Social reading in Taiwan

—Zhuangzi 莊子 Tian Dao 13天道第十三, as translated in Geaney (Reference Geaney2018, 8–9).

Taipei has a robust and visible reading culture, with poems displayed in the subway and on buses, novels and self-help books sold at every convenience store, and a massive publishing expo held annually in February. Part of this reading culture is the proliferation of book clubs of varying degrees of formality and institutional affiliation. In January of 2024, the Taipei City Public Library system listed fifty-three reading groups for adults at its various branches, including foreign language reading groups. With a number of bookstores hosting public reading groups of their own and a good handful of independent book clubs that advertise online, there are probably dozens of reading groups to attend on any given night. Add in the numerous nonpublicized reading groups that are simply gatherings of friends and acquaintances, and there are hundreds of groups meeting each month.

Many factors contribute to the popularity of reading groups in Taipei. Among them is the high value placed on reading and letters within an educational culture that draws upon, though is not straightforwardly a continuation of, Chinese-Confucian educational traditions (Smith and Kong Reference Smith and Kong1991; Wu Reference Wu2026). In the greater Sinosphere, textuality has long been understood to be intimately related to both self-cultivation and statecraft, with centuries of elaborate reflexive thought on textual metapragmatics, including the passage from the Zhuangzi quoted above. Though the relationship between contemporary Taiwanese culture and “Chinese civilization” is fraught, Sinophone literary traditions continue to be intellectually, aesthetically, and philosophically influential for many, in part because of their influence on public school curricula (though less so in recent years, see Chou and Wang Reference Chou and Wang2024). At the same time, Taiwanese educational and literary culture participates in global networks of reading practice through increasing integration into systems of international schooling, publication, and media.

Reading groups in Taipei vary along several dimensions, including the genres of text they read, participant demographics, the format and frequency of meetings, how focused the conversation is on the text, and the reading group’s purpose. Some groups self-describe as primarily educational, others mostly social, still others therapeutic or generally in pursuit of self-betterment. The primary criterion uniting these groups and distinguishing them from similar activities, such as classes or lecture series, is that participants are expected to contribute to the discussion. While there may be a formal or informal discussion leader, leaders do not have sole or final purview over the discussion or text interpretation.

For this paper, I draw on ethnographic data primarily (though not exclusively) from meetings of three groups, as well as interviews with participants in those groups. World Books Reading Group is affiliated with an independent bookstore and led by its owner, Hung Fong, who has a literature degree and formerly taught at a community college. The group meets weekly, completing one book over five or six weeks, and focuses on world literature (in contrast with the Chinese literature group held at the same bookstore). They read Marcel Proust’s In Search of Lost Time in its entirety from April 2023 to November 2024, though I only joined the group for volume four. Over the years, the group has read classic and contemporary literature from around the world, including Dostoevsky’s Brothers Karamazov, James Joyce’s Finnegan’s Wake, Jose Saramago’s Blindness, and Nakagami Kenji’s Wings of the Sun. Most participants are women in their thirties to fifties. While many have been attending the group for years, others are more recent additions. About fifteen readers participated in the session that I focus on in this paper. In general, discussion in this group remains focused on the text, and when the topic does shift to other subjects such as current events, it is with a clear and usually explicit relationship to the text under discussion.

The No Rules Book Club is hosted at a library branch and meets weekly on Monday mornings. During the period on which this paper focuses, there were around ten participants each week, including six core members and a rotating cast at the periphery. The group is comprised mostly of retirement-aged people, about half men and half women. The books they read tend to be of a broadly defined self-help genre, though they profess to not have any limitations on the content of the books they choose. Though the books provide a basis for conversation, the primary goal of the group is for members to share (分享 fenxiang) their thoughts and talk about their lives. Members are encouraged to speak their minds, regardless of any relation to the text under discussion.

