The Madeleine’s Metapragmatics: On Michael Lucey’s “What Proust Heard: Novels and the Ethnography of Talk”

IN 1896, THE FRENCH JOURNALIST Paul Perret claimed that he had distilled Marcel Proust’s aesthetic principles into a phrase: “Desire makes all things bloom, possession withers them all; it’s better to dream one’s life than to live it, although to live it is still to dream.” The line came from Proust’s first book, Les plaisirs et les jours (1896), an odd mélange of stories, essays, and lyrics he’d published in various periodicals, repackaged with floral illustrations by Madeleine Lemaire and musical scores by Reynaldo Hahn. A flop of a debut, the book’s primary consequence was the duel it kicked off between Proust and decadent dandy Jean Lorrain. (Both missed.) But the phrase Perret identified did presciently capture one of the central preoccupations Proust went on to explore in his seven-volume masterpiece À la recherche du temps perdu (1913–’27). Why, that novel asks, do our pursuits entail so much projection and fantasizing? Why, upon obtaining an object, be it erudition or love, can we feel oddly empty? What forms of self-knowledge remain in this illusional world?

For over a century, Proust’s readers have found those questions either life-changing or disingenuous. Perret fell into the second camp. “But no!” he protested of the Plaisirs epigram. “Nothing is less true, less human, less social […] To write is to launch oneself once more into the heart of life; a book is an action, and to keep it wrapped in a dream’s crepuscular color is to remove its power and force.” Yet he was struck by the moments when Proust’s prose diverged from its precepts. Ekphrastic poems and salon satires punctured the collection’s more introspective ruminations, displaying a remarkable eye and ear for contemporary mores. No writer cut off from social reality could have penned such words. These too became central features of À la recherche. Our objects may be illusions, the novel concedes, yet don’t we still pursue them in a certain cultural milieu? What were these milieux during the French Belle Époque, how did people circulate between them, and how did this all change after World War I?

It is this second Proust — the listener more than the dreamer — who interests Michael Lucey. The Proust Lucey evokes in his new book, What Proust Heard: Novels and the Ethnography of Talk, is a certain type of listener, someone who attends not only to words but also to how words function in particular interactions — what Lucey calls “language-in-use.” He’s the Proust who kept manuscript notes listing character argot (Robert de Saint-Loup received cosmique, pratique, and catastrophique), but more importantly the one whose narrator interrupts the story to expound on why those word-person pairings matter.

In those moments, Lucey asserts, the narrator pursues a sort of research agenda, one whose methods resonate not only with turn-of-the-century linguistics but also turn-of-this-century sociology and linguistic anthropology. Instead of asking what a word means (its semantics), these disciplines ask what it does for a certain person, on a certain day, in a certain setting (its pragmatics). All those determinants generate “social indexicality,” the idiosyncratic set of values and meanings we come to attach to all words. When we reflect on why these underlying conditions generate pragmatic outcomes, we’re engaging in metapragmatics.

The word madeleine is the perfect example. Imagine you’re a plus-one at a dinner party, and the host, whom you don’t know well, asks: “Would you like a madeleine?” How do you respond?

A. If you have no knowledge of Proust whatsoever, you will presumably accept or decline based on your dietary and gustatory preferences.

B. If you’ve heard of the author and are reasonably sure your host has too, you may feel inclined to brighten the conversation with a vague reference: “Ah yes, like in Proust — terribly long novel.”

C. If you’ve read a summary or worked your way through one volume, that response might look more like: “Well, I don’t want to spend the rest of this event lost in involuntary memory.”

D. And if you’re a Proust specialist like Lucey, you will likely shudder, recalling the endless dinner parties and receptions where people have made jokes about Proust and madeleines. Sensing another such ambush, you may try to “block” the gambit your host has sprung. Perhaps you will decline the madeleines, explaining that you much prefer dense chocolate desserts, and so you’d best see if there are any brownie bites left. Your host will be doubly distraught, not only because their attempt to engage you in some Proust chit-chat failed but also because you’ve just used the phrase “brownie bites.”

While the nescient (A) and expert (D) attendees end up responding in precisely the same way, their mental and emotional experiences look nothing alike. Different social indices generate different metapragmatic frames, causing “madeleine” to be processed by A as a benign offer and by D as a calculated snare.

