Vous êtes sur la nouvelle plateforme d’Érudit. Bonne visite! Retour à l’ancien site


#1b1t: Investigating Reading Practices at the Turn of the Twenty-first Century

  • Anatoliy Gruzd et
  • DeNel Rehberg Sedo

Corps de l’article

Reading is a fundamental modern activity, influenced but not always confined by structures of geographic space, time, age, culture, or economic status. And although reading is a private activity, many enjoy sharing their reading experiences. The sharing of reading experiences traditionally happens face to face (f2f) in book clubs, with people from similar geographic areas. However, Twitter and other social media and networking sites have the potential to make it easier for people to meet virtually to share their reading experiences. Additionally, technologies to capture and analyze these experiences are able to provide reading researchers and historians with novel investigative tools.

This article provides an historical account of the origins of Twitter’s first book club and thus maps a unique reading community. We illustrate through discourse and social network analysis that Twitter allows an author unique opportunities to participate in reading communities. The platform also gives readers the space to interact with the author and with each other in ways not previously possible. While sometimes difficult to observe and analyze, discourse rules and online performance is part of community maintenance. Sampling readers’ conversations and their responses to the book with new observational and evaluative technologies, we begin to see that the asymmetric structure of Twitter and its potential reach to millions of its users offer readers opportunities for engagement that are not available in f2f groups. However, we also see that there are cultural literacies and skills that readers need to participate fully in the reading group on Twitter.

This paper is a timely attempt to better understand what Bob Stein, the founder of the Institute for the Future of the Book, has called the “context [of the] reading experience”. [1] We ask: What evidence of reading experiences can be found in the unique online environment of Twitter? To begin to answer this question, we investigated the One Book, One Twitter book club (#1b1t). [2] Jeff Howe (who coined the term “crowdsourcing”) believes that the activity of shared reading is quickly changing with the proliferation of social media. Inspired by the popular and successful One Book, One Chicago community reading model, he started #1b1t, effectively moving the usually face-to-face activity into the virtual world.

Whereas investigating historical reading groups, practices, and responses can require the study of readers’ letters, diaries, biographies, club minutes, and marginalia markings, the examination of contemporary collective reading usually utilises methodological tools such as ethnographic studies, surveys, and discourse and textual analysis. [3] These methods are altered, and perhaps even enhanced, when the researcher moves into the virtual world of shared reading. Using Netlytic, [4] a web-based system for automated text analysis and the discovery of social networks, our study pinpoints a particular time in history that opens new conclusions about the spread of knowledge, education, culture, and ideology. An analysis of more than 14,000 “tweets” about American Gods by Neil Gaiman, [5] which was the first selection of #1b1t, provides insight into this world-wide reading group. This paper is one interpretation of shared reading practices at a unique moment during the twenty-first century, in which the future of reading, publishing, and writing seems uncertain.

Build It, and They Will Come: Creating a Reading Community on Twitter

Since his initial call to readers, Howe’s book club has morphed from a one-off summer reading program to an ongoing, more “traditional” book club that sees readers coming together to discuss one book every month. Now titled “1Book140,” the online book discussion group is very different from a group of readers gathering together in one member’s living room or in a local library, which is often the case for Western f2f groups. Readers who participate in #1b1t hail from disparate parts of the globe and really never meet in one space at the same time. Instead, they choose the book they would like to read, share their interpretations of that book, and “chat” with one another in the virtual space that is known as the “Twitterverse” using not more than 140-characters for each post.

In promoting his idea for the book club, which is now sponsored by The Atlantic, Howe wrote, “I have a dream. An idea. A maybe great notion… What if everyone on Twitter read the same book at the same time and we formed one massive, international book club?” [6] Working with the idea that One Book, One Community (OBOC) models of reading (which see one community, region, or nation reading one selected book) create what Robert Putnam, Lewis M. Feldstein, and Don Cohen [7] have called “bridging capital”, the “early adopter,” [8] author, and businessman Howe projected that through Twitter, readers could create “communities of people who have little else in common,” and that the creation of such communities could be accomplished through book discussions.

One has to give Howe credit: his intentions are laudable and they actually parallel the motivations of some—but not all—OBOC producers and funders. [9] While some producers want to promote the activities of their public library, and others want to create relationships across literacy, arts and cultural agencies, all programs work to create some sort of reader engagement. As Danielle Fuller, Amy Thurlow, and DeNel Rehberg Sedo have written, cultural policy in Canada, the United States, and the United Kingdom is currently driven by the notion that expanding participation in programs like OBOC increases the cultural and social capital of citizens, thereby producing socially responsible and politically engaged individuals.

