Content uploaded by Morten Bay
Author content
All content in this area was uploaded by Morten Bay on Nov 14, 2024
Content may be subject to copyright.
7/13/2019 Bay
https://firstmonday.org/ojs/index.php/fm/rt/printerFriendly/9388/7603 1/14
First Monday, Volume 23, Number 11 - 5 November 2018
Political discourse on social media is seen by many as polarized, vitriolic and permeated by falsehoods and
misinformation. Political operators have exploited all of these aspects of the discourse for strategic purposes,
most famously during the Russian social media influence campaign during the 2016 presidential election in
the United States and current, similar efforts targeting the U.S. elections in 2018 and 2020. The results of the
social media study presented in this paper presents evidence that political influence through manipulation of
social media discussions is no longer exclusive to political debate but can now also be found in pop culture.
Specifically, this study examines a collection of tweets relating to a much-publicized fan dispute over the Star
Wars franchise film Episode VIII: The Last Jedi. This study finds evidence of deliberate, organized political
influence measures disguised as fan arguments. The likely objective of these measures is increasing media
coverage of the fandom conflict, thereby adding to and further propagating a narrative of widespread discord
and dysfunction in American society. Persuading voters of this narrative remains a strategic goal for the U.S.
alt-right movement, as well as the Russian Federation. The results of this study show that among those who
address The Last Jedi director Rian Johnson directly on Twitter to express their dissatisfaction, more than half
are bots, trolls/sock puppets or political activists using the debate to propagate political messages supporting
extreme right-wing causes and the discrimination of gender, race or sexuality. A number of these users
appear to be Russian trolls. The paper concludes that while it is only a minority of Twitter accounts that tweet
negatively about The Last Jedi, organized attempts at politicizing the pop culture discourse on social media
for strategic purposes are significant enough that users should be made aware of these measures, so they
can act accordingly.
Contents
Introduction
Background
Related work
Methods
Findings
Discussion
Conclusion
Introduction
Eight months after it opened in theaters, Star Wars fans were still talking about the eighth installment in the
series, The Last Jedi. During this time, media outlets ranging from lightweight pop culture websites to serious
news organizations have covered the “toxic” parts of Star Wars fandom, i.e., fans who hate The Last Jedi and
have gone as far as trying to crowdfund a remake of the film, start Change.org petitions to strike the film
from the Star Wars canon and create videos, Web sites and social media content that criticize the film and
call for the firing of its creators. Supporters of The Last Jedi have called these detractors out as being
predominantly white males with misogynistic views that did not care for the film’s attempts at improving
representation of women and ethnic/sexual minorities in the Star Wars franchise. However, as the study
presented here shows, this is more than a heated discussion among social media users. There is also
evidence that the fan conflict caused by The Last Jedi stems from deliberate and organized social media
influence tactics employed by politically motivated operators, foreign and domestic. This study explores how
these political influence tactics on social media have jumped from political debate spaces to pop culture
discussions — but with the same goals of disruption or persuasion.
7/13/2019 Bay
https://firstmonday.org/ojs/index.php/fm/rt/printerFriendly/9388/7603 2/14
Background
In the National Review, conservative commentator Peter Spiliakos described the conflict as having less to do
with the movie itself and more to do with the political polarization of the Western societies into which The
Last Jedi was inserted: “People on both sides of this divide are trying to drag the Star Wars franchise into a
pre-existing set of obsessions and resentments.” (Spiliakos, 2018). Whether you agree with Spilliakos’ take
on the film or not, this is an intriguing perspective. How does the current state of political discourse and the
use of social media for political influence tactics in the U.S. and other Western nations affect our consumption
of pop culture phenomena such as The Last Jedi?