Cindy’s Salon meets once a month and has been running for over twenty years. There are really two groups: one is based in Taipei and meets at a different coffee shop each month, while the other was started in Beijing and now runs online through WeChat video. Cindy started the Taipei group a few years after she graduated university, as she found working life and later full-time motherhood boring, and she was looking for something to do and a reason to see friends. When she relocated to Beijing for several years for her husband’s business venture, she started the Beijing group. The Taipei group usually meets during weekdays, often for several hours at a time, and the attendees are mostly housewives, like Cindy herself. The Beijing group is a little more varied, with participants calling in from Chengdu, Rome, and Vancouver, in addition to Beijing and Taipei. Cindy recruits new participants to both groups actively through her personal networks.

I participated in all three of these groups regularly over the course of several months as part of my sixteen months of ethnographic fieldwork in Taipei, along with several other groups that met less frequently, some of which also make brief appearances in this paper. In general, I was treated as a regular group member, though I kept my comments to a minimum. For the most part, I kept up with the reading using English translations (and sometimes English originals) for the sake of expediency, but I used the Chinese editions in discussion. In this paper, I do not engage much with the content of the texts under discussion, except insofar as readers find the content to be relevant to their sensory engagement with the books.

Approaches to text in linguistic anthropology

Starting in the mid-1990s in response to the ascendance of the textual metaphor for culture within anthropology, linguistic anthropologists have theorized text as “one, thing-y phase of larger cultural processes of entextualization and co(n)textualization … one kind of metadiscursive interpretation of a phase of discourse, one outcome of a process in which discourse metamorphoses and precipitates as form” (Silverstein and Urban Reference Silverstein and Urban1996, 3). From this perspective, the defining feature of texts is their mobility across time, space, media, and language—texts are a semiotic type, abstracted from different tokens. Indeed, the text is precisely that which has been made detachable from its contextual surround. This theorization has led to productive work formulating a variety of interdiscursive processes to which texts are amenable: de/recontextualization (Bauman and Briggs Reference Bauman and Briggs1990), circulation (Gal Reference Gal2018), citation (Nakassis Reference Nakassis2013a), translation (Gal Reference Gal2015), and transduction (Silverstein Reference Silverstein2003; Keane Reference Keane2013).

Much of this research has focused on unpacking the semiotic labor by which two texts come to be understood as instances of the same thing, tracing out the differences between instances to reveal something about the process of entextualization. The semiotic ideologies and techniques that undergird replicability are culturally and historically specific and are caught up in larger sociohistorical processes of subject formation and institutional power (Inoue Reference Inoue2018). Under the discourse of verbatim as described by Inoue, texts transmitted across distinct sensory channels are understood to be tokens of the same type (“the” text), resulting in a kind of functional immateriality.Footnote ⁴

This immateriality of text is, of course, an ideological achievement. Because text is defined by its decontextualizability, it exists beyond any individual material instantiation. Indeed, in order for text to be mobile, the material form and sensuous experience through which text is encountered must be ideologically bracketed. For a stretch of signs to be a text, it must be metapragmatically interpreted as detachable from the material token through which it is realized.

To demonstrate how texts are taken to be “the same” across different material instantiations, I look to Cindy, who has been participating in and running her own book club for over twenty years now. When I was first invited to participate in Cindy’s Beijing Book Club, the book under discussion in the next meeting was Tiangong Kaiwu (天工開物), known in English as The Exploitation of the Works of Nature. Published in 1637, this hefty encyclopedia of scientific and technical knowledge runs several hundred pages—a somewhat daunting task for a first-time attendee. Cindy suggested that I download the Weixin e-book reader, where I could get the book for free. But don’t worry! She reassured me in writing over LINE messenger. Weixin Reader also lets you listen! I often cook and wash the dishes while listening to a book, it’s very convenient. In this case, not only were all of the members of the reading group expected to have their own copy of the book, whether physical or digital, but the idea of listening to the book, using a different sign-channel all together, was entirely unproblematic.