However, Howe’s intentions do not seem as ideologically neoliberal, even if they are idealistic. Consider, in addition, that during the voting phase to choose the first #1b1t book, he wrote:

I wouldn’t dream of second-guessing anyone with her own action figure, but I think the “One Book, One City” programs make a very industrial age assumption: Namely, that most of our relationships are determined by geography. On the Internet—where affinity is more powerful than geography—that’s just not the case anymore. And so I’d like to ask a slightly more ambitious question: What if everyone in the world read the same book? We have the ideal technology at our fingertips—Twitter. All we need now is the book. [10]

To assume that any formation of a reading group or book club can create “community across geographical, cultural, ethnic, economic, and social boundaries” is a large task. [11] While book clubs vary in size and format, [12] infrequently do they mimic the communities that are formed as a result of OBOC programming. The types of reading communities that form around OBOC events tend to be ephemeral, imagined, and geographically distributed. [13] The public or semi-public space in which much OBOC book talk takes place—such as library breakout rooms, theatres, or via radio or television broadcast—is different from that private space in which f2f book club members often meet. Book clubs that meet f2f and those online book clubs that have a core membership who meet for many years have created an intimacy and a dynamic history that is difficult to duplicate in a city-, region-, or nation-wide reading program. The space that is Twitter, on the other hand, might provide the unique environment that can facilitate community creation through shared reading.

Tweet Analysis

Twitter, founded in 2006, is a social networking and microblogging service for exchanging short messages (140 characters long) that are known as “tweets”. [14] According to a recent report, there are currently about 200 million users on Twitter. [15] The website is especially popular in countries such as the United States, Brazil, the United Kingdom, Japan, and Canada. [16] Unlike other social networking sites, such as Facebook, Twitter promotes asymmetric-type relationships in which connections between Twitter users usually are not reciprocal. In other words, if one user follows another user on Twitter, the second user does not have to follow the first user back. This often leads to the creation of power users who have many more followers than the number of people they follow back; power is thus put in the hands of a few influential users. However, at the same time, the asymmetric nature of Twitter also helps to interlink different online communities and facilitate the flow of information and support between these communities that otherwise would be disconnected. [17]

The data for this study consists of 14,427 tweets posted by members of the #1b1t community that were collected between March 24, 2010, and April 30, 2011. For the purpose of this study, only messages with the “#1b1t” hashtag were collected. The “#1b1t” hashtag is a prior agreed upon keyword used by the members of the book club to identify messages related to the #1b1t discussion. The messages were collected using a website called TweetBackup. [18] For the analysis of the messages, Netlytic, a web-based system for automated discovery, analysis, and visualization of information about online communities, was used. [19] Netlytic has previously been used to analyze various online communities including learning communities, [20] communities of bloggers and blog readers, [21] communities emerging on the i-Neighbors website, [22] a scholarly community on Twitter, [23] and most recently a study of an online community around a popular fan website, TheOneRing.net, dedicated to discussions of Tolkien’s (text) and Jackson’s (film) versions of The Lord of the Rings[24]

Choosing the Book

As in many f2f book clubs and OBOC programs, choosing the book to read is top of mind for the members/readers and it is often a complex process. In f2f and “traditional” online book clubs, the criteria are decided upon by the various members and reflect the implicit or formal norms of the group; some vote on member-generated options, some leave it to one person—the host, for example—and some follow more prescribed lists, such as those provided by Great Books™. [25] OBOC programmers have a larger task in that they often have to consider hundreds, if not thousands, of readers when selecting the book. They must consider the political implications of the book, and in addition to accessibility issues, they think about practical factors such as the number of copies available for libraries and purchase, and the programming opportunities around the book. [26]

In the new environment of Twitter, the book selection process had to be negotiated across time zones and genre preferences. While the main discussions about which book to read took place on Twitter, other social networking platforms were implemented in order to facilitate the process. Reddit [27] was used for voting on the book and on the “official #1b1t crest,” while the blogs— Wired.com and Crowdsourcing.com—helped Howe and his colleagues to explain the book selection process in more detail.

