Time to read: Exploring the timespaces of subscription-based audiobooks

Abstract

With remediation of the printed book into audiobook subscription services, reading by listening is becoming a popular alternative to reading by seeing. This article explores when people read by listening and whether there may be a shift regarding the places and times people read by listening rather than by seeing. Based on a considerable dataset from a Swedish subscription service for digital books, this article reveals that audiobook reading takes place at somewhat different times than expected and that subscribers read significant amounts each day. The findings indicate that the remediation into reading by listening using digital audiobooks may close the gender gap common in reading, as the reading practices of men and women are very similar, with men even reading slightly more than women. The reading practices of young adults are also similar to the larger population. Furthermore, the concepts stationary reading, mobile reading and stationary/mobile reading are introduced.

Full text

1
Author’s pre-print of article accepted for publication in New Media and Society, final version can be
found at: https://doi.org/10.1177/1461444819864691
Time to read: Exploring the timespaces of subscription-based
audiobooks
Elisa Tattersall Wallin and Jan Nolin
University of Borås, Sweden
Corresponding author: Elisa Tattersall Wallin, Swedish School of Library and Information Science,
University of Borås, Allégatan 1, 501 90 Borås, Sweden. Email: elisa.tattersall_wallin@hb.se
Abstract
With remediation of the printed book into audiobook subscription services, reading by
listening is becoming a popular alternative to reading by seeing. This article explores when
people read by listening and whether there may be a shift regarding the places and times
people read by listening rather than by seeing. Based on a considerable dataset from a
Swedish subscription service for digital books, this article reveals that audiobook reading
takes place at somewhat different times than expected, and that subscribers read significant
amounts each day. The findings indicate that the remediation into reading by listening using
digital audiobooks may close the gender gap common in reading, as the reading practices of
men and women are very similar, with men even reading slightly more than women. The
reading practices of young adults are also similar to the larger population. Furthermore, the
concepts stationary reading, mobile reading and stationary/mobile reading are introduced.
Keywords: Audiobooks, reading by seeing, reading by listening, streaming subscription
service, timespace, temporality, remediation, stationary reading, mobile reading,
stationary/mobile reading.
Introduction
This article investigates changes of when and where reading takes place when it is performed
by listening. In focus is an ongoing remediation of reading material as books are being
refashioned into streaming content on subscription services. It is an example of ”the
representation of one medium in another” (Bolter and Gruisin, 1999, p. 45). Though it is rare
that new media entirely replaces old media, essential elements and aspects are renegotiated.
The remediation of literature from print into audio involves written text being turned into
spoken words using a narrator, creating a re-representation with different modalities from the
original medium (Have and Stougaard Pedersen, 2016; Lundh and Smith, 2016). However,
the audiobook is not merely a remediation of the printed book, but also of the storytelling
traditions of oral culture and the practice of reading aloud to others, which has been present
for most of the history of reading (Cavallo and Chartier, 1999; Darnton, 2014). There has also
been a continuous reshaping of the audiobook since its inception, over time shifting from
records to cassette tapes and CDs (Colbjørnsen, 2015b; Rubery, 2016). However, the

2
remediation involved in contemporary practices of reading by listening is quite dramatic and
involves several aspects. Briefly put, Western societies have seen a noteworthy expansion of
reading by listening which can be connected to the development of new economic actors.
Given that, this text addresses the overall question: Given the remediation of the book, is there
a shift regarding the places and times when people read by listening rather than by seeing?
More specifically this article examines an exceptionally large dataset in order to investigate
three broad comparative issues. Although there has been research on the audiobook
publishing industry (Whitten, 2002; Colbjørnsen, 2015b), the audiobook as a new technology
for literature (Phillips, 2007; Rubery, 2011; Rubery, 2016), and audiobook reading (Have and
Stougaard Pedersen, 2016) there is still a substantial lack of previous research.
Reading is a concept that can be understood in many different ways. In this article reading is
defined as the process of making meaning from text, either by seeing or listening to the text.
This as a broad but, still, moderate understanding. Reading could be defined even more
broadly as interpreting, understanding or extracting meaning from various different contexts,
such as when watching a film, playing a video-game or listening to a podcast. Another form
of reading common today is the reading done online, such as when using social media,
reading online newspapers or fanfiction. These extended forms of reading are however not
studied in this article. The particular focus here is on books, both fiction and non-fiction, that
have been recorded, i.e. audiobooks. This type of reading is here referred to as reading by
listening, an alternative to the more established concept of reading by seeing.
Listening, just like reading, can be understood in various ways. While synonymous with the
faculty of hearing, listening can also be used as a metaphor for paying attention or
eavesdropping, even when using other senses than that related to the ear (Crawford, 2009).
Based on Michel Chion’s three different modes of listening (2012), we suggest that reading
by listening can be seen as a form of semantic listening, which entails making sense of spoken
language. This is a distinct form of listening separate from causal listening, i.e. to listen to a
sound to gather clues about its cause or source, and reduced listening, that is, focusing on the
characteristics of the sound itself and not interpreting cause or meaning. However, causal
listening may be used in combination with semantic listening as the tone of the speaker may
inform the meaning of the words (Chion, 2012). We suggest that reading by listening is a
specific form of semantic listening separate from other forms of semantic listening such as
those involved in conversations or consumption of vocal music.
This article engages with and makes contributions to three central issues within contemporary
reading research.
First, there is much speculation regarding if young people have different reading practices
compared to older generations (Ross, McKechnie and Rothbauer, 2018). The print-based
reading activities of young people, or lack thereof, have been given much attention (Moyer,
2011). Reports regarding traditional forms of reading from countries such as Australia and
Sweden declare that young people are reading less and that conventional kinds of literacies
are at risk (Merga and Moon, 2016; Läsdelegationen [The Reading Delegation], 2018). It has
been indicated that sustained long-form reading is becoming less common among the younger
generation (Rutherford, Merga and Singleton, 2018). Other studies have found that many
young adults read extensively using digital technologies (Ross, McKechnie and Rothbauer,
2018). The current text will discuss data regarding the question: Is reading by listening taking