The Star Wars franchise is an interesting object of study in this regard. Even though it was originally targeted
towards cinema-goers who were too young to vote, Star Wars was always double-coded, with layers oriented
towards adults, often involving subjective critiques of contemporary politics. Star Wars movies, books, video
games, TV shows and comics have consistently attempted to convey left-leaning values. For years, the
franchise’s creator, George Lucas, has explained the series’ impetus as partially being a comment on the
Vietnam war, which ended just as Lucas started work on the first Star Wars film. The evil, oppressive,
technologically and economically superior Empire represented the United States, while the far less advanced,
but resilient South Vietnamese forces and their unlikely victory inspired the “good guys” in the Rebel Alliance
(Teague Beckwith, 2017). Political commentary is even more present in the so-called “Prequel trilogy” from
1999–2005, in which the hyper-capitalist Trade Federation, led by Nute Gunray (named after Republican Newt
Gingrich) becomes part of a Separatist Alliance wishing to split the republic. In this parallel to the American
civil war, the Separatist Alliance represents the Confederate States of America. The conflict turns out to be
the work of Senator Sheev Palpatine, who uses his provisional war privileges as Chancellor of the Republic to
consolidate and centralize power at the executive level, similar to the changes in the national security
apparatus that happened under then-President George W. Bush, such as the passing of the PATRIOT Act and
the establishment of the U.S. Department of Homeland Security. The references to the Bush administration
became even clearer, as Anakin Skywalker, about to become Darth Vader, paraphrases Bush’s words about
Western nations and Islamist terrorists: “If you are not with me, you are my enemy.” (O’Connor, 2016).
It should not have been a surprise, then, that a new trilogy in the Star Wars franchise would express equally
left-leaning sentiments. Although they may still have a long way to go (Brown, 2018), the Star Wars films,
books, video games and TV shows produced after Disney’s acquisition of Lucasfilm in 2014 have made an
effort to address identity politics by introducing strong, female protagonists and a better overall
representation of gender, race/ethnicity and sexuality. This was the case in the first entry in the new trilogy,
The Force Awakens, but even more so in The Last Jedi (Watercutter, 2017), which took a no-holds-barred
approach to address issues of gender discrimination, class warfare, the destructive character of masculine
aggression and war profiteering, while still working within the left-of-center frame constructed by George
Lucas in 1977. Criticism of American engagements in the Middle East had already been present in the
anthology film Rogue One from 2016 (Doescher, 2016), so clearly Star Wars was continuing to convey left-of-
center values in the new Disney era. In other words, in the more than 40 years it has existed, politics and
left-leaning political commentary has always been woven into Star Wars’ fabric.
Still, it appears that some fans with right-leaning political views expected the franchise to be politically
neutral, as they went to see the first Star Wars film of the Trump presidency, The Last Jedi. They saw its
arguments for equality of gender, race and class as a new, leftist takeover of Star Wars, even though Star
Wars has always been politically left-leaning. The Last Jedi is unique in that it landed in the Trump era, acting
as a lightning rod at a time when most cinemagoers had chosen a political side for or against the president
and adopted the “obsessions and resentments” of their political camp, with social media acting as the primary
battleground. However, The Last Jedi fan conflict is not just an interesting case because it is a microcosm of
the overall political discourse on social media in the Trump era, but also because it is possible to identify
organized and deliberate attempts at right-wing political persuasion and/or defense of conservative values as
well as sexism, racism and homophobia in the social media discussions about the film. It is important to
stress, of course, that there are also a substantial number of fans who simply think The Last Jedi is a bad film
and who use social media to express their disappointment. Regardless of motive, almost all negative fans
express the belief that they are in the majority and that most Star Wars fans dislike The Last Jedi.
In this paper, I analyze tweets sent to the director of The Last Jedi over the first seven months after its
release. In the collected data, I have discovered political agitation for right-wing values using The Last Jedi as
a placeholder for left-wing positions. Furthermore, it appears that political activists have used bots and sock
puppet accounts to troll left-wing fans, and there is even evidence that Russian influence operators have
inserted themselves into the debate to exploit and exacerbate the conflict, thereby securing more media
attention to the conflict, which again helps spread the perception that America is divided and in chaos.
Thus, this analysis of tweets pertaining to The Last Jedi shows that pop culture spaces on social media are
now also political battlegrounds, vulnerable to the same organized vitriolic polarization, manipulation and
disinformation seen in the usual venues for political discourse online.
Related work
Fandom studies, and studies of fan interactions online are now in their third decade, after the groundbreaking
early work of scholars such as Camille Bacon-Smith (1992), Nancy Baym (1993), and Henry Jenkins (1992).
