Content uploaded by Michele Zappavigna
Author content
All content in this area was uploaded by Michele Zappavigna on Jul 17, 2019
Content may be subject to copyright.
Pre-proof – please so not quote this version. Published as:
Zappavigna, M. (2017). Ambient liveness: Searchable audiences and
second screens. In Hight, C & Harindranath, H. Studying Digital Media
Audiences: Perspectives from Australasia. London: Routledge. pp 150-
172.
10. Ambient liveness: Searchable audiences and second screens
Michele Zappavigna
Abstract
Audiences are becoming increasingly ‘searchable’. A key semiotic resource
supporting such searchability is the hashtag, a form of social tagging allowing
microbloggers to embed metadata in posts. This ‘conversational tagging’1 supports
forms of ambient liveness in terms of how people communicate opinion and sentiment
in ‘real-time’ in ways that are findable by others. While popularly thought of as topic-
markers, hashtags do much more than label posts or label what audiences are talking
‘about’: they construe a range of communicative functions. In this chapter I employ
Halliday’s 2 concept of linguistic metafunctions, to explore how hashtags enact three
simultaneous communicative functions: marking experiential topics, enacting
interpersonal relationships, and organizing text. Corpus-based discourse analysis of
patterns in a specialized hashtag corpus is used to investigate these linguistic
functions and how they enact social processes of ‘ambient affiliation’3. This corpus
consisted of posts with the hashtag, #MKR, indicating that they are about the
Australian TV series, ‘My Kitchen Rules’.
Bio
Michele Zappavigna is a lecturer in the School of Arts and Media at the University of
New South Wales. Her major research interest is the discourse of social media and she
has published two books in this area: Discourse of Twitter and Social Media
(Continuum, 2012) and, with Ruth Page, Johann Unger and David Barton,
Researching the Language of Social Media (Routledge, 2014). Her work also engages
with the practical concerns of applied linguistics. Key publications with this focus
include, Tacit Knowledge and Spoken Discourse (Bloomsbury, 2013), and Discourse
and Diversionary Justice: An Analysis of Ceremonial Redress in Youth Justice
Conferencing (Palgrave, forthcoming) with JR Martin and Paul Dwyer.
Introduction: Second-screen viewing
Posting commentary to social media while watching television has become a
pervasive practice. For example, using Twitter to post an observation about a show in
real-time is known as ‘live-tweeting’4. It typically involves the use of a mobile device,
such as a smart phone or tablet, that acts as an interactive ‘second screen’5. Access to
social media services on these devices means that viewers are increasingly networked
in terms of both the ‘synchronicity’ of media supporting their communication, and the
‘simultaneity’ of other audience members present in the ambient social network6.
Pre-proof – please so not quote this version. Published as:
Zappavigna, M. (2017). Ambient liveness: Searchable audiences and
second screens. In Hight, C & Harindranath, H. Studying Digital Media
Audiences: Perspectives from Australasia. London: Routledge. pp 150-
172.
Second screens allow these viewers to produce their own texts, and to interact with
each other, both directly, and via mass communicative practices such as hashtagging.
This chapter employs corpus-based discourse analysis to explore the communicative
practices viewers engage in when live-tweeting a particular program broadcast in
Australia, ‘My Kitchen Rules’7. In particular the chapter focuses on the kind of
opinion and sentiment construed by these viewers as they forge different kinds of
attitudinal alignment. This type of communication is at once an emergent practice and
embedded in a broader history of people commenting on media via multiple
communicative channels. In exploring this meaning-making I will introduce the
concept of ‘ambient liveness’, that is, how the temporal affordances of social media
inflect communicative practices, in particular, the expression of attitude regarding
televisual media.
The real-time discourse produced by these viewers is available to audience
researchers in ways that have not previously been seen. This discourse has also been
seen as a ‘backchannel’ form of communication 8, in other words, as supplementary to
more primary forms of experience, or more principal forms of media consumption.
Anstead and O’Loughlin term the emergence of audience backchannels as the rise of
the viewertariat, “viewers who use online publishing platforms and social tools to
interpret, publicly comment on, and debate a television broadcast while they are
watching it” 9. Studies of second-screen viewing have focused on a range of genres
such as televised election debates10 and political debates11, current affairs shows12, and
talk shows, as well as differences across genres of TV shows13. Other contexts in
which backchannelling has been documented include collaborative learning
environments14, education15, conferences 16, question-and-answer sessions17, and other
kinds of large events, such as fashion trade shows18. Backchannelling has also played
a role in elections and debates19, and public events and crises20. Some studies adopt a
cognitive focus, for example, aiming to determine how second-screen viewing impact
on factual recall21, influences users’ opinions22, and affects the neural indicators of
engagement with the primary media being consumed23. There has also been interest in
how second screens act as a technology of fandom allowing fans to commune around
a shared media text24. Two-screen viewing not only facilitates real-time audience
interaction during TV shows25, but also enable fans to engage in creative practices
such as roleplaying26 or simulating characters27.
The social connections produced via backchannel communication may form a kind of
‘intangible, clandestine community’28 or ‘conversational shadow’29, operating in
parallel to media or real-world events. This loose notion of social-media supported
‘conversation’ is in keeping with accounts of microblogging as a conversational
technology. For example microblogging has been variously described as a kind of
‘conversational exchange’30, as ‘lightweight chat’31, as ‘prompting opportunistic
conversations’32, as ‘a specific social dialect, in which individual users are clearly
singled out and engaged in a conversation’33 and as constituted by ‘dialogue acts’34.
Pre-proof – please so not quote this version. Published as:
Zappavigna, M. (2017). Ambient liveness: Searchable audiences and
second screens. In Hight, C & Harindranath, H. Studying Digital Media
Audiences: Perspectives from Australasia. London: Routledge. pp 150-
172.
However, linguistic study, of the kind undertaken in this chapter, is required to
understand the particular conversation-like practices that characterise live-tweeting
via second screens, and the particular communicative functions they enact.
Ambient liveness and ‘real-time’
Microblogging, and social media more generally as a communicative platform,
foregrounds time as an important social variable. For example, Twitter describes itself
as “a real-time information network”35 where instant engagement with the social
stream36 is critical. It is now common for feeds of ‘live-tweets’ to appear in news
broadcasts, or during programmes involving audience commentary, such as chat
shows and panels. ‘Real-time’ as a concept ascribes value to an apparent increase in
the pace at which we engage with Internet-mediated resources, and with members of
our social networks. The term is often used uncritically as synonymous to ‘clock-time’
or ‘synchronous time’, where social media users are posting about events almost ‘as
they happen’ via mobile media. The classifier ‘real’, however, also has a legitimating
function: web-based time is positioned as being the equal to the ‘offline’ time of face-
to-face social interaction. At the same time it is ‘hyper-real’, potentially affording
superior access to what is happening in our social networks in any given moment. In
addition, the discourse of real-time appears to be “a central ideological feature of
global capitalism”, involved not only in cyberculture, but in finance culture and the
experience of living in city environments37.