3
place at different times and places, and to a different extent, for young adults compared to the
larger population?
Second, there are similar discussions regarding differences in reading practices of men and
women. Research reveals a gender difference concerning the amount men and women read as
well as the type of formats and genres they engage with (Bergström and Höglund, 2014;
Höglund and Wahlström, 2018; Ross, McKechnie and Rothbauer, 2018). The current article
will therefore discuss data that contributes to the questions: is reading by listening taking
place at different times and places, and to a different extent, for women and men?
Third, there are also persistent discussions regarding if there is a gender difference concerning
young adults, with girls assumed to be reading more than boys (Asplund and Pérez Prieto,
2017; Ross, McKechnie and Rothbauer, 2018). Therefore, the following query is of interest:
is reading by listening taking place at different times and places, and to a different extent, for
young adults of different gender?
Although these questions are broad in character, the large dataset discussed in this text is
specifically directed toward temporal aspects. It should be emphasised that the temporally
based data is used as a fundament for discussing places of reading as well, i.e. timespaces.
Parts of these reflections are speculative in character due to lack of data regarding places of
reading subscription-based audiobooks. Nonetheless, such speculations build upon
understandings of conventions regarding how people organise their lives. Data on patterns of
reading over the space of a day, week or a year, as collected for the current study, supplies
various clues to other dimensions of situated practice due to established societal routines.
The temporal dimension of reading has rarely been investigated. Rothbauer (2009) studied
teenagers reading and place, finding that many experienced temporal restraints on their leisure
reading. Respondents expressed that they would read more, if they only had the time for it
(2009). Similarly, Cedeira Serrantes (2016) found that lack of time was a major factor for
readers, some experiencing guilt when spending time reading rather than studying or working.
Comics proved to be a strategic choice from a time aspect for many, as comics would supply
a reading experience, without being as time-consuming as a novel (Cedeira Serrantes, 2016).
In a study on audiobook reading, Have and Stougaard Pedersen (2016) found that time
previously experienced as wasted by their participants, such as time spent commuting,
became quality time because of their use of audiobooks.
In the following, the remediation of reading through audiobook subscription services will be
introduced as an intervention into the routinized practices of reading handheld documents on
paper. Thereafter, the notions of temporality and timespace will be introduced. The methods
section discusses collection of a unique and very large dataset obtained from BookBeat, a
Swedish audiobook subscription service. Following this, the results of the study will be
presented. The discussion will deal with broader implications: given the data on temporality,
what can be implied regarding changes of the spaces and activities of reading? The article
ends with the conclusion.

4
Audiobook subscription services
Audiobooks are not new inventions. For well over a century books have been recorded on
different technologies and for various different groups of readers. Whether as talking books
on phonograph records for blind people in the early 20th century or as books on tape for
commuters during the 1970’s, recorded books have long faced criticism as an illegitimate or
lazy form of reading (Rubery, 2016). Constraints of the various formats have also
encountered disapproval, as neither records, cassette tapes nor CD’s could hold an entire
novel on one tape or disc. This resulted in either highly abridged audio versions of the printed
books, or in large sets containing anywhere between five and 80 tapes or CD’s for one book
(Rubery, 2016), making it cumbersome in several ways for the reader.
The remediation into subscription-based digital audiobooks brings with it more changes and
possibilities compared to the previous technologies. Not least as it is possible to store one
audiobook in a single digital file. In this sense it can be seen as a remediation not only of the
audiobook into a new format, but also as a remediation of the printed book with the many
dramatic changes in distribution, economic actors, practice and readership. The changes that
come with the remediation from the printed book into the subscription-based audiobook
include:
• Artefact: from paper to intangible streams.
• Active sense: from eyes to ears. We will in the following talk about, and contrast, reading
by listening with reading by seeing.
• Mobility: with content available in ever present smartphones, a wide-ranging catalogue of
books is always carried with the user.
• Distribution and economic model: from local bookshops or libraries to centralized content
providers.
Traditionally, creative works such as literature, music and film have been communicated
through distinctly different types of media, such as books and records. However, with further
development of first the digital book and then cloud-based services, these traditional
distinctions are collapsing. Literary works, music and films are converging, all being
streamed in parallel ways. From the cloud-based service perspective all creative output can
uniformly be described as content. This is a vital shift for the understanding of reading by
listening. Users of streaming services do not need to think of literary works as books with a
material presence that must be used in certain ways. Instead, the book can be consumed as
content similar to other streamed listening material such as podcasts, radio and music.
Similarly, previous formats of the audiobook, with its cassette tapes or CD’s had a significant
material presence that dictated its use.
Critically, streaming content is intangible. It is not possible to touch or carry a stream around,
only a device through which the stream can be provided, such as a smartphone. A digital book
is not something that can be placed on a bookshelf or table. Therefore, remediation generates
fundamental challenges to the traditional book market. When practices shift toward intangible
content such as e-books and streaming audiobooks, people tend to acquire books more
according to the model of the library rather than that of the bookshop. In the library model,
users deal with books through borrowing and returning (Söderholm, 2018).