À la recherche is full of these ethnographic digressions. Again and again, the narrator recounts overhearing or engaging in conversations whose dynamics invite socio-indexical commentary. Here are a few of my favorite examples that Lucey explores. When Albertine uses the word mousmé (young girl), the narrator signals a generalized horror at the trendy term — a Japanese loanword that entered the French language through Pierre Loti’s 1887 novel Madame Chrysanthème. But he also reflects on why he nonetheless finds mousmé pleasant, even seductive when spoken by his love interest. When the Duc de Guermantes uses the phrase quand on s’appelle (with a name like), by contrast, the narrator registers the locution’s oddity, then brainstorms two “laws of language” that could explain why the lowly expression would grace the lips of such an eminent personage. And when the Duchesse de Guermantes wants to typify a “monster” of a woman, her go-to evidence is the use of words like plumitif (a scribbler or pen-pusher).

What made Proust such a marvelous eavesdropper? One answer would be the paranoia of the closet. What Proust Heard is Lucey’s fifth book, and it marks a departure insofar as “sexuality” does not figure as a key titular term. But here too we find the historically nuanced complement to queer theory that made Lucey’s prior monographs — particularly The Misfit of the Family: Balzac and the Social Forms of Sexuality (2003) and Never Say I: Sexuality and the First Person in Colette, Gide, and Proust (2006) — such field-changers in French studies. “We might speculate,” Lucey suggests at one point,

that some of Proust’s interest in understanding the complex play of indexicality within verbal exchanges that occurs in the space between denotational and interactive texts was stimulated by his experience of the perils of verbal communication about his own sexuality.

To use his stylish example, you need a certain metapragmatic frame to realize that you’re being cruised — or that one character in a Balzac novel is being cruised by another. Queer communities continually invent and repurpose terminology, which helps prevent detection by homophobes but bequeaths a confusing archive to posterity. One of Lucey’s career-long feats has been to reconstruct so much of that context for 19th- and 20th-century French literature. Reading his books shifts your frame, making it much easier to hear the Balzacian cruisers. A word to the wise: if your madeleine-proffering host calls himself “a Vautrin,” something’s up.

In What Proust Heard, however, other historical linkages come to the fore, namely the nascent models of linguistic anthropology that circulated in French intellectual circles. Proust, we learn, was a distant relative of Michel Bréal, whose Essai de sémantique (1897) speculated about diachronic language development, and he was friends with Antoine Bibesco, whose father served as president of the Société de linguistique de Paris. These were the years when Auguste Kerckhoffs transformed military cryptography and Jules Gilliéron published atlases mapping regional language use. Yet after citing a passage from Antoine Meillet’s “Comment les mots changent de sens” (1904), whose resonance with Proust’s novel provides superlative evidence for an influence claim, Lucey demurs: “I feel no need to make that particular argument.” It’s both impossible and unnecessary to know exactly what Proust read; it’s enough to read his novel.

That’s especially true if you buy Lucey’s assertion that Proust’s “tools and concepts” surpassed those of both Belle Époque linguists and today’s literary critics. Though What Proust Heard mostly eschews combative idioms, it does allege that we’ve been poor readers of sociologist Pierre Bourdieu and unreceptive to linguistic anthropologists like Michael Silverstein. Literary studies, Lucey claims, hews to outmoded semiotic and speech act models, which incline us to attribute too much signifying power to words themselves. Semiotics explains why “madeleine” means lemon-flavored cookie. Speech act theory explains why the phrase “I’ll take a madeleine” acts on your host to both extend the cookie platter and judge you an amiable guest. J. L. Austin calls those latter two aspects of the utterance its illocutionary and perlocutionary forces; the first is exerted directly in the act of acceptance, the second indirectly by it.

But linguistic anthropologists want a bit more: they want a nitty-gritty account of why guests A and D maintain such different mental states during identical talk. They point out how common it is for our speech acts to fail and take indexical mismatch as a starting point, perhaps even as the norm. “Indexical force, unlike illocutionary force, would not be a quality one imagines a particular speech act to hold in itself,” Lucey asserts. “It would be a demonstrable effect within an interaction on the emerging sense(s) of coherence within that interaction.” Before “Vautrin” is going to elicit any blushes, its gay index needs to be grasped.