Agnate to the concept of real-time is the notion of ‘liveness’38, historically associated
with broadcast media, and used to classify media as being broadcast at the time that it
recorded. With the advent of social media we see new forms of liveness emerging.
Practices such as ‘citizen journalism’, whereby users post about newsworthy events
that happen to occur near them, depend on the value ascribed to liveness as a form of
eyewitness39. Interestingly, the notion also appears to inform more mundane practices,
such as the chronicle of daily routine (e.g. posting about your morning coffee at the
time that you drink it 40). Couldry41 suggests that liveness invokes a potential
connection with central social realities shared at the time they are occurring, and that
the range of media to which this concept can be applied is expanding. He proposes
two new ‘rival forms of liveness: ‘online liveness’, as a form of social co-presence
mediated by Internet technologies, and ‘group liveness’, facilitating continuous
mediation via mobile technologies.
Social media use has collapsed the distinction between these two forms of liveness,
establishing what I refer to as ‘ambient liveness’. This kind of liveness presupposes
that there are always others, co-present in the social stream, ready to engage with the
content posted via a social media service. As such, liveness may be thought of as the
ever-present possibility of the attention of multiple participants in a social network.
This chapter explores the communicative patterns that are associated with such
Pre-proof – please so not quote this version. Published as:
Zappavigna, M. (2017). Ambient liveness: Searchable audiences and
second screens. In Hight, C & Harindranath, H. Studying Digital Media
Audiences: Perspectives from Australasia. London: Routledge. pp 150-
172.
ambient liveness, in particular, how live-tweeting practices afford the opportunity for
forging different attitudinal alignments and forms of ‘ambient affiliation’42.
The chapter begins by introducing the ‘My Kitchen Rules’ (MKR) corpus used as a
case-study, and the reference corpus, HERMES2013, used as a baseline of more
generalised tweeting practices. It then details the model of evaluative language used
for discourse analysis, and applies this model to the discursive patterns observed in
the corpus, with a view to understanding the communicative practices characterising
live-tweeting.
MKR corpus
The MKR corpus explored as a case study in this paper consisted of 6508 tweets (83
793 words) posted from the morning of 18/3/15 until the afternoon of 20/3/15. The
posts were about the television show ‘My Kitchen Rules’ and the selection criteria
used to construct the corpus was to include all tweets during this period containing the
hashtag, #MKR. A limitation of such hashtag-based research is that it does not
necessarily capture all the communication surrounding the primary media, since there
will be posts that do not use the hashtag, and in particular, posts that are replies may
omit the tag. However, hashtags do, at the very least, make obtaining a snapshot of
particular discourse at a particular time, achievable. A truly exhaustive corpus would
be very difficult to create since it is possible to comment on a TV show without
mentioning its name or using an established hashtag, and thus both hashtags and
keyword searches will result in patchiness in terms of exhaustivity. In addition, since
2010, due to its increasing commercial development, Twitter has not made the
‘firehose’ of all tweets available at an affordable price for most research projects 43,
and instead many researchers rely on the ‘garden hose’ feed which is a randomised
(via an algorithm that Twitter do not disclose) selection of the more comprehensive
feed.
The posts in the MKR corpus were scraped from the Twitter Application
Programming Interface (API) using the DMI Twitter Capturing and Analysis Toolset
(DMI-TCAT)44. The corpus collection period aimed to approximate a 48 hour cycle,
and spanned two episodes of MKR that screened in Australia. As Table 1 suggests, the
number of posts across these two days was also reasonably consistent, in terms of
overall tweeting, and in terms of the amount of tweets with links, hashtags and
mentions, as well as the level of retweets and replies (Table 1). This suggests that the
live-tweeting occurring has some level of communicative consistency across different
episodes of the show. Of particular note is the relative infrequency of posts with
replies. If backchannel commentary is to be thought of as conversational, analysis that
moves beyond turn-taking as a conversational strategy is needed.
Pre-proof – please so not quote this version. Published as:
Zappavigna, M. (2017). Ambient liveness: Searchable audiences and
second screens. In Hight, C & Harindranath, H. Studying Digital Media
Audiences: Perspectives from Australasia. London: Routledge. pp 150-
172.
Date Number
of tweets
tweets
with links
tweets
with
hashtags
tweets
with
mentions
retweets replies
18/03/15 2963 245 2956 907 520 157
19/03/15 3243 258 3239 1018 603 188
20/03/15 161 84 159 82 49 7
Table 1 General composition of the MKR corpus.
Tweets were collected on 20/3/15 until 4pm, deliberately stopping collection 3 hours
prior to the screening of the show to demonstrate that minimal posting occurred
outside the time in which the TV show aired (i.e. around 7pm). The frequency of
tweeting is presented visually in Error: Reference source not found, which shows the
frequency of tweets along the y axis, and time along the x axis. This figure also gives
some indication of the number of users compared with tweets, suggesting that there
are users who post multiple times during the broadcasts.
Figure 1 Tweets in the MKR corpus over time.
The concordance software, Antconc45, was employed to explore the communicative
patterns in the corpus, with concordance lines used for investigating particular
instances. Concordance lines are samples of particular lexical or grammatical patterns
presented together with a pre-defined amount of co-text (e.g. 5 words to the left and 5
Pre-proof – please so not quote this version. Published as:
Zappavigna, M. (2017). Ambient liveness: Searchable audiences and
second screens. In Hight, C & Harindranath, H. Studying Digital Media
Audiences: Perspectives from Australasia. London: Routledge. pp 150-
172.
words to the right of the search word). For example concordance lines for the
temporal adverb ‘right now’ are shown in Error: Reference source not found. These
differ from traditional concordance lines, which offer only limited windows of co-text
compared to the length of the entire text, because the entire post can be displayed
since Twitter posts are such short texts.
!
""
#$
%!
Table 2 Concordance lines for ‘right now’ from the MKR corpus.
Evaluative language and live-tweeting
Given that live-tweeting involves responding to some form of primary media, and
affords the opportunity for expressing opinion and making observations, it is
unsurprising that evaluative language plays an important role in this form of
discourse. However, in order to understand the functions of this communication it is
necessary to employ a framework, grounded in linguistic theory, for exploring
evaluative language. The model adopted in this paper is known as the Appraisal
Framework46, developed within Systemic Functional Linguistics. This framework
details three regions of meaning in which evaluation is construed in language: AFFECT
(expressing emotion), JUDGEMENT (assessing behaviour) and APPRECIATION
(estimating value).