7/13/2019 Bay
https://firstmonday.org/ojs/index.php/fm/rt/printerFriendly/9388/7603 3/14
From the beginning, the internal conflicts among fans and in the producer/fan relationship have been part of
the equation, but since the emergence of social media and their many affordances for community-building,
expression and discussion, studies of these conflicts have become more frequent among scholars. Authors
like Stanfill (2018, 2010) and Brock (2015) have brought attention to the heteronormativity and racial biases
that live in many fandoms (and also in fandom studies itself). Jonathan Gray turned his attention to those
who are not fans, or have explicit opinions about, e.g., a TV show and might even become ‘haters’, a group
he referred to as “anti-fans” (Gray, 2003). These are different from “fantagonists”, which is a term introduced
by Derek Johnson (2007) for fans who engage antagonistically with other fans or with content producers, for
example by belonging to certain factions of the fandom who are in opposition to other factions with different
views of the content. In the book chapter introducing the concept, Johnson references Brooker’s (2002)
important work on fan reactions to the Star Wars prequel trilogy and the so-called Special Edition versions of
the three original films. The visceral and vitriolic fan reactions against these installments in the Star Wars
series were so prevalent in the pre-social media world that a documentary, The People vs George Lucas, was
produced on the subject.
In recent years, inspired in part by the #GamerGate controversy as well as other controversies that followed
it, scholars have been studying what is now termed “toxic” fandom. The widespread view among academics is
that some toxic fan practices are symptoms of endemic and omnipresent misogyny in geek culture,
entertainment fandoms and in society as a whole (Massanari and Chess, 2018; Massanari, 2017; Salter and
Blodgett, 2017; Todd, 2015; Bealer, 2011). William Proctor is one of a few scholars trying to moderate this
view. Though he does not deny or downplay the presence of sexist, homophobic or racist fan practices, he
argues that some practices that often described as “toxic” in popular media outlets can be “benign” and
“innocuous”. In his view, these defensive stances are indications of a “totemic nostalgia” which ties the fan’s
past to the present, (Proctor, 2018, 2017) and severing this connection by redefining or changing the
meaning of “totems” from the fan’s past is the psychological equivalent of invalidating that part of the fan’s
personality.
These scholars, and many more, have uncovered and unpacked fandom divides along fault lines of gender,
race and sexuality in much more impressive ways than I could ever aspire to. Hence, I will mostly rely on the
work mentioned above as accepted knowledge, focusing on how these divisions are being exploited for
political and agitation purposes. Literature on the latter will be introduced below.
Methods
Collecting tweets
To ensure that the dataset was not restricted by adherence to certain hashtags or so large that it would
exceed my computational resources to analyze it, I chose to study tweets addressing The Last Jedi director
Rian Johnson directly. Using Twitter’s Advanced Search functionality, I retrieved 18,113 tweets tweeted
directly to the director’s Twitter handle, “@rianjohnson” between 13 December 2017 and 20 July 2018. After
removing retweets and Likes that appeared as individual tweets in the dataset, the number of tweets were
brought down to 1,273. The Last Jedi opened in Europe on 13 December, so this is the earliest that
substantial amounts of fan reactions could be expected. Deadline restrictions meant that the tweet collection
had to be performed on 20 July, but the frequency of tweets sent directly to Rian Johnson about The Last Jedi
had also slowed down substantially at this point — even though the debate was still somewhat alive among
fans elsewhere. The tweets were scraped from Twitter using the Data Miner software. The dataset was
subsequently cleaned, so that errors, duplicates and tweets from Rian Johnson himself were removed. GIFs
and meme images that were unaccompanied by text were removed from the dataset in order to retain text
searchability.
Collection practice
No mass collection of tweets has been performed, since only tweets engaging with Rian Johnson were
collected. Only information stated publicly on the studied Twitter accounts was collected. All information
revealed in this paper was immediately available and observable on public Twitter accounts during the study
period. Only handles of subsequently deleted or suspended accounts are revealed in this paper. The collected
data has been stored on protected computing devices.
Sentiment analysis and coding
To ascertain how the collected tweets were attitudinally constituted, a manual sentiment analysis was
performed, which is described below. As one of the main objectives was to study the political leanings of Star
Wars fans engaging with Rian Johnson (rather than, say, how frequently those fans tweet), series of several
tweets expressing the same sentiment from the same account was reduced to one. This reduced the number
of tweets under analysis to 967 and simultaneously created the foundation for the account analysis to follow,
since each user was now represented by one tweet.
Benefits of this method have been shown in several studies based on analyses of tweets (Borromeo and
Toyama, 2015; Chikersal, et al., 2015; Barclay, et al., 2014) and include higher accuracy than many current
automated systems, the ability to detect satire or sarcasm and a lower noise rate. The disadvantage is that
manual sentiment analyses are too labor-intensive to perform on large datasets, but with 967 tweets, the
dataset in the present study was manageable enough for manual coding and analysis.