Cindy, as I learned when I met her in person a couple of weeks later, has developed a routine for preparing for her two reading groups, which she has been running for many years. First, she usually listens to the book on audio, often while doing housework. Then she reads some reviews and summaries of the book online to help gather her thoughts. Each group has a policy of circulating discussion questions in the group chat before the meeting Cindy uses these questions to prepare a printed document with excerpts of the text on which she then writes detailed, handwritten notes to prepare her to participate in discussion. She keeps these notes in big binders she has amassed over the years. At the end of each in-person meeting, she passes along small stationery cards, and everyone writes down their own brief reflection on the book and the discussion. These she preserves in a scrapbook alongside pictures of the group that gathered that day and menus from the coffee shops they meet at. Cindy has scrapbooks for her reading group going back to 1999, serving as a personal record of the group as it has changed over the years.

Cindy’s multistep reading practice shares many of the hallmarks of inscription and entextualization that have been the focus of linguistic anthropological research. Within her practice, inscription enables increasing degrees of durability and manipulability, allowing text—longer portions of text, summaries, and reduced texts—to migrate across contexts in cycles of decontextualization and recontextualization. Within those cycles of entextualization, text undergoes multiple instances of transduction, transformations across semiotic modalities.

The stationery cards, the pictures, and the menus are all entextualizations of the social event of the book club, designed to capture and preserve the setting and people in attendance as much or more than to record what was said on that day. Cindy’s personal notes serve two purposes: first, Cindy reinscribes portions of the book, as well as her own thoughts about them, for easier access in the interactional here-and-now of the reading group. After the discussion, she keeps the notes to preserve the knowledge that she has acquired through the reading and discussion of the book. In both cases, the material operations she performs on the text provide access to its content, and in some cases, its linguistic form, in a different temporality than the original text.

But of course, text-artifacts do have material properties—it requires semiotic and ideological work to background them, and this work is never fully complete. Furthermore, I argue, the material dimensions of each text-artifact, and the reader’s sensory experience of them, are socially consequential. Cindy’s engagement with each iteration of this text—what might be called her reading of it—makes use of the specificities of its material form. The audio format is useful to her because she can listen to it while she does other activities that require her vision and her tactile engagement (cooking, cleaning). As with the digital format, the aesthetic qualities of the delivery are subsumed to the exigencies of daily life. The audiobook itself is rendered through an automated text-to-speech software included in the reading app. Listening to an audiobook is an embodied, sensory engagement, much though the sensory component may be backgrounded in order to enable the ideological sameness of each iteration. To listen to a book is, as Cindy puts it, convenient (方便fangbian)—hardly a sensory quality, but rather a characterization of how the act of reading might fit into the other activities of everyday living. But the dimensions of that sensory engagement are consequential for how the reader encounters the text—precisely because they have no impact on the content conveyed therein.

Material texts

While we can certainly think of the text Tiangong Kaiwu as an entextualization of an entire social-technical world, for the purposes of this analysis, the inscription itself is the primary social object. That is to say, Cindy and her fellow readers are not trying to reconstruct some original interactional world, but to engage with the text as a text. For Cindy, that text itself is, perhaps somewhat ironically, less durable than Cindy’s personal entextualizations, because of the ways that she chooses to read. She encounters Tiangong Kaiwu through a combination of digital-visual and auditory text-artifacts. The audiobook unfolds through sound waves, and the e-book unfolds through light waves. Yet both of these formats have a degree of ephemerality to them. The e-book unfolds through pixels, arranged and rearranged as the reader progresses through the file, and the audiobook unfolds through the digital encoding of an MP3 format (or similar). Each must be reproduced (the audiobook replayed, the e-book’s pixels reconfigured again) in order to be encountered and read again. But for all this ephemerality, these text-artifacts are no less material than the durability of ink on paper.