Figure 1

The crest of #1b1t

The crest of #1b1t

-> Voir la liste des figures

A Facebook page repeated some of the Twitter conversation, and also acted as another channel through which to direct readers to the Twitter platform. [28] The various media needed to choose one book to read together illustrates that the communication needs of a regionally disparate group of readers are not easily accommodated on one platform and can be facilitated using media that mimic the synchronicity of f2f communication.

Howe wanted the chosen book to be of “general interest,” available in “many, many languages,” and “freely available.” [29] By “freely available,” we assume that Howe meant he hoped the book would be digitally accessible to readers free of charge. That did not happen. Instead, Tweeter-readers chose Gaiman’s novel over Ray Bradbury’s Fahrenheit 451 and 1984 by George Orwell, even though a commenter on Howe’s blog responded to his initial call with support: “I like the idea. In the spirit of Crowdsourcing, selecting a book that is in the public domain and available through Librivox.org [30] and similar sites would eliminate many financial and geographic barriers.” Interestingly, the responses Howe received after his initial blog post announcing his idea were not all positive. [31] Several comments illustrated critiques that some readers have of televised book clubs, while others expressed cynicism and mistrust of publishers appropriating the discussion for financial gain. In the end, the Tweeter-readers chose a book that is very much still on the market. Published by Headline, William Morrow, Harper and Harper Perennial, American Gods is in its tenth edition. [32]

Classified as science fiction or fantasy, [33]American Gods was one book on a list of ten selections. Six titles were nominated by Tweeter-readers: American Gods, Fahrenheit 451, 1984, Brave New World, Slaughterhouse Five, and Catch 22. The “1b1t advisory board” selected four others—100 Years of Solitude by Gabriel Garcia Marquez, Catcher in the Rye by J.D. Salinger, God of Small Things by Arundhati Roy, and Toni Morrison’s Song of Solomon. Howe justified adding more titles in order to broaden the gender representation of the nominated authors:

Here’s our reasoning—the crowd picked six wonderful books, but they’re all written by white men with, well, a healthy disregard for reality. That’s not a dis—hell, Catch-22 and Slaughterhouse Five are two of my all-time favorite novels. But it’s bound to appeal to a limited constituency. And besides, each one of the additional four is a masterpiece in its own right, so who can complain? [34]

Howe and his group may have been trying to appeal to a wider audience, but as participant observers, we noticed early on in the nomination process that there was a skew towards the American classics, if not the American canon. While attempting to take the book club “global,” the chosen book was firmly rooted in American literary culture. That is not to say, however, that the members of the club were all American. Indeed, of the top ten discussants, at least four were not from the United States. One was from India, one from Germany, another from Canada, and another was from either Portugal or Brazil. Twitter profiles do not always provide the location of the user, nor do we always know the gender of the reader. However, users often link their profiles to their blogs, which sometimes provides illuminating information for the researcher.

As figure 2 indicates, the most active Tweeter was Crowdsourcing. This is the handle that Jeff Howe uses, so this finding is not surprising. What is surprising is the number of men involved in the book club. While 4,815 independent Tweeters chimed into the discussion, of the top ten posters, six were most likely men. This split between men and women in the book club is notable: f2f book clubs are usually composed mainly of female readers. [35] Of the top one hundred posters for this group (people who posted nineteen or more tweets in the data set), thirty-seven were male, fifty-three were female, seven were of an unknown gender, and three were organization handles. [36] In the next section, we discuss what the male and female readers talk about in this book club.

Figure 2

Top 10 most active posters on #1b1t—twitterers and their corresponding number of tweets which they posted to the group

Top 10 most active posters on #1b1t—twitterers and their corresponding number of tweets which they posted to the group

-> Voir la liste des figures

Book Talk

In traditional f2f book clubs, readers often come together physically once a month to discuss a book. Someone who has not read the book but attends the book club discussion understands that he or she may hear “spoilers,” that is, details that give away the ending. In online book clubs, the readers often provide qualifiers to the other group members if they discuss the book before everyone else has finished reading it. For example, someone might write, “spoiler alert,” to contemplate a plot twist. Howe and his group decided to outline a schedule for book talk based on chapters, and in writing about it on the Crowdsourcing blog, Marcela Valdez provides a glimpse into how this group thinks about reading practices. She writes:

So now that we know what we're reading, we still have to answer the question: how do you get thousands of people to read one book together without ruining the suspense and twists for anyone? After all, some people like to read in delicate sips of ten or twenty pages, while others—like me—prefer greedy three or four hour binges. And many of you have already devoured the entire book, or are reading for the umpteenth time! [37]

Most of the readers who participated in the conversation appear to have read the book before, or if not before, then very rapidly. According to our analysis, the majority of the discussion took place during the first two weeks (between May 2 and 14). During this period, chapters 1 through 6 (of 20) were on the schedule, [38] but a close reading of the general discussion indicates that the readers were finishing the book as early as the first week. By week eight (June 23–30, 2010), during which the final chapter and the postscript were to be discussed, participation in the discussion dropped off significantly.