5
Digital audiobooks on the commercial market are accessible through platforms referred to as
streaming services, the book counterparts to Netflix and Spotify. However, streaming services
may not technically be an entirely accurate label, as books also can be downloaded and
listened to offline. Instead, we refer to them as subscription services as most audiobook
platforms follow a premium subscription model, where users sign up and pay a monthly fee to
access all the content. Some other media streaming services, such as Spotify, follow a
freemium/premium model (Colbjørnsen, 2015a), where users can choose to either use the
service for free or become paying premium subscribers. Premium users gain access to the
entire catalogue, can download material to use offline, and are not interrupted by
advertisements. Most audiobook subscription services currently active on the market operate
under the premium model. There are, however, differences between the various platforms.
Amazon owned Audible differs from most other services with their restricted premium model.
For their monthly fee users get to pick one audiobook to download and keep, with the option
of purchasing more books if desired. Other services operating primarily on the European
market, such as BookBeat and Storytel, follow the premium model well established by Spotify
and Netflix, where users get access to the entire catalogue for their monthly fee. This is a
noteworthy difference between separate models. In a sense, Audible attempts to uphold the
traditional ideals of the bookstore. Books are from this perspective seen as individual units
and are bought, not borrowed. Even though users are given the status of buyers, the book is
still an immaterial artefact situated on a platform account. There is limited value connected to
that kind of ownership. Storytel and BookBeat have more clearly adopted the notion of book
as content. This concept also resembles the library model, as books are borrowed by the users
instead of bought and owned. It is possible that the usage of the services differs due to the
different premium models. Subscribers of services offering unlimited access to the catalogue
for their monthly fee may have different usage patterns, or may choose to read more,
compared to subscribers of Audible, where users need to pay more to read additional books.
Therefore, various distribution models afford different reading practices.
In Sweden, where this study was performed, the market for digital book subscription services
is growing quickly, both regarding more market actors and increases in subscribers (Dahlgren,
2018). In 2017 it was estimated that 6 percent of the Swedish population subscribed to a
digital book service, consequently 600 000 people out of a population of 10 million
(Nordicom, 2018). A recent extensive survey of Swedish media habits revealed that 30
percent claimed to be regular consumers of audiobooks (Wallin, Tattersall Wallin and
Gunnarsson Lorentzen, 2019). Besides availability of subscription services, readers also have
the option to purchase audiobooks in bookshops or borrowing them from libraries, both in
physical and digital formats. Swedish public libraries now also allow patrons free of charge
access to audiobooks and e-books through a smartphone app (Wallin, 2019). Within the
commercial market, the three largest subscription services for digital audiobooks and e-books,
Storytel, BookBeat and Nextory, represent 15.5 percent of the entire Swedish book market
(Wikberg, 2019). BookBeat, whose transaction logs forms the dataset for this article, is owned
by Swedish media conglomerate Bonnier. Having entered the market in 2016, BookBeat had
in 2019 approximately 150 000 subscribers in four countries: Sweden, Finland, Germany and
the United Kingdom, with Sweden holding the majority of the subscribers (Dahlgren, 2019).
A majority of BookBeat’s users in Sweden are female, over 75%, whilst young people
between 18 and 20 are one of the smallest groups of subscribers (Sandin, 2018). The average

6
BookBeat subscriber is 38 years old, although there is a fairly even distribution of subscribers
ranging from their 20’s to their 50’s (Sandin, 2018).
A practice-based perspective on timespace
Human activities are always situated within spatial and temporal arrangements. Since the
introduction of clock time in the 14th century, the temporal ordering of daily life has become
increasingly well-structured (Dorhn-van Rossum, 1996; Adam, 2006). Clocks and calendars
are used as measurements and trackers of time (Adam, 2006; McKenzie and Davies, 2010).
Days are divided into working time and leisure time, schools run during certain hours of the
day, and the end of one working week is marked by the arrival of the weekend (Dorhn-van
Rossum, 1996; Adam, 2006). Clock time and calendar time are cyclic in character and the six
basic cycles are minutes, hours, days, weeks, months and years. When investigating the
doings of reading by listening, three of these will be in focus. These can be conceptualized as
three different cycles of time: daytime, weektime and yeartime. Data regarding these different
cycles supplies separate forms of insights regarding reading by listening and timespace.
The spatio-temporal audiobook practices can be studied through the lens of Schatzkis activity
timespace theory (2010). Based on the work of Heidegger, activity timespace is opposite
from, and can act as a complement to, the separate concepts of objective time and objective
space. Objective time and space is, according to Schatzki (2010), autonomous from both
human action and consciousness. Whilst humans may both perceive and act on objective time
or space, it would also occur without them. The invention of the clock and the calendar is the
prime example of how people may measure objective time, and organize themselves
according to its succession (Schatzki, 2010). Though often understood as two separate
phenomena, Schatzki notes that time and space are connected, and calls unified time and
space activity timespace.
These timespaces are the fundamental features of human activities (Schatzki, 2010). Data
related to daytime, weektime and yeartime supplies different insights regarding what people
do with their time. Even though temporal data will be in focus in this study, the theory of
timespace suggest that time and space are intrinsically connected and together they form the
essential aspect of action. Humans navigate objective time through routines, i.e. we tend to do
the same kind of activities at assigned places such as at home, at school, in a restaurant or on
public transport. While quantitative temporal data, such as collected for this article, is
incomplete it can still serve as a foundation for some guesswork regarding situated practices.
Timespaces are either derived from, or are dependent on, social practices (Schatzki, 2009).
Even though activities may be similar, they are characterized, and individualized, by the
objective times and spaces in which they occur.
In this article, a quantitative dataset will be used to infer clues to the temporality of the
activities of remediated reading by listening. Involved activities take place in literally
thousands of places in Sweden and involving a myriad of different practices. Nonetheless,
data on the temporality of audiobook reading will supply insights on the possible timespaces.