The choices between these different meanings are represented in Error: Reference
source not found as a system network47, that is, a network interrelated options that are
organised paradigmatically, in terms of ‘what could go instead of what’, rather than
syntagmatically in terms of structure48. The examples shown in the boxes are taken
from HERMES2013. Each example illustrates a type of ATTITUDE and the network
may be further specified to greater levels of delicacy depending on the kind of
analysis for which it is being used. ATTITUDE that is INSCRIBED is realised via explicit
Pre-proof – please so not quote this version. Published as:
Zappavigna, M. (2017). Ambient liveness: Searchable audiences and
second screens. In Hight, C & Harindranath, H. Studying Digital Media
Audiences: Perspectives from Australasia. London: Routledge. pp 150-
172.
evaluative choices, while ATTITUDE which is INVOKED is suggested via linguistic
choices that imply evaluation but do not employ evaluative lexis.
&'(
)*(
(+*,-.(//
""
$ $ 01 2 3
4$
% " ! " 51$
!1 6
do Deband Evaeverserveamainwithout banana leaf? #mkr.
00
$
Figure 2 The system of attitude49.
For example, consider the types of ATTITUDE realised in the following posts about the
contestant Drasko:
Drasko kinda annoys me.....#mkr
Drasko is a walking fiasco #MKR
Drasko and Bianca are my fav couple #mkr
In the first post Drasko is the trigger of negative emotion, with ‘annoys’ realising
negative AFFECT. In the second post he is the target of negative moral opinion, with
‘walking fiasco’ realising negative JUDGEMENT in terms of capacity. In the final post,
together with his partner, Drasko is assessed in terms of aesthetic opinion, with ‘fav’
realising positive APPRECIATION.
Comparing the MKR corpus to a reference corpus
In order to understand the communicative practices observable in the MKR corpus,
and make claims about how this live-tweeting might differ from general tweeting, it is
necessary to use a larger, randomised Twitter corpus as a baseline. This is what is
known in Corpus Linguistics as a ‘reference corpus’. The reference corpus used in
this study was HERMES2013, a 100 million word corpus of approximately 10 million
randomised tweets collected in 2013. The MKR corpus contained less @mentions
(3% of the total words) compared with the reference corpus (6.5%). This is perhaps
due to the preference, which we explore later in the chapter, of referring to contestants
in the show using vocatives rather than @mentions. Assessing the difference in
hashtagging levels is problematic, since a hashtag was used in the selection criteria for
the MKR corpus. Approximately 19% of posts in the MKR corpus were retweets, a
Pre-proof – please so not quote this version. Published as:
Zappavigna, M. (2017). Ambient liveness: Searchable audiences and
second screens. In Hight, C & Harindranath, H. Studying Digital Media
Audiences: Perspectives from Australasia. London: Routledge. pp 150-
172.
similar level to the reference corpus (25%). These kinds of structural differences are
not particularly illuminating. What is more interesting is to compare the different
kinds of meanings that are made in the corpora.
Unsurprisingly, the discourse in the MKR corpus, conformed to a more restricted field
(semantic domain) than the reference corpus. A keyword analysis was performed to
determine which words in the MKR corpus were key when compared with
HERMES2013, that is, which words occurred at a significantly different frequency to
the reference corpus (Table 3).
7 89 :;8< <:=>:=;?
@ 51
$
8>@ >==7>9>
= =9> ?=<8>77
? =7@ =<@7878
8 @>8 =:@=@7=
: =88 =7;9@@;
; 0$ =@@ @:>=:9:
> 0$
!
7>: @::@;?=
< ""
A
7;> @8?>@7:
79 7>8 7<??9<8
77 7>< 7::<:?:
7@ @=; 7:??>?;
7= 7== 78>8>?=
7? B 7@7 7??@;@7
78 $ 7;8 7=@>9=:
Table 3 Keywords in the MKR corpus using HERMES2013 as a reference corpus
Pre-proof – please so not quote this version. Published as:
Zappavigna, M. (2017). Ambient liveness: Searchable audiences and
second screens. In Hight, C & Harindranath, H. Studying Digital Media
Audiences: Perspectives from Australasia. London: Routledge. pp 150-
172.
It is also unsurprising that tweets to the official @mykitchenrules account are key.
More interesting is the general types of words that were key:
oMKR-specific hashtags e.g. #MKRpubcrawl
oContestant names e.g. Drasko
oCompetition-related lexis e.g. teams
oCooking-related lexis e.g. cooking
oEvaluative language about the above e.g ‘fantastic food’
The sections which follow later in the chapter explore the types of meaning associated
with each of these and how they afford forms of ambient affiliation. As Error:
Reference source not found suggests, discourse about the show’s contestants was most
prevalent area of semantic difference between the corpora. The discourse was
typically highly evaluative in nature, spanning the three types of attitudinal meaning
introduced in the previous section. The most frequent key attitudinal lexis in the
corpus (Table 4) was predominately APPRECIATION, typically invoking JUDGEMENT
targeted at the capacity of the contestants.
:< ! ;> ?7<=;
7
4$"
;7 ?> ?9:<=
?
.!%
1 !"2
79
:
C @8 @8@<:
7
&001#01
7=
<
?: 7;>;@
8
$$D$
E(&#F#FF--/G/)#F
E(.(&,*'H,-
78
>
=7 78;7=
:
1$
,,0
7:
:
79= 787?>
=
"
!$
51$
7<
@
1$ =? 7@>8<
8
$1"$1
Table 4 Key evaluative lexis in the MKR corpus
Pre-proof – please so not quote this version. Published as:
Zappavigna, M. (2017). Ambient liveness: Searchable audiences and
second screens. In Hight, C & Harindranath, H. Studying Digital Media
Audiences: Perspectives from Australasia. London: Routledge. pp 150-
172.
For example, consider the concordance lines for ‘raw’ in Table 5. What without
context might appear to be a purely ideational reference to an uncooked state, is in
fact charged with negative value, invoked through co-occurring exclamation of
incredulity (e.g. 2, 3, 4) and mock-delight (e.g. 1, 6). While this evaluation is realised
in the texts as APPRECIATION, it invokes JUDGEMENT of the abilities of the contestants
who have made the dishes being assessed. Rallying around such shared negative
assessment is a form of communing affiliation in which the live-tweeters create
ambient solidarity through shared critique. We will return to this form of affiliation
later in this chapter.
7 ""
)2
0
@ 2!!00
"
= !1
I
?
2
8 02
:
"
; &!!0!
> 21$!
!J
J
!
1$
Table 5 Concordance lines for ‘raw’.
Communing: The function of hashtags in live-tweeting
Social tagging, for example inserting a hashtag such as #MKR into a social media
post, plays an important role in live-tweeting. Hashtags are a form of linguistic
innovation51 that have spread from microblogging to different forms of social media
and have been integrated into other mediated contexts such as television and
advertising. Hashtags are prefixed by the # symbol which at a typographic level, acts
a particular kind of linguistic marker, indicating the beginning of a tag, and its special
status as metadiscourse. The # symbol has an interesting semiotic history, originating
Pre-proof – please so not quote this version. Published as:
Zappavigna, M. (2017). Ambient liveness: Searchable audiences and
second screens. In Hight, C & Harindranath, H. Studying Digital Media
Audiences: Perspectives from Australasia. London: Routledge. pp 150-
172.
with the pound symbol as an abbreviation derived from the Latin libra pondo (pound
in weight) and subsequently taking on a range of uses depending on context, for
instance as an ‘octothorpe’ on touch tone phones developed by Bell Labs in the
1970s52. Hashtag use may have arisen “from a history among computer programmers
of prefacing specialized words with punctuation marks, such as $ and * for variables
and pointers, or the # sign itself for identifying HTML anchor points” 53. Special
characters of this kind are typographic conventions used in markup languages to
indicate when something is part of the ‘markup’ rather than the ‘content’ being
annotated.