Materiality appears in classic linguistic anthropology scholarship on entextualization in two ways: through the concept of the text-artifact and through treatments of transduction. Silverstein distinguishes text in the expansive sense, which is “a completely socio-spatiotemporal entity,” from its materialization, which he refers to as the text-artifact (Silverstein Reference Silverstein2005, 18, note 2; see also Silverstein and Urban Reference Silverstein and Urban1996). Though this separation is theoretically justified, in practice, “text” has been the primary object of interest for linguistic anthropologists, while the “text-artifact” has remained largely removed from our analytical gaze. In this original designation, it is not clear whether text-artifact is a type-level designation—such as a book, or the 1995 Penguin Classics softcover edition of Ethan Frome—or a token-level designation, such as my personal copy of Ethan Frome that sits on the bookshelf beside my desk. In this paper, I use the term text to mean a stretch of linguistic signs with the property of decontextualizability. I use text format to refer to kinds of text materializations, such as print book, e-book, and audiobook, as well as text-artifact to refer to a specific materialization: my copy of this edition of Sodome et Gomorrhe.

In this paper, while I address the materiality of texts, I do so in a way that differs from much material text scholarship. Scholarship on the history of material texts generally approaches the relationship between technological developments (such as the printing press), textual forms (such as the book), and modes of engaging with text-objects (such as silent reading). In approaching ethnographic material using tools of semiotic analysis, I have access to a degree of granularity in the everyday use of texts that is not available to the historian.

In Nakassis’s (Reference Nakassis2013b) commentary on Government of Paper (Hull Reference Hull2012), he theorizes the place of materiality in a semiotic account of social life, including the social life enabled by and through graphic artifacts. Nakassis’s treatment of materiality provides an important insight, namely that “materiality is mediated by actual events of encounter and (mis)use and … by how those events are reflexively framed for and by participants” (2013b, 403; emphasis in original). In this paper, I similarly present these “actual events of encounter and (mis)use,” moving beyond affordances to examine the readers’ actual engagements with texts in various material instantiations. This approach provides a perspective that standard histories of material texts cannot. Attending to the material dimensions of text formats and text-artifacts, rather than the content within, provides insight into how readers are supposed to interact with books, i.e. how their materiality shapes the stereotypical or intended interaction between reader and text. However, it is also the case that texts once materialized are available for many kinds of uptake, by virtue of their very materiality. Because my methods are ethnographic, I have access to the real-time practices by which readers engage with text-artifacts. I also analyze their explicit talk about texts, and their talk about texts’ material properties, however those properties may be perceived or reflexively formulated for action.

In dealing with the realm of materiality, this paper also deals with affordances. Any text-artifact materialization has material properties—as such, it has affordances to (specific sets of) humans in relation to certain activities (Keane Reference Keane2018). Humans engage with the material features of an object—in the case of a print book, features such as its dimensions and layout, whether it lays flat or falls closed—through their sensors. The way those material properties shape sense experience can be called an affordance. How those affordances are perceived is mediated by the sense organs themselves, as well as how the semiotic agent might have been socialized to attend to some features and ignore others.

With this in mind, let us return to the cat and the bookmark. The book does not invite the cat to read, because he has never learned (and cannot learn) that patterned arrangements of ink on paper can be linguistic signs. Instead it affords a stable surface to land on and a mobile, colored object (the ribbon bookmark) to bat about. But to the reader, who has been socialized into reading, this physical object invites a different set of sensory engagements. The affordance is not in the object but rather is best understood as a triadic sign relation: the printed words invite reading for those who know how to read (in this case, those who know how to read traditional Chinese characters). The print book has other affordances as well: it has pages that can be turned and it can be written upon. Each of these possible actions is a sensory engagement with the text-artifact. Reading is a visual engagement; listening to an audiobook is an auditory engagement. The touch of the finger over the ink is haptic, the turning of the page kinetic, the writing of marginalia once again kinetic and also visual. The realities of sense-perception are quite a bit more complex than what I describe here, but what is relevant for our purposes is how these different senses are metapragmatically categorized as distinct sign-channels. Because these channels are distinct, a text can be transmitted over multiple channels simultaneously, such as when one person reads a text out loud while others read silently. Readers understand different text formats to engage different senses in distinct ways and act accordingly.