7
Method
In order to acquire data about when people listen to streaming audiobooks, it became
necessary to gain access to transaction logs detailing the use of a digital audiobook
subscription service. Such non-intrusive methods have been used to collect information on
user behaviour in various digital collections (Huurnink, Hollink, van der Heauvel et al.,
2010), including study of temporal aspects of use (Jansen, Spink and Pedersen, 2005). This
method was chosen as it details the actual usage of the chosen platform, and is able to reveal
user patterns from a large group of users. This type of information would be difficult to obtain
in any other way. There are ethical problems involved in collecting personal data in this kind
of investigation. In order to avoid such issues, it was decided not to attempt access to the
transaction logs themselves. Rather, queries would be provided to a subscription service,
which would supply the authors with statistical results. Therefore, no personal data was
collected.
A collaboration was set up with BookBeat, a Swedish subscription service for digital books
active in several European countries. In this study, only the users on the Swedish market
during the 12-month period between April 2017 and March 2018 are included. BookBeat was
unwilling to share the exact number of subscribers during the time of data collection,
probably due to the competitive market. This is one of the drawbacks of collaborating with a
company. However, this is compensated by a much larger dataset than would probably be
possible otherwise. Though it is difficult to say exactly, it is probable that the data consists of
at least 80 000 users. This is an exceedingly large dataset compared with most conventional
studies on media habits, which usually build upon a few thousand respondents.
Four questions were formulated by the authors that would make it possible to understand the
specific patterns of reading among young people, men/women and the whole group of
subscribers. The cyclic notions of daytime, weektime and yeartime were used to understand
data on reading in connection with temporality. The four questions could therefore be
formulated the following way:
1. How many minutes per day does the average BookBeat user read audiobooks? What is the
use of the average man/woman in comparison to the average young man/woman?
2. How is the reading distributed over the day for the average BookBeat user? Is there a
difference in use between the average man/woman and young man/woman?
3. How is the reading distributed over the days of the week for the average user? Is there a
difference between the average man/woman and young man/woman?
4. How is the reading distributed over a year for the average user? What is the use of the
average man/woman in comparison to the average young man/woman?
Informed by these questions, BookBeat carried out log searches within their system, providing
the authors with statistical data in the form of averages. This data did not contain enough
information to calculate median levels or standard deviation. Visualisations of the data was
created by the authors, offering a descriptive overview of the time-related usage of
audiobooks on the BookBeat platform.
Limitations and ethical considerations

8
It is worth noting that the people in this study are presumably very interested in reading, as
they are willing to pay a monthly fee to a subscription service for digital books, and must
therefore not be taken as a representative sample of the larger population. The pricing for the
audiobook platforms active on the Swedish and European market is currently set at about 40–
50 percent higher than the prices for streaming services offering film, tv and music.
The category of “young men/women” is always problematic. Initially the intention was to
collect data on 16-19-year olds, the age group typically in upper secondary school in Sweden,
with the larger group of all users. However, since users need to be 18 to register with
BookBeat, it would not be possible to specifically study younger teenager in this case.
Therefore, the age group of young adults was chosen to be people between ages 18-20.
Included into the category of all users was anyone with a Swedish BookBeat account who
listens to audiobooks.
Whilst the data reveals temporal aspects of audiobook use, it cannot reveal how, or in what
places or situations, that audiobook reading was carried out. It is also impossible to know
whether the user was actively listening to the book, and if it was the owner of the account
doing the listening. It is possible that a subscriber may share the access to the account with
another person, for instance a parent might let their child listen to an audiobook using the
parents account. However, since it would only be possible to play one book at a time on a
single account, multiple users could not use the same account simultaneously, unless listening
together in the same time and place. Nonetheless, it is likely that listening is carried out
predominantly by the account holder. Whilst being aware of these limitations, the data
gathered will be used following two assumptions: 1) The person playing the book had the
intention of reading, and so all minutes will be considered actual reading minutes, and 2) It
will be assumed that the person registered to the account is the one doing the listening. These
assumptions are quite reasonable given the extensive amount of data involved.
This data was collected by BookBeat and provided to the authors during the spring of 2018,
before the implementation of the new European data protection law, GDPR. Again, it is
important to note that individual user’s transaction logs were not studied or accessed by the
authors. Nor has there been any focus on what type of books the users read. Instead, the study
has only been concerned with temporal audiobook usage patterns of the whole group of
subscribers. At the time of data collection, it was stated in the service user agreement that the
customers usage of the platform would be documented and used for development, marketing
or personalized recommendations. It was also stated that BookBeat may share information
with their collaborators.
Results
In the following, four figures based on the transaction logs provided by BookBeat will be
presented and analysed.
Daytime reading
During a cycle of daily routine various activities take place. Some of these are routinized and
performed similarly or almost in the same way every day; such as wake up activities, meals
and bedtime routines. There are other routines that are different depending on work/school
day or weekend. Traditionally, reading has been part of daily routines for many people, such