Zappavigna54, drawing on Halliday’s55 notion of linguistic metafunctions, identifies
three important discursive functions that hashtags can play in live-tweeting:
An interpersonal function of enacting ambient relationships and construing
ambient community.
A textual function of organizing the post (acting as a form of punctuation and
affording aggregation (intertextually associating the tweet with other tweets
containing the same tag).
An experiential function of classifying the post a being about a particular
topic.
Table 1 presents the 10 most frequent hashtags in the corpus. The most obvious
function of hashtags in the MKR corpus is an experiential function of indicating the
topic of the tweet. Indeed, this function was leveraged in creating the corpus, the
assumption being that tweets tagged with #MKR were about the show. An example of
this topic-marking function is the following, where without the hashtag providing
context, the particular target of the positive AFFECT construed in the tweet is unclear:
I need a lot of tissues tonight. Omg! I love both of the teams #MKR
The tag indicates that the teams appraised are part of the TV show. However, hashtags
can have functions beyond such topic-marking, in particular interpersonal functions
involved in how attitudinal alignments are negotiated in the discourse. For example:
@mykitchenrules cook-off! Good luck girls! #mkr #mkr2015
#soexcited
In this post the hashtag, ‘#soexcited’, is an evaluative metacomment with the
interpersonal function of construing AFFECT regarding the dimension of the TV show
referenced in the body of the post. The broader social function of this type of tag is to
invoke solidarity with a potential network of co-emoters who share the sentiment
presented in the tweet.
Pre-proof – please so not quote this version. Published as:
Zappavigna, M. (2017). Ambient liveness: Searchable audiences and
second screens. In Hight, C & Harindranath, H. Studying Digital Media
Audiences: Perspectives from Australasia. London: Routledge. pp 150-
172.
7 8<=8 "(%$
@ @978 @<@ H$"$2$!
!!%&'()*
= 0$
!
7;: /*%$+,-
? ""
A
7?9 %../
8 $
?: -,1B1
%!01*$-!
Table 6 Most frequent hashtags in the MKR corpus.
While the most frequent tags have a predominately experiential function, at the other
end of the spectrum there was a set of 419 hashtags (with a frequency of one) that
construe highly interpersonal meanings (Table 7). These tags clearly do not have
marking a particular topic as their primary function, and it is unlikely for anyone to
conduct a search using such idiosyncratic tags. Instead the main function appears to
be to enact evaluative metacommentary with a humorous edge, for example:
"It's a lot of pressure in a commercial kitchen!" No shit, Sherlock.
#MKR #CaptainObvious
The hashtag in this post construes negative JUDGEMENT as a metacomment targeted at
the quoted speaker.
Hashtag Example
#iwanttobebestfriendswith
allofyou
Seriously, #MyKitchenRules? My two favourite teams against each
other?! This is madness! #iwanttobebestfriendswithallofyou #seriously
#MKR
#katandandjefailbigtime &&4(K($"$CI
1$!01I0
%1123"
#lovelygirls unlucky Sheri n Emilie #chinup #lovelygirls #mkr
#snobberyatitsbest Wrong...a large amount of pub goers won't want to eat "raw" steak..
#stubbornwomen #MKR #snobberyatitsbest
#wankers I've been to a few pubs in life but never had Ling en Pappiote #wankers
Pre-proof – please so not quote this version. Published as:
Zappavigna, M. (2017). Ambient liveness: Searchable audiences and
second screens. In Hight, C & Harindranath, H. Studying Digital Media
Audiences: Perspectives from Australasia. London: Routledge. pp 150-
172.
#mkr
#keepingitclassy Bearded clams as finger food #keepingitclassy #mkr
Table 7 Examples of hashtags with a frequency of 1.
The hashtags in the corpus were able to make a large variety of interpersonal
meanings about the TV show, the contestants, the judges, and other viewers. For
instance, interpersonal hashtags spanned three of the four different types of speech
function defined by Halliday and Matthiessen56: statements, questions, offers and
commands. For example, the following is an instance of a hashtag realising a
statement that invokes negative AFFECT:
Is it wrong that I just bought Pete Evans saying 'The Score Is A Ten'
as a ringtone? #WeNeverGotATen #MKR
Questions realised in hashtags were often rhetorical and used to produce additional
attitudinal metacommentary:
So Devo! Neither team deserves to be eliminated! #MKR
#Whygodwhy @mykitchenrules
Commands often realised some form of JUDGEMENT, for instance the following
example targeted at other viewers in the social network:
OMG! Stop with the fat hate at Celine. Pathetic and horrible people -
improve yourself instead of criticising others #mkr #StopBullying
The only function not performed was offers of goods or services, a semantic function
more typical of promotional discourses.
Hashtags could evaluate a particular target in the tweet (e.g. a contestant), the entire
body of the tweet (often a situation presented in the show) or the entire post, including
other tags. The evaluative hashtags in the corpus were observed to realize all the
systems of ATTITUDE57 introduced earlier. For instance there were tags construing
AFFECT (expressing emotion), often the viewer’s reaction to an event in the show:
Time sure does fly in that kitchen! The stress is really starting to kick
in now #keepcalm #mkr2015 #mkr @mykitchenrules
The live-tweeters also reacted to the show with hashtags by realising JUDGEMENT
(assessing behaviour), often as a humorous display, for example:
If none of these teams doesn't do a chicken parma, they should all go
into elimination. #unaustralian #pubfood #mkr
This form of evaluation was also targeted at the discourse of other viewers:
Note to self: stop reading #MKR tweets if I don't want to despair the
state of humanity. #sexismeverywhere #bodyshamers
Pre-proof – please so not quote this version. Published as:
Zappavigna, M. (2017). Ambient liveness: Searchable audiences and
second screens. In Hight, C & Harindranath, H. Studying Digital Media
Audiences: Perspectives from Australasia. London: Routledge. pp 150-
172.