7/13/2019 Bay
https://firstmonday.org/ojs/index.php/fm/rt/printerFriendly/9388/7603 4/14
As can be seen in the examples below, the language used in the dataset contained very little nuance.
Because of the unequivocally positive, negative or neutral nature of the collected tweets, forcing a rating
upon them for coding purposes could entail overinterpretation, raising the risk of bias significantly. I believe
the advantages of inference-driven generalizability and the assessment of sample quality would be undercut
by this high risk of bias, and since the size of the dataset is significant, yet still manageable, I chose instead
to adhere to a simple categorization of tweets as negative, positive and neutral.
An example of a negative tweet: “@rianjohnson your movie is the worst. I hate you for ruining Star
Wars.#kathleenkennedy #RianJohnson #TheLastJedi #starwars #thelastjediawful” Other negative tweets
included phrases such as “You ruined Star Wars”, “Please don’t make anymore Star Wars movies” and tweets
calling the film a “travesty”, “dumb”, “terrible”, “career-endingly bad”, “abhorrent” and “awful”. Other users
stated that Rian Johnson had committed a “war crime” or should “be in jail”. Only four tweets out of the 967
were more gradual in their negative sentiment, with one user admitting to being “in the camp that did not
enjoy it though I did appreciate the cinematography” and another calling a Johnson a “great filmmaker”, but
also stating that “I couldn’t disagree with your take on Luke more”. One of the four tweets came from a user
that adopted a slightly less negative stance towards the film months later. The four less unequivocal tweets
mentioned above were coded as negative.
On the positive side, tweets were similarly unequivocal. This is an example of a positive tweet: “@rianjohnson
TLJ is a true piece of art! The symbolism and the chemistry between Rey and Kylo is just amazing. Thank you
for such a great film.”. Other positive tweets used words such as “great”, “beautiful”, “fantastic” and
“wonderful”.
Finally, a number of tweets can be said to be value-neutral in that they engage in discourse on the film, other
Star Wars productions or general entertainment industry topics without passing judgment on The Last Jedi.
An example includes this tweet about another of Rian Johnson’s films: “@rianjohnson Having binged Jean-
Claude Van Johnson this weekend, I keep wondering if you’ve seen it, if so, what you thought of all the
Looper vs. Timecop arguments and lastly, how many people have already asked you about this?”. It seems
reasonable to assume that those who actively chose to engage with Rian Johnson directly on Twitter in the
seven months after the release of The Last Jedi without expressing negative sentiments towards the film are
also unlikely to actually harbor such negative sentiments, at least to a significant extent.
It is also important to note that the abovementioned removal of GIFs and meme images did not do a
disservice to the coding of negative tweets, as even a quick overview of Rian Johnson’s Twitter account from
the study period makes it clear that positive or neutral memes/GIFs far outnumber negative ones. Suspected
use of sarcasm or irony was accounted for by consulting the context of the tweet.
Account analysis
With the dataset now consisting of one tweet per account, it was possible to turn towards an analysis of the
accounts of those users who had tweeted negatively. This analysis was performed to ascertain whether the
users behind the negatively-tweeting accounts were simply sufficiently dissatisfied with the film that they felt
the need to express this directly to the director, or whether some — and how many — had a different,
politically-motivated agenda driving their negativity. The analysis was also performed to identify any outside
influence from foreign actors and how many Twitter bots participated in the discussion.
For the account analysis, I divided the negatively-tweeting accounts into three categories, coding them as
Political agenda, Troll/Sock Puppets/bots and Real fantagonists. This was done by visiting each of the
negative twitter accounts and reading through tweets sent during the study period. In cases where the
account was highly active and contained large amounts of tweets or retweets, I would conduct searches for
specific terms on the accounts to find tweets that could help place the account in one of three categories. The
specific criteria for each category are specified below.
1. Political agenda
These are accounts that are not solely Star Wars-related but most likely belong to real human beings, rather
than bots or sock puppeteers (see below), and from which several/frequent tweets have been sent that can
be characterized as overtly political. The tweets on these accounts would typically only occasionally or even
rarely be about Star Wars, but also consisted of comments on current affairs or personal matters — the latter
being one of the indicators that the owner of the account is not using the account as a sock puppet or
exclusively for trolling activity. For example, these accounts would also often include selfies or family photos
containing objects or persons that were consistent over time.