Figure 3

The number of tweets posted by the #1b1t club members over time

The number of tweets posted by the #1b1t club members over time

-> Voir la liste des figures

Through the textual analysis function of Netlytic, we can generate a tag cloud visualizing the English words mentioned most during the discussion (see figure 4). The size of each word in the cloud represents how many times it was used in the data set relative to other words.

Figure 4

Tag cloud of the most frequently used words in the book club discussion (discovered by Netlytic, visualized by IBM’s Many Eyes)

Tag cloud of the most frequently used words in the book club discussion (discovered by Netlytic, visualized by IBM’s Many Eyes)

-> Voir la liste des figures

After eliminating pronouns, contractions, and common words like “and” and “are,” we begin to see the focus of the discussions. Not surprisingly, “@crowdsourcing” (mentioned 2,106 different times), “read” (2,252), and “reading” (1,739) were a large part of the book talk. The author, “@neilhimself,” was mentioned 2,234 times, indicating that the Tweeter-readers often included the author in their discussions. In some instances, such as the one below, Tweeters illustrate that authors remain celebrities in contemporary print culture:

2010-05-18 11:45:14
@neilhimself omg. its you! love your work. #1b1t / @fatimoir

In the Twitterverse, the readers were anxious to include an author who was familiar and active in the environment, [39] and Gaiman himself “tweeted” support for the book when nominations were made:

2010-04-16 16:10:51
@Crowdsourcing I’m now in this weird position where I’d like to plug #1b1t more, but won’t because I get accused of "flooding" the vote / @neilhimself

When we analyzed the conversations that included “@neilhimself”, readers usually mentioned the book and included the author as a way of compliment:

2011-01-09 02:34:17
I thoroughly enjoyed reading American Gods by Neil Gaiman along w/the rest of the world last year. @1B1T2010 #1b1t @SBookstaver @neilhimself / @twindy5

While choosing a book because of its author is not uncommon—and is one of the top ways of selecting a book [40]—this kind of para-intimate interaction with an author is something that the online environment enables more so than the physical f2f environment. [41]

Netlytic indicates that the word mentioned most often was “gods” (4,374 times). Those who were not participant observers in #1b1t might assume that the discussion was about the main characters in the book, which as the title suggests is about gods. However, analysis shows that the talk is about the title instead of the main characters in the text. We discovered this finding because Netlytic allows for contextual expansion through a click on the word in the word cloud, which provides more nuanced and richer data. Consider, for example, the following Tweet:

2010-04-30 17:49:45
@1B1T2010 I’m doing #1B1T to promote literacy and score cultural cache. I’ve already read Ame. Gods and hated it, but maybe in a group...? / @dsbigham

First, we see that this reader thinks that engaging in the book club will provide him/her with cultural capital and that reading the book again with the help of other readers might change her/his initial response to the book. Then, by searching our data set for another post by @dsbigham, a retweet, or a response, we can evaluate whether or not the goals he/she set out were accomplished. @dsbigham did not post again so we might assume that the book club did not influence her/his original opinions about the book.

The social norms of book clubs often mandate certain discourse and performance expectations that the group itself creates and maintains. [42] This seems to be consistent across book club spaces and platforms. In this book club, a reader has to be able to express an idea in fewer than 140 characters, and he or she has to do so (most effectively so as to be understood by the majority of other users) in English. Readers also illustrate that there is a discussion threshold that some cannot cross. Consider, for example, that until May 7, 2010, “gods” was mainly present in tweets attached to mentions of the selected book’s title. After May 7, the contextual analysis shows that readers began questioning and interrogating the gods that make up the characters of the book. At one point, readers contemplate the platform in which the discussion is taking place. A reader with the handle of “wiivie” indicates that he or she is discussing a book about gods in a communication platform that is inherently narcissistic for many people. The tweet is retweeted and philosophically commented upon:

2010-05-09 12:36:19
RT @wiivie: #1b1t interesting discussing a book about gods in a medium based on attracting followers or following others. / @scarcraig

To which another user responded:

2010-05-09 08:26:59
@wiivie brilliant point! "The meaning of life is to write your name on a wall and hope when the wall is gone, your name remains"? #1b1t / @magnushimself

This particular thread is interesting in that Gaiman’s gods are “old” gods who were carried from the old country to the new in conflict with “new” gods represented by money, the media, and new technology.

