class and caste stratification – could really help them counter their marginalisation and stigmatisation.

This is evident in the Domino’s rumour that I chose to illustrate the constant, dynamic presence of caste in these students’ lives, and which shows how the discursive ‘castelessness’, in Deshpande’s (2013) terms, of the middle classes obscures what is in practice a deeply consequential element of social organisation. In later chapters, I return to this to argue that this ideological ‘castelessness’ is a key part of what propels the discourse of social-mobility-through-English, as it reframes social mobility as an individualised pursuit, open to anyone, if only they try hard enough.

The problem, I argue later, is that these remain politically very difficult conversations to have, particularly in the current climate of rising suspicion of anything ostensibly ‘anti-national’; it is partly for this reason that I took the steps mentioned on p. 99 to protect my participants’ identities. This raises difficult questions – questions that I continue to grapple with – about the ethical and political challenges at stake in critical research and in our engagement with projects of social transformation.


Satish Deshpande. 2013. “Caste and Castelessness: Towards a Biography of the ‘General Category.'” Economic and Political Weekly. 48: 32–39.

Katy Highet. 2021. Becoming English speakers: a critical sociolinguistic ethnography of English, inequality and social mobility in Delhi, UCL Institute of Education. https://discovery.ucl.ac.uk/id/eprint/10122328/

—–2022. “She will control my son”: Navigating womanhood, English and social mobility in India, Journal of Sociolinguistics (early view online version available open access)

Michael Lucey on his book, What Proust Heard

Interview by Ilana Gershon


Ilana Gershon: In what sense are Proust and the other authors you discuss – Balzac, Eliot, Woolf — attuned to what linguistic anthropologists know talk can do in the world?

Michael Lucey: I think all those writers had, in their lives, a lot of practical experience of salon culture, or other kinds of social rituals (dinner parties, garden parties, receptions, visits) where talk was very formalized. And I think part of what made them great novelists was their ability to abstract themselves from their circumstances and start to think about what was going on as people talked to each other in these contexts, and in ways that might sometimes seem, to an uninformed observer or even when you are caught up in it, pointless or inane or insignificant. Then they had to think about how to write this down, and how to make it part of a novel. The work of writing these scenes of talk must have been instructive. In their different manners, Balzac, Eliot, Woolf, Proust would find ways to frame such a scene so that they could focus on its internal poetics or so that they could develop a sense of how what was going on in that scene might be linked to preceding or subsequent scenes of talk they would also compose.

Now of course they are composing these scenes of talk, as opposed to transcribing them. Maybe they include bits of language that they’ve overheard and remembered, but I think mostly their compositional effort involve finding ways to convey what they have grasped about the social poetics of talk, or the way a scene of talk can function within a narrative structure. And one thing that narrative structure could be said to be about would be tracking interdiscursivity – how one ritual scene of talking could be linked to another. Another would be the question of the indeterminacy or non-finalization of context: a narrative surrounds talk with context, but also allows the context to shift as the narrative continues. (Think of the important function of rereading in coming to understand a novel. The context has shifted the second time you read a scene, especially if you’ve read the rest of the novel by that point.)

Often people like to say that novels deal with certain cultural topics: adultery, inheritance, ambition, social mobility, some kind of historical crisis or transformation. Bakhtin, on the other hand, wrote once that “the novel can be defined as a diversity of social speech types (sometimes even diversity of languages) and a diversity of individual voices, artistically organized.” It’s really good to keep that angle in mind. Probably novels are multifunctional, so no definition is going to capture all of it, but the novelists I write about seem to be interested in the artistic arrangement of speech types as part of a project of understanding large social phenomena that happen in part through language being used.

But also it’s good to remember that novelists are craftspersons who learn their craft by studying other novelists. And composing scenes of talk can be thought of as a kind of acquired skill. There’s a moment in an interview of Ali Smith by Gillian Beer where Beer asks about a dinner party scene in one of Smith’s novels, There but for the, and Smith recounts how she mapped out the table before writing the scene and adds that she had just read “Penelope Fitzgerald talking about Lawrence.” Fitzgerald says, according to Smith, that “Lawrence is good with more than three people in a room. Not very many writers can do more than three people in conversation.” (But Proust could! And Woolf!) So that set Smith to wondering if she could have a dinner party with nine people and keep up “a rhythm.”  Novelists don’t necessarily need to read Goffman on frames for interaction (although maybe some of them do). They get to that idea by reading other novelists and deciding what works and what doesn’t, or how something can be made to work.

Ilana Gershon: This book is, in a sense, about “how the experience of social reality can be communicated in a literary work such as a novel.”  Could you say a bit about what this approach lets you know about the novel as a genre?