For accounts with high tweeting/retweeting activity, search terms such as “Trump” and “SJW” (see below)
were employed to discover tweets that might indicate a political stance, e.g., for or against the current U.S.
president.
2. Troll/Sock Puppet/Bots
Several of the accounts could be identified as fully or semi-automated bot accounts, accounts specifically
used for trolling, the latter often involving accounts with a false identity, also known as sock puppets. See
below for definitions of these terms and how they were detected/identified in the dataset.
3. Real fantagonists
These accounts are the most likely to belong to the human beings claiming to be behind them, who are Star
Wars fans who were simply disappointed by The Last Jedi for reasons other than political ones and expressed
7/13/2019 Bay
https://firstmonday.org/ojs/index.php/fm/rt/printerFriendly/9388/7603 5/14
their discontent to Rian Johnson on Twitter. These are fans who tweet very frequently about Star Wars, and
who also often did so before the study period. Some of these accounts would contain the occasional
politically-oriented tweet or retweet, but not enough to indicate an agenda or position like the accounts in the
Political Agenda category.
Identifying sock puppets, trolls and bots
Bots are software robots that post content to Twitter with little or no human control involved, mostly for
strategic purposes (Fitzgerald and Shaffer, 2017), and though bots are widely used for strategic
communication in commercial contexts, they are increasingly used in political contexts as well, as witnessed
in the U.S. presidential election in 2016 (Gorwa, 2017; Bessi and Ferrara, 2016). For bot detection, I used the
Botometer tool created by Indiana University Network Science Institute (IUNI) and the Center for Complex
Networks and Systems Research (OSoMe [Observatory on Social Media], 2018), which relies on very large
sets of social media data and machine learning to ascertain the likelihood of a Twitter account being operated,
fully or partially, by a bot.
Sock puppets are fake accounts where the user profile represents a user other than the person actually
controlling it. Kumar, et al. (2017) show how sock puppet account owners use their accounts to post content
they would perhaps do not want to include on accounts disclosing their real identities. They may also operate
several such accounts. According to Kumar, et al., sock puppets post more frequently than regular users and
participate in more, mostly controversial discussions than regular users (but rarely initiate them). They are
often treated harshly by the rest of the community and use more swearwords.
A troll is a human actor that “teases people to make them angry, or somebody who offends people, or
somebody who wants to dominate any single discussion, or somebody who tries to manipulate people’s
opinion (sometimes for money)” [1]. Most often, trolls use sock puppet accounts or fake identities to avoid
repercussions, but a small number of trolls are willing to be identified by using their actual identities on their
profiles.
Mutlu, et al. extracted a range of troll characteristics from a dataset of 95,578 tweets posted by 3,321 users.
Trolls may send more than 50 tweets a day, often exceed a 70 percent retweet rate (70 retweets per 100
tweets sent), have follower/following ratios of 0.4 and below and often don’t change the visual presentation
of their profile very much from the default settings. Posts from trolls/sock puppets contain simplified
language with shorter sentences and words with fewer syllables. Finally, sock puppets/trolls and bots often
use handles that are “variations on a single ‘real’ name, variations on a celebrity name, and long strings of
alphanumeric garbage (often after a ‘real’ name)” (Fitzgerald and Shaffer, 2017).