As we indicated above, by the end of the allotted duration of the book club, discussion had ebbed significantly. However, several active readers adhered to the schedule laid out by Howe, and provided interesting interpretations of the book’s finale. We include a snippet of the conversation here to illustrate how several readers work out the final plot with one another:

2010-07-03 14:36:09
The "great reveal" was an interesting twist. Is there a deeper metaphor here or is this just plain fantasy? #1b1t_18c #1b1t / @grazen

2010-07-03 14:37:57
The old gods and the tech gods decided to get along because their battle was meaningless? #1b1t_18c #1b1t / @grazen

2010-07-04 16:16:54
@timemastertim He was without guilt, a man who’s father was a god, he sacrificed himself and was resurrected to save the world. #1b1t / @grazen

2010-07-04 18:21:02
#1b1t @grazen Using pagan beliefs was a way to incorporate the heathens to Christianity. Local gods->saints, local rites->christian rites / @FSkornia

2010-07-04 18:21:47
#1b1t Essentially, the Christian Church subjugated the local customs and said, hey look, you’re all christians now, pay us money / @FSkornia

2010-07-04 18:25:28
@FSkornia There is a Christ like quality to Shadow though. Christ was a friend to thieves & promised one a place in Heaven. #1b1t / @grazen

2010-07-04 18:29:08
@grazen There maybe similarities, but "son of a god" idea is not unique to Christianity, could show signs of other cultural influences #1b1t / @FSkornia

By analysing the discussion as it happened, which is possible through direct observation or by backing up the tweets, we can observe how these two readers (@grazen and @FSkornia) interpret not only the main character’s symbolic nature, but also critique contemporary institutionalized power inequities. Their conversation indicates a comfort level with one another and with their own abilities to interpret the deeper meaning in the book’s message. In fact, this is precisely how new online connections tend to form on Twitter and existing ones are re-enforced. To find out how common such conversations were among members of this group, we analyzed the communication network among members of this online community using social network analysis.

Social Network Analysis

One way to gauge the level of interactivity in this group is simply to count how many messages were conversational in nature. On Twitter, it usually means that a message would contain the @ sign followed by a Twitter handle. In Twitter slang, this means that the message refers to someone on Twitter, whose handle is mentioned in the text either directly or (by reposting that person’s original tweet) indirectly. In our data set, 9,750 (68%) out of 14,427 tweets contained the @ sign followed by a Twitter handle. This suggests a very high level of conversational-type messages posted by the book club members, which is about three to five times higher than the Twitter average (25.4% [43] or 12.5% [44]). However, the question remains whether these conversational-type messages address members of this group, or whether people are simply referring to others outside of the #1b1t community.

To address this question, Netlytic was used to discover who mentions whom in the messages, in order to uncover who interacts with whom within this reading club. The result of this procedure was an interaction network consisting of 3,361 club members and 6,314 connections among them (see figure 5 below). Since this network represents who talks to whom in the book club, only people who mentioned (or were mentioned by) at least one other club member are included in the network. The resulting network was then visualized and analyzed using the social network analysis software called ORA. [45] The position of each node in this network visualization is based on a popular “force-directed” algorithm that attempts to display the network with the least number of crossing connections. [46]

Figure 5

Interaction network of the #1b1t community

The size of the nodes represents the number of neighbouring nodes with direct links.

Interaction network of the #1b1t communityThe size of the nodes represents the number of neighbouring nodes with direct links.

-> Voir la liste des figures

One interesting observation is that there are two distinct clusters of users in this network—one that is centered around Jeff Howe (@crowdsourcing) and another around Neil Gaiman (@neilhimself). This demonstrates that both Jeff Howe and Neil Gaiman, who are influential Twitter users, [47] bring their own group of followers into the conversation of this club. This also reconfirms the positive impact on the book club of having an author with a strong Twitter following who is also joining in the conversation himself. An active author is something that other book clubs run through an online social networking site might consider during the book selection process. Selecting a book and a discussion platform in which the author of the selected piece has an active presence may be a good starting point for creating active discussions and building up an online community. But it is not only an author’s influence that can help to build an online community. The promotion of the club in popular online media may also help to bring more readers to the club; such appears to be the case with the online magazine, Wired, whose Twitter account has a substantial group of nodes attached to it (see figure 5).