Some readers might protest that speech act theory can accommodate these factors, and Lucey’s focus on its illocutionary over perlocutionary tenets at times risks a straw man. But literary studies can certainly gain by engaging with thinkers such as Silverstein. As Lucey shows in a series of “interludes,” these methods are apposite for many writers besides Proust. Balzac, George Eliot, Virginia Woolf, Nathalie Sarraute, and Rachel Cusk serve as his examples; mine would be Henry James. When Lucey invokes defeasibility — one interlocutor refusing closure in a language exchange — I thought at once of the conclusion to James’s The Wings of the Dove (1902):

“I’ll marry you, mind you, in an hour.”

“As we were?”

“As we were.”

But she turned to the door, and her headshake was now the end. “We shall never be again as we were!”

Unlike Proust, James tends to withhold metapragmatic explanation, requiring his reader to perform talk’s ethnography. Wasn’t the speech act clearly a pressing proposal, we wonder? Why does the “we” have an expiration date for one character but not the other?

The plot of James’s novel provides many reasons. Another explanation is that all language expires, or as Lucey puts it: “Indexical signs are time-sensitive.” The significance of words ineluctably drifts, for both a given person and a given milieu. Your relationship to the word “madeleine” changes once you read Proust, and before 1913, those frames didn’t exist. Likewise, the pandemic seems to have irrevocably altered our relationship to terms like “mask” and “unprecedented.” But those associations will age. If madeleines turn out to be a primary vector of COVID-19, Proust will lose his socio-indexical monopoly.

What’s uncanny about À la recherche is that it theorizes these very processes — the way literary works are themselves language-in-use defined by time-specific frames. The novel’s early chapters, for instance, reflect on the unique impact of childhood reading, when, to compensate for withholding her bedtime kiss, the narrator’s mother opens a present from his grandmother. The package contains four novels by George Sand, and she selects one, François le Champi (1848). While gran-sanctioned on account of its literary merit, the novel features some PG-13 content, so mom is forced to double as a live-action censor. She’s such a talented reader, however, that her tone provides what was “not to be found in the words themselves […] endowing the imperfect and the preterite with all the sweetness to be found in generosity, all the melancholy to be found in love.” Sensuality sneaks back in through her vocal performance, and the impact is unforgettable. When, six volumes and many decades later, the narrator stumbles on Sand’s novel, it elicits an ecstatic digression:

If, even in thought, I pick from the bookshelf François le Champi, immediately there rises within me a child who takes my place, who alone has the right to spell out the title François le Champi, and who reads it as he read it once before, with the same impression of what the weather was like then in the garden, the same dreams that were then shaping themselves in his mind about the different countries and about life, the same anguish about the next day.

For a long time, I’ve thought of these moments as intertextual madeleines — little activators of the narrator’s and the reader’s involuntary memories of prior literary works. Lucey offers a useful revision to that framework, pointing me less toward the semiotics of George Sand and more toward a mother’s sonorous voice.

Other scholars refer to those interpretive shifts using the lexicons of reception, aesthetic judgment, and actor networks. And I would have liked Lucey’s book to dialogue more with thinkers like Rita Felski and Sianne Ngai, who foreground the affective valences of literature as language-in-use. Lucey favors a term from Bourdieu, calling his approach a distinctive “uptake” (reprise) that constructs a context to hear what the novel says. À la recherche, he observes,

[was] initially taken to be primarily a psychological novel, a novel about time and memory, later a novel about art and aesthetic experience, a philosophical novel, a phenomenological one, and perhaps only later still a novel with a deeply anthropological and sociological bent […] Many real-time moments of uptake were required for it to acquire some of these ways of apprehending it.

As usual, the novel is one step ahead, making sure to address its many readers, and its own indexical demise. Having decided to dedicate his life to writing, the narrator observes:

No doubt my books too, like my fleshly being, would in the end one day die. But death is a thing that we must resign ourselves to. We accept the thought that in ten years we ourselves, in a hundred years our books, will have ceased to exist. Eternal duration is promised no more to men’s works than to men.

It’s hard not to read the passage as a retort to Perret’s 1896 review: a blazing assertion that Proust knew “a book is an action” we thrust into the social world. À la recherche’s revolution was to painstakingly depict how one such world was itself constructed by a million desires and dreams. What I’ve always heard from Proust is that every book is wrapped in a “crepuscular color”; today, I might call that the reader’s metapragmatic frame.

¤

Colton Valentine is a PhD candidate in English at Yale.