Russian trolls
In addition to the characteristics mentioned above, certain parameters indicate that a troll may part of a
Russian social media influence campaign. Several studies have shown how trolls have been employed by the
Russian Federation in order to sway public opinion in the West through social media and online forums
(Kaminska, et al., 2017; Zelenkauskaite and Niezgoda, 2017; Giles, 2016; Lange-Ionatamishvili and Svetoka,
2015; Nimmo, 2015) The U.S. intelligence community as well as social media platforms have acknowledged
that such a campaign was ongoing during the study period, targeting the U.S. midterm election in November
2018 (Diamond, 2018; Gleicher, 2018). From studies of the Russian influence campaign of 2016, researchers
have constructed comprehensive lists of criteria for detection of Russian trolls. Analyzing 27,000 tweets from
1,000 Twitter users submitted to the U.S. Congress as part of a list of 2,752 accounts identified by Twitter as
having ties to the Russian “troll farm” Internet Research Agency, Zannettou, et al. found that Russian trolls
usually target “very specific world events” and “political threads”, i.e., they insert themselves in, and often
solely focus on, discussions that take up space in American media discourse. They also find that Russian trolls
“... adopt different identities over time, i.e., they ‘reset’ their profile by deleting their previous tweets and
changing their profile name/information” [2] and prefer the Twitter Web client over the mobile client. They
predominantly report their location (if at all) to be in the U.S., Germany and Russia, mostly tweet in the
languages of these nations, and can be observed randomly switching from one to the other. They also often
pretend to be sources of information or news outlets and “nudge” other users to follow them. Their accounts
are often created just before or just after events relating to topics that the troll is focused on. Llewellyn, et
al., analyzing the same 2,752 alleged Russian troll accounts, found that Russian troll activity peaked at
certain points in time. Zannettou, et al. also noticed this, finding that Russian troll tweets had two peaks on a
typical day during the 2016 presidential election, where their activity was higher than the established
baseline for Twitter. Tweets from Russian trolls exceeded the base line between 7am and 10pm and then
again — more substantially — between 12 noon and 5pm UTC. Russian trolls also like to retweet other trolls,
contact prominent users through mentions and they frequently attempt to organize “political events and
abusive behaviour and harassment” [3].
In the present study, the Twitter accounts were categorized after how many of these criteria they met, and
those that met a large majority of them would only be placed in the Troll/Bot/Sock Puppet or Russian Troll
categories after an extra, qualitative inspection. It is important to mention here that it is widely considered
almost impossible to attribute any sort of cyberattack or strategic operation in the online sphere to an actor
with absolute certainty, due to the many opportunities for anonymization, masking of IP addresses, adoption
of false identities online, etc. (Lobel, 2012; Dipert, 2010; Geers, 2010). This also applies to the trolls and
bots found in this study. Though the above described classification has been performed with rigor and the
accounts in Troll/Bot/Sock Puppet category fulfill most of the criteria listed, these accounts should be viewed
as having a high likelihood of being trolls/sock puppets, bots and Russian trolls. No absolute attributions are
claimed.
7/13/2019 Bay
https://firstmonday.org/ojs/index.php/fm/rt/printerFriendly/9388/7603 6/14
Findings
Negative-to-positive/neutral fan ratio
Among the 967 tweets analyzed, 206 expressed a negative sentiment towards the film and its director, which
is 21.9 percent or a little more than one in five fans. This number includes all negative tweets analyzed, i.e.,
also those who came from the 44 accounts identified as bots, sock puppet accounts and trolls. It also includes
61 users who showed clear political agendas in their tweets against the film. Thus, the number of fans whose
tweets are purely motivated by a negative stance towards the film is 101 or 10,5 percent.
Overall, 50.9 percent of those tweeting negatively was likely politically motivated or not even human.
Some bots — A lot of trolls
Using the Botometer mentioned in the Methods section, 11 out of the of 206 accounts expressing negative
sentiments were identified as bots.
I identified 33 of the 206 negative accounts as trolls and/or sock puppets. Besides meeting a majority of the
detection criteria mentioned in the Methods section, these accounts would mostly tweet or retweet right-wing
messages alongside their attacks on Rian Johnson and The Last Jedi. All of these troll/sock puppet accounts
were created — or became active after being dormant for months or years — during the study period, most
of them around the time when The Last Jedi opened in theaters and the first negative fan reactions were
posted to Twitter. This may indicate that the accounts were created or revived specifically for trolling behavior
related to The Last Jedi or to use that debate as a platform for activities related to other subjects.
Seven of the 16 had auto-generated handles consisting of a very common, English name followed by a series
of seemingly random digits, and five of those seven had not uploaded a profile image. According to the
studies mentioned above, this combination is a typical characteristic of Russian troll accounts.
Sixteen of these 33 troll/sock puppet accounts appear to be Russian trolls, or at least possess several of the
Russian troll characteristics noted earlier. Seven of the 16 had auto-generated handles consisting of a very
common English name followed by a series of seemingly random digits, and five of those seven had not
uploaded a profile image. According to studies mentioned earlier, this combination is characteristic of Russian
troll accounts. One such account tweeted at Rian Johnson, with only little lingual variation from tweet to
tweet, that he “ruined Star Wars” no less than 13 times during a three-week period, along with other tweets
disparaging the director. In addition to these tweets, the vast majority of other tweets on the account were
retweeted messages from Donald Trump and black conservative activist and commentator Candace Owens.