Thus far, I have focused mostly on how individual readers engage with texts in different formats. But, I argue, the material properties and affordances of text-artifacts are particularly important for social reading practices. Phoebe, an intermittent participant in reading groups of various kinds, commented explicitly in our interview on how text format and its material affordances shape the way that readers participate in reading groups. She has a preference for audiobooks, with the exception of self-help books that might have written exercises—these she gets from the library in hard copy and makes copies of the worksheets for her own use. She commented on how different reading media (閱讀方式 yuedu fangshi) enable people to participate in reading groups differently. Physical books, she noted, are easiest to reference back to, because you can flip through to find the thing you were looking at. One can do the same thing with e-books, but it’s not as smooth in her experience. Meanwhile, audiobooks make it basically impossible to recall specific passages. One remembers the general gist of what you’ve read but often can’t point to specifics, and that limits or changes how you can participate in discussion. Phoebe recalled that in one of the groups she went to, there was a specific participant who had the physical copy of the book and was always able to bring up quotes more readily than her peers as a result.

When I met the person Phoebe described, I found the same thing. The book we read was distributed for free in PDF form on the website of an international nongovernmental organization (NGO), the same NGO that hosted the reading group itself. The group leader had also pointed us to a ten-page summary available alongside the book, for those of us who didn’t have time to read the whole thing in detail. When I arrived, I saw that one member had taken it upon herself to print out the entire book. She placed the stack of printer paper, stapled together, on the table in front of her and flipped through it to provide specific quotes to anchor her claims about the book.

In all of these cases, the material dimensions of different text formats and their affordances impact how reading as an activity fits into the reader’s life. They also shape how the reader engages with that text socially, particularly when it comes to questions of citation and durability. The possibilities for how a reader participates in the reading group are shaped by each reader’s sensory engagement with a text, as conditioned by the format of their text-artifact.

Transduction

In the prior section, I focused on the affordances of different text formats in general, both for individual readers and for group reading. Yet, affordances are neither singular nor determinative (Keane Reference Keane2018). The same text format, it turns out, affords several different sensory engagements. This is in part because of the nature of affordances and material objects, and in part because of the textual ideology of reproducibility that undergirds social reading. Each of the examples above makes use of the ability of text to be materialized in multiple formats, in ways that do not just preserve the denotational sense of the linguistic strings therein but also their arrangement, genre, and word choice. The relevant textual ideology holds that the text is the same regardless of the sign-channel through which it is perceived. Inoue (Reference Inoue2018) dubs this ideology verbatim, one in which it is taken for granted that what is spoken can be written down, further dissecting the technological labor required to create inscriptions that are putatively the same as some original verbal production. What is on display in reading groups is the inverse of this same text ideology: what is written down may also be read aloud.

Transduction has taken on two distinct meanings within linguistic anthropology. In the first, the term is used to analyze those components of translation that go beyond the denotational to include the broad set of semiotic phenomena that make text meaningful and effective, including indexical and pragmatic components of discourse (Silverstein Reference Silverstein2003; Harkness Reference Harkness2017). The second sense of transduction indicates how a text may be transformed from one semiotic modality into another (Keane Reference Keane2013). It is this latter sense in which I use the term. In both cases, the relevant metaphor is of a turbine that transmits mechanical energy of one kind (such as the movement of water) into another (the turning of the turbine). Within sound studies, the concept of transduction as the transformation of sound as it changes media has been used in attempts to bridge the gap between the material and the semiotic (Helmreich Reference Helmreich, Novak and Sakakeeny2015).