9
as reading before going to sleep. In the case of reading by listening facilitated by subscription
services, it should be noted that reading material is ubiquitously present for smartphone
owners.
Figure 1 is concerned with “daytime”, i.e. how many minutes per day, on average, that
different BookBeat user groups listen to audiobooks. The data is from the 12-month period
starting with April 2017. Young women between ages 18 and 20 listen on average 90 minutes
per day, whilst young men of the same age group listen 100 minutes per day. Meanwhile, the
larger group of all women with a BookBeat account listen 94 minutes per day and all men 98
minutes. Given the questions that we want to investigate: generational and gender differences,
it is striking that there are only marginal statistical differences among these groups.
Interestingly young men have the highest daytime reading of all groups, with men of all ages
close behind. This means that the men in this study on average listen slightly more than
women, which is noteworthy considering the common notion that men read substantially less
than women (Asplund and Pérez Prieto, 2017; Ross, McKechnie and Rothbauer, 2018;
Höglund and Wahlström, 2018). This data shows that all the groups studied in this material
average around the same amount of reading time on a daily basis. This is notable as, again,
studies on reading books tend to show a gender gap. The current study implies that
remediation from reading by seeing to reading by listening using a subscription service may
serve to close this gender gap.
The second major takeaway from the statistics is the substantial time involved. 90-100
minutes per day on average is a considerable amount of time spent reading every day,
contrary to common claims that young adults and men read very little. Since the amount
presented here is the average amount, there will also be BookBeat users who read less and
users who read more.

10
As of writing, there is no other study to compare with in regards to the quantity of time actual
users spend specifically on audiobook services each day. However, it should be emphasized
that this is a large number even when compared to other activities as measured in other
studies. According to a recent large Swedish survey, the average person spends 22 minutes
reading books of all formats daily, with audiobooks representing only 3 of these minutes
(Nordicom, 2018). This can be compared with 50 minutes on social media, 52 minutes
playing music and 97 minutes watching television. Young adults ages 15-24 are fairly similar
to the larger population concerning books and television, but spend exceedingly more time on
social media and music, with 121 minutes per day on social media and 109 minutes listening
to music (Nordicom, 2018). These figures represent estimations of time spent using various
media from the larger population, including non-users, as opposed to the statistics gathered
for this article which only represent users of audiobooks on the BookBeat platform.
In studies by Rothbauer (2009) and Cedeira Serrantes (2016), the participants expressed a
clear wish to read more, but felt they lacked the time for it. Their findings show that many
people want to read more than they actually do, and actively attempt finding solutions in order
to be able to do so, in one form or other. The remediation of the printed book into the
subscription-based digital audiobook makes it easier for people to read whilst doing other
things. By reading by listening, and by having the audiobook streaming through a
smartphone, the eyes and hands of the reader are unrestricted, except when using the app to
start or stop the book. This offers them the possibility to engage in other activities whilst also
reading their book, thus perhaps finding time for reading where there was not time before.
Results from the current study, then, implies that remediation from reading by seeing to
reading by listening may serve to increase daily time spent on reading by average persons.

11
Figure 2 allows for a closer and somewhat different view of “daytime”, illustrating how
audiobook listening using BookBeat is distributed over an average day in January 2018. At
first glance, it becomes apparent that the reading is distributed quite evenly throughout the
day, although with a dip between 1am to 5 am for all users. This decline in reading during the
night is to be expected, as most are likely to be asleep. There is also a slight decline starting
from 5 pm until about 9 pm. This drop in audiobook reading is interesting, as it falls during a
time that would be considered traditional leisure time. Most people would be out of school or
work after 5 pm, and it could be expected that a leisure activity such as reading would be
carried out during this time. Between 10 pm until about midnight there is a rise in reading
again, which assumedly would be bedtime reading.
There is a surprisingly continuous reading activity going on during the day, from about 7 am
to 4 pm. There are some variations between the groups during this time, with the group of all
women reading slightly less than the others, young men slightly more, and all men and young
women somewhere in the middle. During the daytime it could be expected that a majority of
people would be in either school or work. Though a smaller part of the user groups may be
out of education and work, or working during odd hours, they should not significantly impact
the statistics for the whole group. Therefore, it must be assumed that many people listen to
audiobooks during the day, possibly during breaks and at lunchtime. Some people may also
have jobs where audiobook use is acceptable during the working day, and some students may
be allowed to listen to audiobooks during lessons. It is possible that the flexibility and
portability of digital audiobooks mean that they could listen for a few minutes here and there,
instead of just for long stretches of time. However, these are only speculations as to why the
data looks the way it does, the timespaces of these listening activities would need to be
researched further in order to better understand what is going on behind the figures.
Weektime reading
While a substantial amount of activities are routinized, some may be distributed unevenly
over the cycle of the week. Working days Monday-Thursday are for many people similar in
character. This should particularly be evident in such a large dataset as is examined in this
study. It is common that activities during Friday, often the last day of the working week,
include somewhat different activities. The weekdays of Saturdays and Sundays are similar but
can also vary in some ways. Again, these are broadly adopted norms within Western societies.
As such, we expect them to be formative for the data. Individual patterns from people
working shifts or work on the weekend will be embedded in the data but not as explicit as the
most broadly available patterns.

12
Figure 3 concerns “weektime” during the twelve-month period between April 2017 and
March 2018 on the BookBeat platform. Women of all ages have the most evenly distributed
reading over the week, with only the slightest decline over the Saturday. The other groups of
young users and all men read more during the weekdays, with a clear decline in their
audiobook use during the weekend. Young men stand out slightly with a steeper decline in
reading over the weekend, and a more distinct increase in listening during Wednesdays.
The weekend decline is interesting in relation to what was revealed in Figure 2, where it
seemed like a large amount of listening is carried out during daytime, in what would be
considered the school day or working day, or in close proximity to it. If the listening is related
to school or work, this would make the decline over the weekend understandable, as most
would be out of school or work during Saturday and Sunday. It is possible that there is a
social aspect to the drop in listening during evenings, shown in figure 2, and the drop during
weekends, shown in figure 3. These traditional leisure times may be when people are
engaging socially with their friends or families, or taking part in other activities where
audiobook reading would not be suitable.
It is also possible that the various routinized activities of the weekdays accommodate for
audiobook reading better than weekend routines. Possibly, audiobook reading, with its mobile
affordances, is related to being in motion, such as commuting, exercising or doing errands. If
this is the case, perhaps people choose to engage in others forms of reading when they are at
home and have the opportunity to sit and hold a book in their hands. Perhaps, reading by
listening is not competing with print or e-book reading, but rather with other types of audio
such as listening to podcasts, radio or music in situations where headphones are suitable, but
when perhaps reading by seeing is not.