The most interesting group of users is actually not those who are only connected to Howe’s or Gaiman’s account, but those who are connected to both. There is a group of nodes in between the two mentioned clusters (circled in figure 5 and labelled as the “community core”). This group of people is especially important for the club: it forms the core of this community. The core consists of members who are most actively involved in the discussion. They post frequently, but what is more crucial from the online-community-building perspective is that they frequently refer to each other by name (handle), more often than to people outside of this core. As a result, this group of nodes is characterized as having a higher density (a ratio of existing connections to the total number of possible connections) and higher reciprocity (a ratio of mutual connections, where each person mentioned one another at least once, to the total number of connections). These two factors often indicate stronger relationships among online members. [48] And having a strong community core is important because it may determine the longevity of an online community. Detecting the presence of a core group may indicate the presence of a successful online community. Furthermore, moderators of this or similar groups may use this information to decide whom to start following from this group, as well as whom to invite directly to participate in the next iteration of the book club.

In sum, the network representation of online interactions reveals interesting communicative patterns among members of this club, patterns that would not be possible to identify using content analysis alone. The network analysis also confirmed that Twitter was not used merely as a posting board, but rather as a platform to discuss and connect to other members of this reading community.


In this paper, we set out to understand the context of the specific reading experience on Twitter, a unique virtual environment that allows readers to communicate with one another across time and space in 140 characters or fewer. Through Netlytic and the analysis of the tweets, we see that reading models overlap across platforms and geographical areas. Consistencies between f2f and online reading communities are especially profound during the negotiations of choosing a book to read together. That is, choosing a book to read together in #1b1t mimics f2f-book club dynamics and mass-reading-event participants’ preoccupations. Furthermore, like f2f (and other on-line) book clubs, #1b1t had both extremely active commenters and also other readers who “dropped in” only occasionally. However, the #1b1t community seems to be a strong one, and one of the likely reasons for this strength is that Twitter allows for the book talk to include the author. The author-reader interaction, as seen in the social-network analysis, enhances reader relationships. At the same time, the strength of this community comes not just from the author-reader interactions, but also from the reader-reader interactions. It is because of the reader-reader interactions that the community core emerged around members who often replied to each other’s tweets. We attribute this to the Twitter’s ability to support the formation of weak social ties even among strangers in this group.

Like most book clubs and community-organized reading programs, there appears to be certain skill sets that people must have in order to participate in #1b1t. They must be able to read in English, they must be able to interact on Twitter, and they must have an understanding of the evolving discourse rules that may not reflect spoken language rules. This means that the Tweeters have to know how to talk about books. Consider, for example, that there were over 4,800 people who posted at least one tweet in our data set, but there were likely other Twitter users who simply followed the conversation without posting a single tweet using #1b1t. The readers must have the confidence to interpret what they are reading to an “audience” of thousands of other readers. This last point highlights an opportunity for further research. We think that #1b1t might be considered a quasi-private book club in that a reader does have to have a certain set of skills to participate. Readers may find pleasure in the shared reading experience, including a reader who only observes but does not comment, which can be easily done by following the designated hashtag. However, to participate fully in this platform, there is an element of performance. Further research would help us better theorize how this new technology might influence what gets said about a book, and indeed, how those utterances are taken up by the wider community.

While our findings provide new ways of interpreting contemporary reading practices, this study provides only a portion of that which is possible. Areas of future research could extend this analysis to include more data such as responses on the #1b1t blog posts, more hashtags (such as #NeilHimself), as well as Facebook posts. We could also examine interviews with the author, the organizers, or the readers. Or we could choose one specific question to zero in on, such as how the gender balance might be reflected in the tweets or how the readers thought about high-low cultural divides in the short-listed books. Future analysis might examine how the readers respond to the convergence of media. For example, we could analyze the data set to understand reader responses to the news that American Gods is set to be adapted for the big screen.

Book discussion on Twitter and other online environments offers a plethora of opportunities to researchers of reading, illustrating how readers interact in virtual communities. While some cultural critics might be worried about the death of reading, our analysis suggests that not only is reading alive and well, but also that there are exciting times ahead for reading research.

Parties annexes