Michael Lucey: There are so many different ways of reading novels. If I want to insist (as I do) that someone like Proust or Balzac is using the novel as a way of conveying a certain experience of social reality, part of my effort has to be pedagogical, showing how to read in order to access this experience that these novels can offer – but that the novels don’t necessarily offer to all readers. I think, for instance, that many people, ones with certain kinds of entitlements, like to move through the world (both the material and the social world) without having to pay too much attention to the social forces they are encountering and interacting with. Some people aren’t allowed not to pay attention, of course. So there aren’t just novels out there in the world, there are also ways of reading novels. We could say that the novel as a genre is a social achievement, and so are ways of reading novels. The experience of social reality can only be communicated by a novel if a way of reading that is attuned to that experience can be brought to bear upon it. And of course different novels understand social reality differently. For instance, the naturalist novel (say Zola or Dreiser or Wright) is often characterized as intently focused on the inevitability of certain outcomes due to implacable social structures and to social forces that cannot be resisted. This can make them frustrating to read for some. The novelists I’m writing about seem more aligned with Bourdieu’s observation that the social world is “a space of immanent tendencies” such that “everything is not equally possible or impossible for everyone at any given moment.” They are interested in how the virtual topography of that social space might be brought home to a reader. Representing scenes of talk interdiscursively linked across time and involving a number of principal speakers also moving through time and social space is one strategy for revealing that topography.  These novels suggest that there are instruments that can give us access to an otherwise virtual social topography, and a novel can be an instrument like that.

Ilana Gershon: In attending to novels with the tools linguistic anthropologists have honed, what kind of translation work did you have to do?  How does having an author in the mix change the analysis?

Michael Lucey: Literary criticism has notions like text vs. subtext (“what is the subtext here?”) or denotation vs. connotation, or explicit vs. implicit forms of meaning, or style, or tone, or even intertextuality, that open onto linguistic anthropological territory, but not usually in a systematic way and not always in a way that allows you to think not only about the internal workings of a text but also about its own ongoing entextualization(s). There is also sometimes a tendency to treat literary texts as sacred aesthetic artifacts that house various mysteries; there can also be a tendency to view context as a fixed explanatory frame linked to a text’s original circumstances. So one challenge is to make sure that contextualization, entextualization, indexical presupposition and indexical entailment, speech genre and ritual as they are understood by linguistic anthropologists (along with concepts such as language ideology or metapragmatic function or indexical order) don’t get assimilated to already existing understandings in the field of literary criticism. This involves subtle shifts in how you think about the literary object of analysis and those shifts can be hard to hold onto.

Authors and narrators are a fun part of the challenge of translating linguistic anthropological concepts and methods into the study of the novel. Of course we know, in a basic sociology of literature kind of way that authors have their own situation from which they “speak.” We can also think of novelists as “in conversation” with other novelists or with writers of other kinds. As Bakhtin said, “the novel as a whole is an utterance just as rejoinders in everyday dialogue or private letters are…” But – and this is the important aspect for What Proust Heard — it is an utterance that includes other imagined utterances. Then what the concept of “narrator” adds is the idea that the entirety of a novelistic utterance can be taken to be constructed by the author as someone else’s—an imaginary someone else usually.

If you spend time with Proust’s correspondence or with biographies of him, you learn not just that he was an attentive listener, but that he was an amazing talker, and that sometimes he would find himself engaged in outlandish and excessive social interactions that included frenzied verbal exchanges that rival anything to be found in his novel. Indeed, it seems clear that he transposes some of his own social and verbal excesses into his novel – sometimes assigning them to the narrator, but sometimes also to other characters. It is as if he was experimenting with hearing himself, with trying to understand what he might have sounded like to differently attuned ears, or what something he said might sound like if said by somebody not quite like him. So his novel becomes, in part, a representation of multi-perspectival listening. If linguistic anthropology is attentive to the differences that can be discovered in different ways of saying the “same” thing, Proust’s novel is interested in something related: different ways of hearing the “same” sounds. Sometimes those sounds are linguistic ones, but they might also be musical ones, or other kinds of sound. Proust’s novel is not just an investigation of how to hear other people’s language; it also investigates why certain people hear what they do and why other people hear something else.

Ilana Gershon:  You analyze narrators who track talk in ways similar to linguistic anthropologists, even exercising a talent for verbatim re-contextualizations that is a professional necessity for linguistic anthropologists.  What does Proust suggest about what the consequences might be for living life with the attentiveness of a Bourdieu-loving linguistic anthropologist?