13
Yeartime reading
Figure 4 displays “yeartime” reading on the BookBeat platform from April 2017 to March
2018. Young users differ from the larger group of users in that their reading is much more
varied throughout the year, with greater increases and declines compared to users of all ages.
The female and male groups of all users display a more even listening distribution throughout
the year, although with a moderate increase over the summer months and at the start of the
year. The young users have a strikingly visible rise in listening throughout the summer
months, with young women listening more during July and August, and the listening of young
men significantly increasing during the month of August. There are also other, more modest,
spikes in reading throughout the year, such as young men’s increase in October and
November, and an increase in listening for young women in January, following a decline
between November and December, similar to that of the groups of all men and women.
The stark increase of young people reading during the summer months stands in contrast to
what was shown in figures 2 and 3, where it seemed young people did most of their reading
during the day on weekdays, appearing as though their reading was disconnected from leisure
time, rather linked to the school day or the working day. Between the months of June and
August, Swedish schools and universities are closed for holidays, and the month of July is
traditionally a holiday month for people in the workforce as well. Many young people do have
summer jobs during their holiday. One possible explanation to the stark increase in audiobook

14
reading is related to summer jobs. It is feasible that many young people are employed to do
routine tasks where it is possible to listen to a book whilst working. An alternative
explanation can be that young people out of school have other timespace practices during the
summer, resulting in different reading practices.
It is also worth noting the slight increase for young women and all men and women between
December and January, this is also a time where many people have time off. Schools usually
shut down for two weeks over the Christmas and New Year period. Contrary to the other
groups, young men have a small decline in reading during this period. There are difficulties,
as noted before, in explaining such trends in the data. It is only possible to speculate without
access to more data.
It is quite notable that yeartime patterns of reading for all men and all women are practically
identical. Again, this study tends to indicate that remediation from reading by seeing to
reading by listening may serve to close the gender gap.
Though perhaps obvious, it is worth noting that the statistics for the whole groups of
BookBeat users reveal little about individual usage. One person may be doing all their reading
as bedtime reading only, whilst another is listening in short spurts throughout the whole day.
The spikes in the various figures represent times where a large group of people read
audiobooks at the same time.
Discussion
Conventional reading by seeing tends to be restricted in various ways. These restrictions
become evident when alternative formats and practices emerge such as reading by listening
using subscription-based audiobooks. Even though it is possible to perform reading by seeing
when standing or walking, it is cumbersome. To a substantial extent, traditional reading
practices have been performed either sitting or lying down. The visual senses need be totally
focused on the words on print and the primary orientating appendixes of humans; hands and
arms, are usually engaged by holding the book. Contrary to this, reading by listening frees up
both the visual sense and the hands for other tasks. In addition, there is no need to sit or lie
down. It is perfectly possible to walk around and perform various tasks.
What emerges is therefore almost the inverse of traditional reading by seeing. Earlier it was
mostly performed seated or lying down. With reading by listening it instead becomes mostly
standing or walking, as audiobook use is a highly mobile form of reading. This is one way in
which the timespaces of reading by seeing, which tends to be stationary, and reading by
listening, with the potential of mobility, differ. The possibility of mobility afforded by reading
by listening creates different timespaces. Although the audiobook has always been more
mobile than the printed book, books on tape became popular with commuters in the 1970’s
(Rubery, 2016), previous formats were less mobile and much more inconvenient to use on the
go compared to the subscription-based audiobooks of today.
In order to discuss the practices connected to the remediated book it is useful to introduce two
new concepts, describing two modes of reading: stationary reading, which can be associated
with reading by seeing and mobile reading, associated primarily with reading by listening.

15
With reading by listening it becomes possible to engage with a narrative while driving, biking
or walking, for instance. That said, there are certain practices of reading where mobile and
stationary reading overlap. In certain mobile situations it is possible to read by seeing, such as
when riding on a bus or train. This thus becomes stationary/mobile reading. These different
modes of reading are illustrated with the “reading mobility shift model” in Figure 5 below.
Figure 5 illustrates that the remediation of the printed book is connected to a mobility shift.
This also involves a shift in focus. Reading by seeing tends to need a different type of focus
compared to reading by listening. Traditional visual reading practices have a strong
dependence on the eyes and body of the reader. Simply put, if the eyes or body of the reader
is needed elsewhere at the same time, they need to stop reading. There is also a particular
form of decoding of text carried out when reading with the eyes, which demands
concentration. This does not lend itself well to also carrying out other tasks at the same time.
Reading by listening meanwhile does not demand the use of the eyes or the body, meaning
the reader can focus on other things at the same time as listening to a book. However, a
different form of focus is needed when reading by listening, without this concentration the
reader would lose comprehension quickly. In both cases, the reader needs to create meaning
from the book and it is this similarity that marks both these activities as reading in this study.
There are different well-established notions of time in society, such as the division between
working time and leisure time. It is expected that during working time, traditionally 8–4 or 9–
5 on weekdays, people engage in work or education. Evenings and weekends are often
considered leisure time, or free-time, as people are out of work or school and may engage in
whatever activities they choose. However, it appears as though in the digital age, leisure time
activities are changing (Vilhelmson, Elldér and Thulin, 2018) and the separation between
these times are becoming blurred. Digital culture makes it possible to work from home, and
carry out leisure activities at work or in school (Agger, 2011; Vilhelmson, Elldér and Thulin,
2018). This is also indicated in the findings of this article, where it is revealed that BookBeat
users listen to audiobooks to a surprising degree during daytime on weekdays, with less
during evenings and weekends. This continuity in reading over different times throughout the
day, and during the working day, could in part be due to the remediation of the printed book,
and analogue audiobook, into subscription-based audiobooks. As books are repackaged as
streaming content, they become accessible in new ways. Digital audiobooks are especially