In addition APPRECIATION (estimating value) was deployed in hashtags such as the
following, which, interestingly, were accompanied by viewer’s images of their own
food:
#MKR Love both Mussels and Beef Carpaccio #delicious
@mykitchenrules http://t.co/wI7xcO9Qxo
Special dinner with friends #yummy #MKR http://t.co/l9F8Pd29gj
Integrating this type of hashtagged metadiscourse into social life may be the first
semiotic movement toward what I term ‘searchable talk’: computer-mediated
communication where we actively leverage the affordances of metadata to make our
discourse more ‘bondable’, in the sense of enabling ambient, ad hoc connections with
both putative and real mass audiences. As we have seen, while topic-marking tags
might invite alignment with an ambient community of co-searchers actively
attempting to find out what people are saying about something or within some
specific semantic domain, interpersonal tags instead invite alignment with putative
co-evaluators or co-emoters. The latter, while unlikely to search for the specific
interpersonal tag, are ambiently co-present in the social network, potentially sharing
the same feelings and values. It may be the case that as our social relations shift in
tandem with changes in the social connections made possible with emerging social
media technologies, that individuals may engage in purely interpersonal search,
actively seeking texts construing similar emotions or values #whoknows!
Enacting solidarity: Addressing and evaluating contestants in real-time
Evaluation of the contestants in the show is one of the most common communicative
patterns in the corpus. The most frequent evaluative lexis in the MKR corpus is
shown in Table 8 together with the most frequent 3 word pattern in which this
appraisal occurred. This provides a synoptic perspective on some of the most frequent
domains of evaluation. The instances of evaluation were determined using the word
frequency list and then discarding non-evaluative usage (e.g. uses of ‘like’ as a
softener).
.45
7 @<>
@ 7<; $
Pre-proof – please so not quote this version. Published as:
Zappavigna, M. (2017). Ambient liveness: Searchable audiences and
second screens. In Hight, C & Harindranath, H. Studying Digital Media
Audiences: Perspectives from Australasia. London: Routledge. pp 150-
172.
= 7?>
? ! 7=9 !
8 $ 7;8 $
Table 8 Most frequent evaluative lexis in the MKR corpus.
As this table suggests, the most frequent appraisal in the corpus is positive in polarity.
For instance the most frequent evaluation was ‘love’ (positive AFFECT) most often
used in the 3-gram, ‘love both teams’. This phrase co-occurred with additional
patterns of AFFECT (shown underlined in Error: Reference source not found),
construing the strong emotional reaction the viewers present themselves as having to
the fate of the contestants. This outpouring of sentiment also has the function of
forging alignments with other viewers who share the same positive feeling.
$2!,ILI
!!"0!1!
12,
51$G!$4$
0%!II6,
,"$!
51$
F!1,MN=
51$
,$"(!
11$!/(
/($11(
,(01O$%&
51$.,
51$!6
,
Table 9 Concordance lines for 'love both teams'.
Negative assessment of characters was also frequent, for instance the pattern
‘contestant + is so’ (Error: Reference source not found, ATTITUDE shown in bold)
Pre-proof – please so not quote this version. Published as:
Zappavigna, M. (2017). Ambient liveness: Searchable audiences and
second screens. In Hight, C & Harindranath, H. Studying Digital Media
Audiences: Perspectives from Australasia. London: Routledge. pp 150-
172.
P$
H*$-!
$!$"
0
Table 10 Concordance lines for ‘contestant + is so’.
Another prominent communicative practice was addressing comments to participants
in the TV show. This was most often the contestants, but could also be the judges, or
the MKR twitter account. With the exception of this account, the forms of address
were not typically @mentions, presumably because the twitter accounts of the
contestants may have not been widely known or promoted. The live-tweeters often
addressed the participants in the show as if they were potential conversational
interlocutors. This occurred by using the character’s name as a vocative (term of
address):
Hey, Drasko, you really passionate about becoming a chef? Start
peeling potatoes, bitch! #MKR
As this example suggests, the discourse directed at the contestants was predominately
interpersonally charged. For instance this is apparent in concordance lines for ‘you’
(Table 11, vocatives shown underlined; ATTITUDE shown bold underlined).
F 7
$
1
/$1I)$0$
1$8."2
*I
,-.!"11$
$!1$#
!
&-..1
9#9
0$!
3&2!1
1
Table 11 Concordance lines for 'you + vocative'.
Commentary of this kind could spawn conversational exchanges with other viewers.
An example is the exchange in Table 12, beginning with User 1 addressing a
contestant with the vocative, ‘Colin’. This conversation centers on shared negative
Pre-proof – please so not quote this version. Published as:
Zappavigna, M. (2017). Ambient liveness: Searchable audiences and
second screens. In Hight, C & Harindranath, H. Studying Digital Media
Audiences: Perspectives from Australasia. London: Routledge. pp 150-
172.
APPRECIATION (shown bold underlined) of this contestant’s hair, with each
accumulating instance of evaluation keeping the conversation flowing and solidifying
the attitudinal alignment enacted. This kind of discourse has a communing function,
whereby the coupling of Colin’s hair with negative APPRECIATION becomes a value
that is shared by User 1 and 2, realizing a shared social bond within the ambient
network.
:);,-.!"11$$!1$
#1$!
:';5'7!,
:);5'@!1$!""!01
"(&,,)!$1
:';5'7!!!2
!!?
,
:);5'@
",,!$
Table 12 An exchange negotiating aesthetic appreciation.
Another pattern involving viewers directly addressing contestants is ‘hey + vocative’
which is also highly evaluatively charged (Table 13, ATTITUDE underlined). The
instances of negative JUDGEMENT inscribed in these posts might be thought of as
reminiscent of futilely shouting abuse at the TV. The affordances of social media
mean, however, that there is the chance that the contestants might view the unfolding
feed of comments and react to the commentary. In addition, the abuse is performed to
the ambiently co-present audience communicating via the MKR hashtag. Thus this
discourse becomes a kind of rallying affiliation, aligning viewers around shared
negative evaluation, and perhaps forging a sense of shared superiority amongst the
viewers.
-<1(*0)$
P3/B@978
80&21$10$"I
/002
+!$"/'((,,
'*,1
(((1$"0
!01$!
=24$1$1$$0$"
1$
Pre-proof – please so not quote this version. Published as:
Zappavigna, M. (2017). Ambient liveness: Searchable audiences and
second screens. In Hight, C & Harindranath, H. Studying Digital Media
Audiences: Perspectives from Australasia. London: Routledge. pp 150-
172.
H$Q1$2I
$0
Table 13 Concordance lines for ‘hey + vocative’.
Other patterns of this kind included ‘dear + vocative’:
Dear Kat, every time you cook it's a complete disaster. Please go
home. #mkr
and ‘command + vocative’:
Use a peeler Josh! #MKR #MKRPubCrawl
Again the main function enacted in these examples appears to be both expressing
personal assessments about the contestants, at the same time as aligning with the
ambient audience. In many instances they are accompanied by underspecified
outbursts of ATTITUDE in the form of exclamations such as ‘oh my god’ or ‘WTF’:
'0!$"";
>
>.0""
1$F
"IJ0$1IJ
20$$!"
/(!I
.&I!00I)H&H(/
Table 14 Examples of exclamations in the corpus.