In his 2013 article, Keane analyzes spirit writing as a process of transduction. Spirit writing encompasses linguistic engagements that draw upon the material properties of writing, such as ingesting the written words of a prayer or creating an amulet from them to harness their spiritual power. Keane writes: “Semiotic transduction aims to tap into the power that can be obtained by the very act of transforming something from one semiotic modality to another” (Reference Keane2013, 3). I take up this notion of transduction to describe those moments in which the materiality of text is made relevant because it is transformed from one sign channel into another.

In a more mundane fashion, transduction features heavily in social reading practices. Reading passages aloud in particular is a common strategy to coordinate readers, to ensure they are literally on the same page. In the opening vignette, a portion of which I reproduce below, group participants coordinate their reading by reading aloud, i.e., transducing a stretch of written text into an auditory form and heard form, as well as by providing directions for interacting with the physical book (page numbers and line numbers).

These readers employ two distinct strategies for locating the correct passage of text. Speaker B provides page numbers and descriptions of where on the page to look for the text (in the middle, eighth line from the bottom). Hung Fong takes a different approach, reading out the first phrase of the paragraph. In this instance, the phrase in a normal day is not being presented for its denotation, but rather as an indexical marker to help the other reader find the passage at hand, reading out a verbatim citation of a visually prominent written string. She is not saying “it’s the paragraph where they talk about what happens in a normal day” but rather the specific linguistic string 在平常的日子裡 is what the reader needs to find. In this sense, it is the transduction of the text across sign channels, not through denotational translation but through simple replication between writing and speech, that enables the coordination.

Though discussion in the World Books Reading Group is focused closely on the primary text, in the No Rules Book Club, it is not uncommon for other texts to be cited in the course of conversation for a variety of reasons. In these cases, reading aloud allows for the real-time unfolding of a new text in the shared space of the reading group.

In the No Rules Book Club, Annie has struggled with the book we’re all supposed to be reading and starts talking instead about some other things she’s read that have felt meaningful to her. She encountered a writing, a poem or perhaps a litany, that spoke to her, and she wanted to share it with the group. She pastes the poem into the group’s LINE chat, which participants use to communicate between meetings, and then also reads it aloud. Other participants haven’t encountered this text before the moment Annie brings it up.

“I sent you[pl] this, by Hellinger, ‘I allow.’ I think it’s very well written. I’ll read it aloud for you all to read, okay? It says, I allow.”

Annie begins reading aloud from her phone. She recites for about a minute, speaking in a loud, clear voice, without any trace of her usual slight stutter. These factors, along with changes in volume, prosody, and the poem’s slightly elevated linguistic register, all mark her speech as a direct transduction of the written poem, even for those of us who are not following along at the link she had pasted into the chat.

Annie also produces several hypercorrections in her pronunciation, both on lexical items that are marked as high register and on some that are very commonly used. Spoken Taiwanese Mandarin has two prominent mergers in comparison to Beijing Mandarin: retroflex consonants zh- [ʈ͡ʂ], ch- [ʈ͡ʂʰ], and sh- [ʂ] merge with the alveolar series z- [t͡s], c- [t͡s^h], s- [s], and syllable-final [eŋ] merges with [en] (Kubler Reference Kubler1985). Though these mergers are widespread both in Taiwan and in many parts of mainland China, they are nonstandard and have some indexical association with lack of education. As such, hypercorrections wherein dentals are realized as retroflex consonants are well-documented as well (Chung Reference Chung2006). While these hypercorrections have largely been discussed for their speaker-focal indexicality, in this case the phatic, i.e. channel-focused indexicality effects are more relevant. As Annie reads this poem aloud, she produces ruci (如此), which can be glossed as thus or thusly, as [ruchi] several different times. Similarly suo (所), which is a nominalizing particle associated with classical Chinese and other formal and especially written registers, is realized as [shuo] multiple times. These hypercorrections further serve to mark this stretch of speech as not speech at all, but in fact a spoken transduction of a written text.