16
mobile, which is illustrated in the mobility shift model. This implies shifts in the timespaces
of reading that require much further study.
Conclusion
Subscription-based audiobooks are growing substantially in popularity, not least on the
Swedish market where this study was performed. This is a significant remediation of not only
previous audiobook formats and the ancient tradition of listening to books being read aloud,
but of the printed book and the practice of reading by seeing. The rise of subscription services
for digital audiobooks is making reading by listening an increasingly common mode of
reading. This study has discussed empirical data collected about reading by listening using
BookBeat, a subscription service for digital audiobooks. With the lively emergence of
subscription-based services overall, and the lack of attention in previous research to
audiobook services, the results presented here must be seen as substantial and, to some extent,
surprising.
With a focus on the timespaces of reading by listening, and especially on the temporalities on
audiobook reading during daytime, weektime and yeartime, this study revealed that users
listen to audiobooks during slightly different times than expected. Due to broadly adopted
routines in western societies, certain temporalities are connected with certain spaces and
activities. The patterns of daytime and weektime were fairly similar across the different
groups studied, and a large part of the reading was carried out during, or in relation to, the
working day. However, there was a marked difference as to yeartime reading practices.
Particularly, young people had significantly higher increases in reading during the summer
months, in direct relation to the summer holiday. Apart from this, there was no marked
difference in the reading patterns of young people compared to the larger group of all users of
BookBeat. The reading was very similar both in distribution over daytime and weektime, and
in the amount being read each day on average. Furthermore, there was no significant gender
difference in the reading practices revealed by the data. Men spent a little more time reading
each day compared to women, but their reading was otherwise very similar in regards to the
distribution over daytime, weektime and yeartime. There was however a slight dissimilarity
between young men and young women, with young men reading on average 10 minutes more
per day compared to young women. This reveals that the remediation into reading by
listening using subscription-based audiobooks may close the gender gap so common in
traditional forms of reading by seeing.
Furthermore, this study found that the users of BookBeat spent significant amounts of time
reading, between 90–100 minutes per day on average. This implies that the remediation from
reading by seeing to reading by listening may serve to increase the daily time spent reading,
especially due to the mobility of subscription-based audiobooks.
The difference in mobility between reading by hearing compared to reading by seeing
prompted the introduction of three concepts in this article, stationary reading, mobile reading
and stationary/mobile reading. Whilst stationary reading mainly was associated with
traditional forms of reading by seeing, mobile reading was in turn associated with reading by
listening. Stationary/mobile reading refers to travelling situations where there is a choice
between the two modes. When travelling as a passenger it is equally doable to read by seeing
as by listening. For instance, sitting on a bus, it is possible to select the preferred reading

17
mode. Buses all over the world have traditionally been places where passengers have engaged
with various reading materials such as books, newspapers and magazines. However, today
there is a choice. The reader frequently chooses to don a headset. Indeed, the listening
equipment may already have been engaged before boarding the bus.
Acknowledgements
The authors are indebted to BookBeat for production of data from transaction logs. We also
would like to thank Dr Mats Dolatkhah for important input on reading practices and
temporality at an early stage of this study.
References
Adam B (2006) Time. Theory, Culture & Society 23(2-3): 119–138.
Agger B (2011) iTime: Labour and life in a smartphone era. Time & Society 20(1): 119–136.
Asplund S-B and Pérez Prieto H (2017) Young working-class men do not read: or do they?
Challenging the dominant discourse of reading. Gender and Education 30(8): 1048–1061.
Bergström A and Höglund L (2014) A national survey of early adopters of e-book reading in
Sweden. Information Research 19(2).
Bolter JD and Gruisin R (1999) Remediation: understanding new media. Cambridge, Mass:
MIT Press.
Cavallo G and Chartier R (eds.) (1999) A history of reading in the West. Oxford: Polity.
Cedeira Serrantes L (2016) When Comics Set the Pace. The Experience of Time and the
Reading of Comics. In McKechnie L, Skjerdingstad KI, Rothbauer PM et al. (eds.) Plotting
the reading experience: theory, practice, politics. Waterloo, Ontario: Wilfrid Laurier
University Press, pp. 217–232.
Chion M (2012) The Three Listening Modes. In Sterne J (ed.) The Sound Studies Reader.
New York: Routledge, pp. 48–53.
Colbjørnsen T (2015a) A Spotify for …Whatever. A Cross-Industrial comparison of
Streaming Services in the Media Industries. In: 4th International Symposium on Media
Innovations, Brussels, Belgium, 2 June 2015.
Colbjørnsen T (2015b) The accidental avant-garde: Audiobook technologies and publishing
strategies from cassette tapes to online streaming services. Northern Lights 13(1): 83–103.
Crawford R (2009) Following You: Disciplines of Listening in Social Media. Continuum:
Journal of Media and Cultural Studies 23(4):525–535.
Dahlgren S (2018) Storytel passerar 300 000 abonnenter i Sverige - streamingmarknad värd
650 Mkr? In: Boktugg. Available at: http://www.boktugg.se/2018/01/19/storytel-passerar-