All of the patterns of positive and negative evaluation considered in this section are
examples of ‘ambient affiliation’58, whereby viewers are aligning around couplings of
ideational meaning (e.g. contestants) and shared ATTITUDE. This affiliation is ambient
in the sense that the viewers do not necessarily have to engage in direct conversational
exchanges with each other to enact this sense of shared values, instead they merely
need to occupy the same semantic space within the social network, coordinated, for
example, by hashtags and other kinds of metadata (e.g. location).
Pre-proof – please so not quote this version. Published as:
Zappavigna, M. (2017). Ambient liveness: Searchable audiences and
second screens. In Hight, C & Harindranath, H. Studying Digital Media
Audiences: Perspectives from Australasia. London: Routledge. pp 150-
172.
Metadiscourse about live-tweeting
While most posts were about the show, there was a set of self-reflexive posts about
the practice of live-tweeting itself, typically realising an attitudinal stance. This
discourse included posts that refer to the activity of live-tweeting and to tweets as
communicative events. Four main semantic functions were identified (Table 15,
ATTITUDE shown in bold). For example these included posts expressing desire, via
AFFECT, about having one’s tweet broadcast on the screen during the show, in other
words, a desire for visibility. Parallel to this was pleasure at being recognised in this
way, again involving inscribed positive AFFECT and invoked AFFECT via exclamations.
In addition there were posts assigning positive social esteem via JUDGEMENT to the
viewer whose tweets was shown on screen. At the same time there was negative
JUDGEMENT of the propriety of other viewers’ tweeting, often featuring hashtags
commanding them to change their behaviour (e.g. ‘#benice’), as well as
APPRECIATION regarding humorous tweets, and positive assessment of Twitter as a
backchannel in general. This metadiscourse, aside from having the function of
enacting the kind of solidarity already discussed in relation to attitudinally charged
posts, foregrounds live-tweeting as a communicative space in which different kinds of
relationships with other viewers can be enacted.
"1 4$1!
*$
N0
$"
F4$!1!!1
5'751$!1
!1III
*1$
1
5')$!!4$
7
&"
!
P$"!
0000
1!
!#,1
/
&"! .?.!!$
!
Pre-proof – please so not quote this version. Published as:
Zappavigna, M. (2017). Ambient liveness: Searchable audiences and
second screens. In Hight, C & Harindranath, H. Studying Digital Media
Audiences: Perspectives from Australasia. London: Routledge. pp 150-
172.
/4$!?& ?
& ?
!!!
Table 15 Functions of metadiscourse about tweeting.
Such ATTITUDE typically occurred in tweets without replies, but there were instances
where the evaluation was negotiated in conversation-like exchanges, often involving
humour. For example, the exchange shown in Table 16 involves negotiation of
ATTITUDE regarding the visibility afforded by a tweet appearing on the TV screen
(ATTITUDE in bold). The appreciation realised in the exchange is targeted at the
verbiage of the tweets (e.g. ‘PG rated’, ‘tamest, lamest’, ‘sardonic’) coupled with
self-deprecating humour, construed via hyperbole (e.g. ‘downfall’). The tenor of these
posts is playful, and the ATTITUDE fuels the unfolding of the exchange in a similar
manner to the kinds of saturated evaluation seen in casual conversation. Even posts
without replies, such as the observations shown in Table 15, retain a sense of potential
conversationality in relation to the ever present ambient audience who might reply,
favourite or retweet the post.
User 1: @mykitchenrules Thanks for airing my tweet(s), that's very PUBlic :) #mkr
User2: @User1 @mykitchenrules I think none of mine are PG rated to be aired lol
User1: @User2 @mykitchenrules Well, you might have a point there
User3: @User1 @mykitchenrules How come my tweets don't make it to the telly???
User 4: @User1 @mykitchenrules Three years I've been waiting....even the tamest,
lamest ones don't work. Sigh.....
User 5: @User1 @mykitchenrules You're a star!
User 1: @User5 @mykitchenrules thanks very munch
User 1: @User4 @mykitchenrules Oh my, and you have such a nice profile pic as well,
very classy
User 4: @User1 @mykitchenrules allegedly. Sardonic sarcasm might be my downfall
User 1: @User4 @mykitchenrules perhaps, not that there's anything wrong with that
Table 16 A conversational exchange involving metadiscourse about live-tweeting.
Conclusion
Pre-proof – please so not quote this version. Published as:
Zappavigna, M. (2017). Ambient liveness: Searchable audiences and
second screens. In Hight, C & Harindranath, H. Studying Digital Media
Audiences: Perspectives from Australasia. London: Routledge. pp 150-
172.
Social media discourse offers an important resource for audience researchers, seeking
to understand new forms of hybrid media use. This chapter has explored the
communicative practices involved in live-tweeting television via a second screen,
applying a corpus-based discourse analysis method. This method drew upon the
Appraisal framework, a linguistic model of evaluative language. The value to
audience research of using an analytical approach, grounded in linguistics, is that it
offers a systematic way of understanding audience practices that is based on textual
evidence about patterns in communication. This type of evidence, as we have seen,
can be both quantitative, telling us something about the communicative tendencies of
audiences, and qualitative, providing details about the particular stances adopted by
audience members through inspection of concordance lines.
An important dimension of the discourse analysed in this chapter was the use of
hashtags to coordinate ideational content and forge attitudinal alignments. Several
communicative practices were evident: communing around shared values via
interpersonal hashtags, enacting solidarity with other viewers through addressing and
evaluating contestants, and construing metadiscourse about live-tweeting as a
communal activity. The MKR audience studied actively leveraged the affordances of
hashtags to make their discourse more ‘bondable’ within the social stream. Social
tagging, and social media discourse more generally, will undoubtedly become an
increasingly important data source for audience studies. The challenge will be to
apply systematic and theoretically sound models of language in order to understand
the meanings audiences are making with these new semiotic resources.
Notes
1 Huang, Jeff, Katherine M Hornton, and Efthimis N Efthimiadis, "Conversational Tagging in Twitter," In Proceedings
of the 21st ACM conference on Hypertext and hypermedia, 173-78, Toronto, Ontario, Canada: ACM, 2010.
2 Halliday, Michael Alexander Kirkwood, and Christian Matthias Ingemar Martin Matthiessen, An Introduction to
Functional Grammar, 3rd ed (London: Arnold, 2004).
3 Zappavigna, Michele, "Ambient Affiliation: A Linguistic Perspective on Twitter,” New Media & Society 13, no. 5
(August 1, 2011 2011): 788-806
4 Schirra, Steven, Huan Sun, and Frank Bentley, "Together Alone: Motivations for Live-Tweeting a Television Series,"
Paper presented at the Proceedings of the 32nd annual ACM conference on Human factors in computing systems,
Toronto, Ontario, Canada, 2014.
5 Lochrie, Mark, and Paul Coulton, "Mobile Phones as Second Screen for Tv, Enabling Inter-Audience Interaction," In
Proceedings of the 8th International Conference on Advances in Computer Entertainment Technology, 1-2. Lisbon,
Portugal: ACM, 2011.