Annie finishes up her turn at talk by saying that she’ll stop reading because it’s too long, suggesting “你們自己看” (You[pl] read yourselves). Notable here is that the verb she uses for read is kan (看), which has the more general meaning of to look or watch. When used as a verb for textual engagement, it entails that the reading be done silently. In contrast, she described her own activity using the verb nian (念), which entails reading aloud.

Annie metapragmatically casts kan as an individual activity, whereas nian makes the text public, collective, and shared. Reading it aloud more effectively brings this stretch of text into the group, more than simply sending it to the group chat where presumably everyone could read it. The auditory experience of the text requires co-presence in the unfolding of the shared space–time of the reading group, in the emergent temporality of the face-to-face encounter. If the text had been distributed to all members ahead of time, with the expectation that it would serve as a shared reckoning point for discussion, this audio-channel reproduction would not have been necessary.

Diverse sensors

Attention to the sensory dimensions of reading acts brings to light the differences between readers as sensors. In Cindy’s account, the possibility for diverse sensory engagements afforded by different text formats expands the ways the text can fit into her life. Letting me know that I could listen to Tiangong Kaiwu on audio was an attempt to make the social reading more accessible for me; the difference between my experience of the text and hers, or between hers and someone who read it in print form, was entirely unproblematic. But this is not always the case.

In one meeting of the No Rules Book Club, Mengjue suggested we talk about what book to read next, since we were coming to the end of our current book, a self-help book called Life Leverage. This group selects their books through an informal democratic process, where anyone can suggest a book to read and then the group chooses among them based on interest and availability from the library. The book we were finishing up was a bit of a flop, to put it mildly. Though most participants found something useful in it, several expressed that it was difficult to read, or boring, or simply irrelevant to their life.

Mengjue’s reflections on the book take a slightly different tack. She shares that she has had a difficult time keeping up with the reading at all, in part because she has a lot going on in her life at the moment, and in part because she doesn’t have a copy of the book of her own, so she couldn’t read it even when she wanted to. This observation in turn spurs a brief conversation about the kind of legwork necessary when selecting a book—talking to the librarians about how many copies we as a group could reserve.

Mengjue then brings up another concern: “And also the print can’t be too small. Because for some of us, reading characters that are too small won’t work well. We really need to resolve this ahead of time, because it’s really important for our mission.”

A full ten seconds of silence elapse, with only the sound of the fans running, as it seems that no one else quite knows what to say. Finally, Professor Chu, the informal leader of the group says, “This is a specialized conversation, let’s not speak of it for now.”

In response to this dismissal, Mengjue changes the topic and proceeds down a different path of inquiry. But at the end of this discussion, Mengjue brings up her concern about the size of the typeface again.

Mengjue frames her material concerns—about the availability of copies for all members, and about the size of print for those with poor eyesight—with regard to the stated goals and principles of the group, which center around the idea of “open.” The ethos the group aspires to is one in which all are welcome and encouraged to discuss whatever they feel, and there are no set rules of engagement. Mengjue sees certain material considerations of the most recent book to be at odds with those core values, because they have resulted in exclusion.

Yet the group leader’s initial response is that her complaint is too specialized—zhuanye (專業), an adjective meaning pertaining to one individual or pertaining to a specialized field of knowledge—and therefore isn’t appropriate for the group to discuss. When Mengjue raises the concern a second time, it is more or less simply ignored, in favor of more familiar discussion about the content of the book.

In this interaction, we see two competing understandings of the relationship between individual acts of reading and social reading, and how the materiality of texts is relevant to that relationship. For Mengjue, attending to material concerns—to the text-artifacts’ availability and to the diverse sensory experiences readers have with those more-or-less identical texts—is crucial to stewarding the group’s collective ethos of open communication and participation. For Professor Chu, the material concerns are a distraction from what is shared—namely the text as a legisign, i.e. text at the level of semiotic type, generalized away from any specific materialization or any individual sensory experience.