Michael Lucey: Once when I was given a talk based on some of the material that went into What Proust Heard, one of my friends in the audience said afterwards something to the effect of, “it’s enough to make you scared to ever speak again.” And it’s true that you might not exactly want to have the experience of “someone like you” listened to by someone like Proust and then finding yourself reading a novelistic transposition of “what you sound like.” In What Proust Heard in the last interlude where I’m looking at novelists other than Proust, I take up Rachel Cusk’s Outline trilogy. That trilogy an example – quite different from Proust, but in a revelatory way – of a novelist who represents people talking in a way that is a little scary because of what she is capable of hearing in the speech of others. Definitely don’t strike up a conversation with her on a train or a plane or at a party – unless the experience of hearing yourself heard that might result from it is one you could depersonalize in a salutary kind of way. Could you manage to find it interesting to learn what people like you sound like to someone like Cusk’s narrator?

But I think on a more general level this is about learning to respect the complexity of the phenomenon of language-in-use. It might seem like novels, which take complex verbal exchanges and render them as written representations of imaginary instances of direct or indirect discourse (accompanied to varying degrees by novelistically-styled pragmatic and metapragmatic commentary) would have a tendency to over-emphasize the view that language mainly serves to communicate thoughts and feelings, or perhaps that it allows people to deceive other people in dastardly ways, or to enable the pursuit of personal or group ambition. But then you encounter a novelist like Proust, or a lesser-know novelist like Robert Pinget, whom I wrote about in my earlier book, Someone, and you see that these novelists didn’t just listen to individual speakers. They listened to worlds of talk in and through which individual speakers moved, and they found ways of writing novels to convey what that way of listening might mean. And so they can really inspire you to think: wow, I wonder if I could listen like that, and if I did, what I would hear.

Ilana Gershon: How do you engage with the psychological in What Proust Heard?

Michael Lucey: I guess I might say that novelists like Proust or Pinget, or even Cusk or Woolf, encourage an attention to the sociological side of the psychological. Concept of “psychologies” are, after all, social facts, cultural concepts that organize our listening and our processing of what we hear. So, of course, we hear “the psychological” when we listen to other people. But if we notice that other people hear the same sounds differently, then we might start to think that what we hear as psychological is, to some extent, indexical of something about us and our own positioning in culture. Proust’s novel is so often read as a deep psychological exploration of the narrator’s experience of the sensory world, the social world, and of time. And of course it is that. But it also turns out to be about the linguistic registers that make the conveyance of that experience possible and about how they interact with other registers–both proximate and distant; it is about how you acquire a register in which to register your version of psychological experience, how you came to that register, how you master it, how you address people from inside it, how you recruit people to it, and so on.  There’s a way in which Proust’s novel becomes a rich occasion for thinking about the interactivity of the sociological and the psychological – an interactivity that can be investigated through a linguistic anthropological analysis. It’s perhaps the same kind of interactivity that exists between the concepts of habitus and field for Bourdieu.

Ilana Gershon:  Are some literary texts more amenable to this analytical toolkit than others?

Michael Lucey: Well, I think you can always, as a reader, be listening to what a given novelist sounds like, how their novels sound, how their characters sound. Flaubert, for instance, wrote in his letters about how much he hated dialogue in novels, and critics have calculated that whereas a Balzac novel might be comprised of around 50% dialogue, Madame Bovary is only somewhere around 20%, perhaps because Flaubert found there to be something aesthetically debased about spending too much time writing dialogue. You can’t turn to Flaubert’s novel for the kind of study of language-in-use that you can find in Balzac or Proust. But Flaubert is also famous for his obsession with how his sentences sounded; he had a practice of declaiming them persistently to make sure they sounded appropriate to him, to make sure they conveyed what he wanted them to convey on some level other than the denotational one. He talked about his work in his gueuloir, gueule being a term for mouth in a very informal register, as if he had to get a mouth-feel and/or an ear-feel for his sentences before he could be satisfied with them. We might think of this practice as a one of linguistic improvisation, where the resources of a linguistic habitus are mobilized in the service of what were for Flaubert new kinds of indexical effects, as if he were aiming for utterances that could index an interplay of conflicting registers held tensely together by phonic and rhythmic effects. This is partly why Bourdieu finds Flaubert to be such a richly sociological writer, even though this might not have been Flaubert’s explicit aim. Bourdieu gives a marvelous description of what it might mean to listen closely to what Flaubert has done by “mak[ing] of writing an indissolubly formal and material search, trying to use the words which best evoke, by their very form, the intensified experience of the real that they have helped to produce in the very mind of the writer . . . oblig[ing] the reader to linger over the perceptible form of the text, with its visible and sonorous material, full of correspondences with a real that is situated simultaneously in the order of meaning and in the order of the perceptible.”

So I think you will have to choose the right tools from the toolkit, and refine them appropriately, to deal with each novel. You won’t be able to deal with Flaubert or Proust or James or Woolf or Baldwin or Morrison or Ali Smith or Marie NDiaye in the same way. But some version of the toolkit will probably always be illuminating.