6 Laursen, Ditte, and Kjetil Sandvik, "Talking with Tv Shows: Simultaneous Conversations between Users and
Producers in the Second-Screen Television Production Voice," Northern Lights: Film & Media Studies Yearbook 12, no.
1 (2014): 141-60.
7 ‘’My Kitchen Rules’ is an Australian cooking TV game show in which contestants’ cooking skills are judged by a
panel of celebrity chefs and hosts. The initial format of the show was two teams competing against each other to cook a
three-course meal, with variations on this format in later seasons. The show has been running since 2010.
8 Atkinson, Cliff, The Backchannel : How Audiences Are Using Twitter and Social Media and Changing Presentations
Forever (Berkeley, California: New Riders, 2010), Sarita, Yardi. "The Role of the Backchannel in Collaborative
Learning Environments,” Paper presented at the Proceedings of the 7th international conference on Learning sciences,
Bloomington, Indiana, 2006.
9 Anstead, Nick, and Ben O’Loughlin, "The Emerging Viewertariat and Bbc Question Time: Television Debate and
Real-Time Commenting Online,” The International Journal of Press/Politics 16, no. 4 (2011): 440-62.
10 Kalsnes, Bente, Arne H. Krumsvik, and Tanja Storsul, "Social Media as a Political Backchannel," Aslib Journal of
Information Management 66, no. 3 (2014): 313-28.
11 Anstead, Nick, and Ben O’Loughlin, "The Emerging Viewertariat and Bbc Question Time: Television Debate and
Real-Time Commenting Online,”, Pedersen, Sarah, Graeme Baxter, Simon Burnett, Ayse Göker, David Corney, and
Carlos Martin, "Backchannel Chat: Peaks and Troughs in a Twitter Response to Three Televised Debates During the
2014 Scottish Independence Referendum Campaign," Paper presented at the CeDEM15: Conference for E-Democracy
and Open Government, 2015.
12 D'Heer, Evelien, and Pieter Verdegem, "What Social Media Data Mean for Audience Studies: A Multidimensional
Investigation of Twitter Use During a Current Affairs Tv Programme," Information, Communication & Society 18, no. 2
(2015/02/01 2014): 221-34.
13 Buschow, Christopher, Beate Schneider, and Simon Ueberheide, "Tweeting Television: Exploring Communication
Activities on Twitter While Watching Tv," Communications - The European Journal of Communication Research 39,
no. 2 (2014): 129-149.
14 Yardi, Sarita, "The Role of the Backchannel in Collaborative Learning Environments,” In Proceedings of the 7th
international conference on Learning sciences, 852-58. Bloomington, Indiana: International Society of the Learning
Sciences, 2006.
15 Ebner, Martin, Conrad Lienhardt, Matthias Rohs, and Iris Meyer, "Microblogs in Higher Education - a Chance to
Facilitate Informal and Process-Oriented Learning?," Computers and Education 55, no. 1 (2010): 92-100.
16 McCarthy, Joseph F, and dana boyd, "Digital Backchannels in Shared Physical Spaces: Experiences at an Academic
Conference," Paper presented at the Conference on Human Factors and Computing Systems (CHI 2005), Portland,
Oregon, USA, April 2-7, 2005 2005, Reinhardt, W, M Ebner, G Beham, and C Costa, "How People Are Using Twitter
During Conferences," Paper presented at the Creativity and Innovation Competencies on the Web, Proceedings of 5,
EduMedia conference, Salzburg, 2009, Grosseck, Gabriela, and Carmen Holotescu, "Microblogging Multimedia-Based
Teaching Methods Best Practices with Cirip.Eu.," Procedia - Social and Behavioral Sciences 2, no. 2 (2010): 2151-55.
17 Harry, Drew, Joshua Green, and Judith Donath, "Backchan.nl: Integrating Backchannels in Physical Space," In
Proceedings of the 27th international conference on Human factors in computing systems, 1361-70, Boston, MA, USA:
ACM, 2009.
18 Bisker, Solomon, Hector Ouilhet, Steve Pomeroy, Agnes Chang, and Federico Casalegno, "Re-Thinking Fashion
Trade Shows: Creating Conversations through Mobile Tagging," In CHI '08 extended abstracts on Human factors in
computing systems (Florence, Italy: ACM, 2008), 3351-56.
19 Hawthorne, Joshua, J. Brian Houston, and Mitchell S. Mckinney, "Live-Tweeting a Presidential Primary Debate:
Exploring New Political Conversations," Social Science Computer Review, 31, no. 5 (2013): 552-62.
20 Vieweg, Sarah, Amanda L Hughes, Kate Starbird, and Leysia Palen, "Microblogging During Two Natural Hazards
Events: What Twitter May Contribute to Situational Awareness,” Paper presented at the Proceedings of the SIGCHI
Conference on Human Factors in Computing Systems, 2010.
21 Van Cauwenberge, Anna, Gabi Schaap, and Rob van Roy, "“Tv No Longer Commands Our Full Attention”: Effects
of Second-Screen Viewing and Task Relevance on Cognitive Load and Learning from News,” Computers in Human
Behavior 38, no. 0 (9// 2014): 100-09.
22 Cameron, Jaclyn, and Nick Geidner, "Something Old, Something New, Something Borrowed from Something Blue:
Experiments on Dual Viewing Tv and Twitter," Journal of Broadcasting & Electronic Media 58, no. 3 (2014): 400-19.
23 Pynta, Peter, Shaun AS Seixas, Geoffrey E Nield, James Hier, and Emelia Millward, "The Power of Social
Television: Can Social Media Build Viewer Engagement? A New Approach to Brain Imaging of Viewer Immersion,"
Journal of Advertising Research 54, no. 1 (2014): 71-80.
24 Highfield, Tim, Stephen Harrington, and Axel Bruns, "Twitter as a Technology for Audiencing and Fandom: The#
Eurovision Phenomenon," Information, Communication & Society 16, no. 3 (2013): 315-39.
25 Lochrie, Mark, and Paul Coulton, "Mobile Phones as Second Screen for Tv, Enabling Inter-Audience Interaction,"
In Proceedings of the 8th International Conference on Advances in Computer Entertainment Technology , 1-2. Lisbon,
Portugal: ACM, 2011.
26 Magee, Rachel M., Melinda Sebastian, Alison Novak, Christopher M. Mascaro, Alan Black, and Sean P. Goggins,
"#Twitterplay: A Case Study of Fan Roleplaying Online," In Proceedings of the 2013 conference on Computer
supported cooperative work companion, 199-202, San Antonio, Texas, USA: ACM, 2013.
27 Bore, Inger-Lise Kalviknes, and Jonathan Hickman, "Continuing the West Wing in 140 Characters or Less:
Improvised Simulation on Twitter," The Journal of Fandom Studies 1, no. 2 (2013): 219-38.