Metapragmatic discourse about transduction and reveals similar dynamics with regard to how the material dimensions of text impact its sociality. Although by default any writing can be read aloud, there are complex considerations governing what texts and which contexts are best suited to such transduction. While reading aloud for the purpose of coordination seems largely unproblematic, the impact such a transduction might have on comprehension is open to debate.

As a founding member of the No Rules Book Club, Meilan is in the habit of reading aloud to her husband in the evenings, and she often reads books from the reading group to him, though he does not participate. In our second meeting for a new book, members shared their experience trying to make sense of the book, which most of the group found to be challenging to read and understand. When Meilan mentioned that she had read the book aloud to her husband as usual, Professor Chu protested. “This book isn’t appropriate for reciting (朗誦 langdu). This book is a book that needs to be understood. If you read it aloud, whatever you don’t understand, you’ll understand even less.”

Meilan had a different perspective: “I wanted to read it aloud to my husband so he could understand it, but in the end even he didn’t understand.”

Teacher Wang, who had joined the group a few months prior, suggested yet a third modality which might help with comprehension. When he read it, he said, he focused on the pictures and diagrams to make sense of the difficult text.

In this case, all three readers are dealing with the same text in the same format—they all borrowed their copies from the same library, even. Rather than focusing on the particular affordances of specific formats, then, this interaction brings to the fore the efficacy of that core property of texts—the ability for a text to be the same across media. This same text format provides several different options for how one could engage with it: read/look (看 kan), recite 朗誦 (langdu), read aloud (念 nian), or look at pictures (看圖 kantu). But while the affordances of this text might be material, the material properties of the text are not the only forces shaping how readers might engage with it, and what might count as successful engagement with it. Unlike Cindy, who gladly listened to a fifteenth-century encyclopedia on audiobook, these readers find that the content of the book, which they all agree is challenging to understand, also has an impact on how it ought to be engaged with. But they differ in terms of how they choose to engage with the text, and why.

Each of these sensory engagements also has implications for the sociality of the text, whether the initial reading act is individual or already involving multiple participants. For Professor Chu, recitation is an aesthetic experience, only appropriate for poetry in which, presumably, the sound of the words is relevant to the reading, interpretation or experience of the text. Reading aloud a text like this makes what is already difficult to understand yet more difficult to understand.

Yet, Meilan sees the possibility of a different way to go about making sense of the text—in the past, when she has read aloud to her husband, sometimes he understands more or different things than she does, and so ultimately that helps her to make sense of the text. Or perhaps, as Mengjue suggested, the benefit of reading aloud is that the text unfolds more slowly, and by reading slowly, one eventually can form ideas about denotational content, even if at first it is elusive. Teacher Wang’s comment about the pictures in the text shows that for these readers, part of what makes the text successful is the interplay of two semiotic modes, purely linguistic and linguistic-pictorial, both of which unfold through the visual-channel. The difference is that while the written text can be seamlessly transduced into audio-channel, spoken text, the diagram resists such transduction, or minimally, these readers are not practiced in such a transduction.

What then is transformed when a text is read aloud such that it might affect the ease of comprehension? The transduction of the visual-channel text into the auditory-channel recitation transforms the unfolding of linguistic signs across space into their unfolding through time. This has two consequences. First, multiple people perceive the text unfolding at the same time. Although there are differences between sensors based on their distance from the sound source (see Glazer, same volume), the sonic experience of a spoken text is at least putatively shared by those who are co-present, as when Annie read “I Allow” for the whole group. As such, anyone present can contribute to trying to comprehend or interpret that text. Yet by virtue of that same shared sonic experience, only the reader can adjust the speed at which a spoken text unfolds. The reading cannot be stopped and restarted by anyone other than the one who reads aloud. Strategies for comprehension that readers may have developed for individual, silent reading cannot be applied to the text made sonic and thus made public.