18
300-000-abonnenter-sverige-hela-streamingmarknaden-vard-650-mkr/ (accessed 28 May
2019).
Dahlgren S (2019) Bookbeat nu över 150 000 abonnenter, ska växa ännu mer 2019. In:
Boktugg. Available at: http://www.boktugg.se/2019/01/02/bookbeat-nu-over-150-000-
abonnenter-ska-vaxa-annu-mer-2019/ (accessed 28 May 2019).
Darnton R (2014 [1986]) The first steps toward a history of reading. Australian Journal of
French Studies 51(2): 152–177.
Dohrn-van Rossum G (1996) History of the hour: clocks and modern temporal orders.
Chicago: University of Chicago Press.
Have I and Stougaard Pedersen B (2016) Digital audiobooks: new media, users, and
experiences. New York: Routledge, Taylor & Francis Group.
Huurnink B, Hollink L, van den Heuvel W et al. (2010) Search Behaviour of Media
Professionals at an Audiovisual Archive: A transaction Log Analysis. Journal of the
American Society for Information Science and Technology 61(6): 1197–2010.
Höglund L and Wahlström E (2018) Traditionell och digital bokläsning. In Andersson U,
Carlander A, Lindgren E et al. (eds.) Sprickor i fasaden. SOM-undersökningen 2017.
Göteborg: SOM-institutet, pp.249–264.
Jansen BJ, Spink A and Pedersen J (2005) A Temporal Comparison of AltaVista Web
Searching Journal of the American Society for Information Science and Technology 56(6):
339–570.
Lundh AH and Smith B (2016) Two King Lears: The Meaning Potentials of Writing and
Speech for Talking Books. Proceedings from the Document Academy 2(1), article 6.
Läsdelegationen [The Reading Delegation] (2018) Barns och ungas läsning - ett ansvar för
hela samhället (SOU 2018:57). Stockholm: Kulturdepartementet.
McKenzie PJ and Davies E (2010) Documentary tools in everyday life: the wedding planner.
Journal of Documentation 66(6):788–806.
Merga MK and Moon B (2016) The impact of social influences on high school students’
recreational reading. The High School Journal 99(2): 122–140.
Moyer JE (2011) What Does it Really Mean to ”Read” a text? Journal of Adolescent & Adult
Literacy 55(3): 253–256
Nordicom (2018) Nordicom-Sveriges mediebarometer 2017. Göteborg: Nordicom.
Phillips D (2007) Talking books: The Encounter of Literature and Technology in the Audio
Book. Convergence 13(3):293–306.
Ross CS, McKechnie L and Rothbauer PM (2018) Reading still matters: what the research
reveals about reading, libraries, and community. Westport, Conn.: Libraries Unlimited.

19
Rothbauer PM (2009) Exploring the Placelesseness of Reading Among older Teens in a
Canadian Rural Municipality. Library Quarterly 79(4): 465–483.
Rubery M (ed.) (2011) Audiobooks, Literature, and Sound Studies. New York: Routledge.
Rubery M (2016) The Untold Story of the Talking Book. Cambridge, Mass.: Harvard
University Press.
Rutherford L, Merga MK and Singleton A (2018) Influences on Australian adolescents’
recreational reading. Australian Journal of Language and Literacy 41(1): 44-56.
Sandin N (2018) On the hunt for new audiobook listeners? Let data lead the way. In The
Bookseller. Available at: https://www.thebookseller.com/futurebook/hunt-new-audiobook-
listeners-let-data-lead-way-713131 (accessed 28 May 2019).
Schatzki T (2009) Timespace and the organization of Social Life. In Shove E, Trentmann F
and Wilk RR (eds.) Time, consumption and everyday life: practice, materiality and culture.
Oxford: Berg. pp 35–48.
Schatzki TR (2010) The Timespace of Human Activity. On Performance, Society, and History
as Indeterminate Teleological Events. Lanham: Lexington Books.
Söderholm J (2018) Borrowing and lending tools. The materiality of X-lending libraries. PhD
thesis, University of Borås, Sweden.
Vilhelmson B, Elldér W and Thulin E (2018) What did we do when the Internet wasn’t
around? Variation in free-time activities among three young-adult cohorts from 1990/1991,
2000/2001, and 2010/2011. New Media & Society 20(8): 2898–2916.
Wallin B (2019) Selling and Lending E-books: Changes in the Swedish literary field. PhD
thesis, University of Borås, Sweden.
Wallin B, Tattersall Wallin E and Gunnarsson Lorentzen D (2019) Bokläsning och den
svenska bokmarknaden. In Andersson U, Rönnerstrand B, Öhberg P et al. (eds.) Storm och
stiltje. Göteborgs Universitet: SOM-institutet
Whitten R (2002) Growth of the Audio Publishing Industry. Publishing Research Quarterly
18(3):3–10.
Wikberg E (2019) Bokförsäljningsstatistiken, helåret 2018. Stockholm: Svenska
bokhandlarföreningen och Svenska förläggareföreningen. Available at:
https://www.forlaggare.se/sites/default/files/rapport_helar_2018_webb.pdf (accessed 28 May
2019).

20
Author biographies
Elisa Tattersall Wallin is a doctoral student at the Swedish School of Library and Information
Science, University of Borås, Sweden. Her research focuses on the temporality of audiobook
reading practices.
Jan Nolin is a professor at the Swedish School of Library and Information Science, University
of Borås, Sweden. His research interests include social media studies, transparency
movements, and strategies concerning academic involvement in Agenda 2030.