28 Yardi, Sarita, "The Role of the Backchannel in Collaborative Learning Environments,” In Proceedings of the 7th
international conference on Learning sciences, 852-58. Bloomington, Indiana: International Society of the Learning
Sciences, 2006.
29 Shamma, David A, Lyndon Kennedy, and Elizabeth F Churchill, "Conversational Shadows: Describing Live Media
Events Using Short Messages,” Paper presented at the Proceedings of the Fourth International AAAI Conference on
Weblogs and Social Media, Washington DC, 2010.
30 Honeycutt, Courtney, and Susan Herring, "Beyond Microblogging: Conversation and Collaboration in Twitter," In
Proceedings of the Forty-Second Hawai’i International Conference on System Sciences, no page numbers available,
Los Alamitos, CA: IEEE Press, 2009.
31 Starbird, Kate, Leysia Palen, Amanda L Hughes, and Sarah Vieweg, "Chatter on the Red: What Hazards Threat
Reveals About the Social Life of Microblogged Information,” Paper presented at the Proceedings of the 2010 ACM
conference on Computer supported cooperative work, 2010.
32 Zhao, Dejin, and Mary Beth Rosson. "How and Why People Twitter: The Role That Micro-Blogging Plays in
Informal Communication at Work,” Paper presented at the Proceedings of the ACM 2009 international conference on
Supporting group work, 2009.
33 Grosseck, Gabriela, and Carmen Holotescu, "Microblogging Multimedia-Based Teaching Methods Best Practices
with Cirip.Eu.," Procedia - Social and Behavioral Sciences 2, no. 2 (2010): 2151-55.
34 Ritter, Alan, Colin Cherry, and Bill Dolan, "Unsupervised Modeling of Twitter Conversations," Paper presented at
the Human Language Technologies: The 2010 Annual Conference of the North American Chapter of the Association for
Computational Linguistics, 2010.
35 Twitter. "Twitter About Page,” https://twitter.com/about.
36 Posts to services such as Twitter, Weibo and Facebook will have timestamp information appended to them indicating
when the post was published. While revealing location data is usually optional, this temporal data is usually
automatically generated since it is central to delivering chronologically unfolding ‘streams’ of content to the ambient
audience.
37 Hope, Wayne, "Global Capitalism and the Critique of Real Time," Time & Society 15, no. 2-3 (2006): 275-302.
38 Couldry, Nick, "Liveness,“Reality,” and the Mediated Habitus from Television to the Mobile Phone," The
Communication Review 7, no. 4 (2004): 353-61.
39 There is debate about the value of citizen journalism which has been criticized for invoking “the drama of
instantaneity, which is compelling and engaging for readers, but not necessarily compatible with fact checking
processes of western paradigms of journalism” Papacharissi, Zizi and Maria de fatima oliveira, “Affective news and
networked publics: The rhythms of news storytelling on# Egypt”, Journal of Communication, 62 (2012): 266-282,
p.279.
40 Zappavigna, Michele, "Coffeetweets: Bonding around the Bean on Twitter,” In The Language of Social Media:
Communication and Community on the Internet, edited by P Seargeant and C Tagg (London: Palgrave, 2014), 139 -60.
41 Couldry, Nick, "Liveness,“Reality,” and the Mediated Habitus from Television to the Mobile Phone," The
Communication Review 7, no. 4 (2004): 353-61.
42 Zappavigna, Michele, "Ambient Affiliation: A Linguistic Perspective on Twitter,” New Media & Society 13, no. 5
(2011): 788-806
43 Bruns, Axel, and Jean Burgess, "Researching News Discussion on Twitter: New Methodologies," Journalism
Studies 13, no. 5-6 (2012): 801-14.
44 Borra, Erik, and Bernhard Rieder, "Programmed Method: Developing a Toolset for Capturing and Analyzing
Tweets," Aslib Journal of Information Management 66, no. 3 (2014): 262-78.
45 Antconc (Version 3.4.3) [Computer Software]. Waseda University. Available from http://www.laurenceanthony.net/,
Tokyo, Japan.
46 Martin, J.R, and P.R.R White, The Language of Evaluation: Appraisal in English (New York: Palgrave Macmillan,
2005).
47 System networks are an alternative to modelling language as a catalogue of structures. This kind of systemic
orientation to meaning arose out of the Firthian tradition in linguistics which asserted the need for a distinction between
structure and system, that is, between syntagmatic and paradigmatic relations in language (Firth, 1957).
48 Halliday, Michael Alexander Kirkwood, and Christian Matthias Ingemar Martin Matthiessen, An Introduction to
Functional Grammar, 3rd ed (London: Arnold, 2004).
49 Based on Martin and White (2005).
50 Since #MKR was part of the selection criteria it can be ignored.
51 Cunha, Evandro, Gabriel Magno, Giovanni Comarela, Virgilio Almeida, Marcos André Gonçalves, and Fabrício
Benevenuto, "Analyzing the Dynamic Evolution of Hashtags on Twitter: A Language-Based Approach," Paper
presented at the Proceedings of the Workshop on Language in Social Media (LSM 2011), 2011.
52 Houston, Keith, Shady Characters: The Secret Life of Punctuation, Symbols, and Other Typographical Marks (New
York ; London: WW Norton & Company, 2013).
53 boyd, D, S Golder, and G Lotan, "Tweet, Tweet, Retweet: Conversational Aspects of Retweeting on Twitter," Paper
presented at the Proceedings of 43rd Hawaii international conference on system sciences (HICSS), 5-8 January 2010,
Honolulu, HI, United States of America, 2010.
54 Zappavigna, Michele, "Searchable Talk: The Linguistic Functions of Hashtags,” Social Semiotics 25, no. 3: 274-
291.
55 Halliday, Michael Alexander Kirkwood, Language as Social Semiotic: The Social Interpretation of Language and
Meaning (Baltimore: University Park Press, 1978).
56 Halliday, Michael Alexander Kirkwood, and Christian Matthias Ingemar Martin Matthiessen, An Introduction to
Functional Grammar.
57 Small caps are used here to differentiate the technical use of these terms from their everyday counterparts.
58 Zappavigna, Michele, "Coffeetweets: Bonding around the Bean on Twitter,” In The Language of Social Media:
Communication and Community on the Internet, edited by P Seargeant and C Tagg, 139 -60. London: Palgrave, 2014,
Zappavigna, Michele, "Ambient Affiliation in Microblogging: Bonding around the Quotidian,” Media International
Australia no. 151 (2014): 97-103, Zappavigna, Michele, "Ambient Affiliation: A Linguistic Perspective on Twitter,”
New Media & Society 13, no. 5 (2011): 788-806, Zappavigna, Michele, Discourse of Twitter and Social Media.
Continuum Discourse Series (London: Continuum, 2012), Zappavigna, Michele, "Enacting Identity in Microblogging,”
Discourse and Communication 8, no. 2 (2014): 209-28.