Eleven of the 16 accounts suspected of being run by Russian trolls almost exclusively tweeted about The Last
Jedi, Rian Johnson or right-wing U.S. politics, typically retweeting personalities from the right or alt-right.
Eight of the 16 had their primary Twitter activity within the timeframe established by Zannettou, et al. The
coordinated efforts typical of Russian troll activity were also seen in a somewhat offensive tweet originally
posted by the account @1popculturefan, which has since been deactivated during one of Twitter’s purges of
Russian-controlled accounts post-study period. The same tweet, along with several other double postings,
was also tweeted by a still-active account using the deliberately misspelled name of a well-known Hollywood
actor (another indicator of Russian troll activity), with the same “1” in front of it. @1popculturefan is not the
only account that has been deleted or suspended during troll and bot purges by Twitter after the data
collection took place. The same applies to these accounts that met many of the Russian troll criteria during
the study period: @MarcoSo94862885, @VPalmera and @ThatNikkaGeeked.
Six of the 16 accounts have an extremely high retweet rate. Nine accounts have been through the “resets”
that are highly characteristic of Russian troll accounts, in which the account name (not the Twitter handle) is
changed, tweets from before a certain time are deleted, or they go dormant for long periods of time, but then
become active when certain events occur that other Russian trolls also comment on. Four accounts present
themselves as a type of news source and encourage other users to follow them.
An example of an account that almost sums up these Russian troll characteristics carries an auto-generated
handle and has almost exclusively tweeted disparagingly about The Last Jedi, and engaged in “anti-SJW”
rhetoric. In the middle of these tweets, the account all of a sudden tweeted in support of Donald Trump
during the latter’s visit to the United Kingdom, which was met with large protests in London. Besides not
containing any personal information, having no profile picture and other Russian troll characteristics, this
account only posted nine tweets in more than a year from its creation in January 2017 to February 2018, all
of them regarding a particular Anime series on YouTube. From April 2018 onwards, during the lead-up to the
release of The Last Jedi for the home video market, and in anticipation of Solo: A Star Wars Story opening,
the account suddenly came to life and began tweeting frequently about The Last Jedi, identity politics and
Rian Johnson. After the data was collected, the account has been through a “reset” where many of its public
tweets have been deleted, leaving only its replies.
One of the above-mentioned accounts which has now been deleted by Twitter exhibited some of the lingual
characteristics that researchers have observed in Russian troll tweets. On 11 July, @MarcoSo94862885
tweeted: “So, now explain why Mark Hamil didn’t like Luje in TLJ?”. The simplicity and brevity of the
language, as well as the fact that this purported Star Wars fan doesn’t spell “Hamill“ or “Luke” correctly, are
all indicators that raises suspicions. In combination with other characteristics (auto-generated handle, lack of
personal matter on account etc.), the verbiage suggests that the user may be a Russian troll. The literature
7/13/2019 Bay
https://firstmonday.org/ojs/index.php/fm/rt/printerFriendly/9388/7603 7/14
suggests that Russian trolls prefer Web clients over mobile apps for Twitter posts, and thus auto-correct and
word suggestions on smartphones would not disguise poor English skills.
No similar accounts were found on the positive/neutral side, where most accounts signaled that they
belonged to an identifiable human being that had a high likelihood of being real (i.e., the accounts contained
personal photos etc.) There were also no bots found among the accounts posting positive/neutral tweets.
Political agenda
Sixty-one accounts could be characterized as having a political agenda. Unlike troll/sock puppet accounts or
bots, these accounts were mostly characterized by the same human identifiability mentioned in the Methods
section.
A majority of the accounts in this category would tweet frequently and positively about party-based politics,
e.g., about President Trump and his administration or retweet supporters of the president or the president’s
own tweets. In example of more issue-based political activism, the user behind one of the analyzed Twitter
accounts replied to a post from Rian Johnson in which a quote from The Last Jedi was used on a sign
protesting the National Rifle Association. The user tweeted directly to the director: “Whoever was over quality
control at Lucasfilm was obviously having ‘naptime’ when you presented the script for The Last Jedi. After the
crap I witnessed in that movie it figures you’d be a gun grabber.”
Some were less focused on politics of party or specific issues and more on identity politics, posting anti-
feminist or anti-homosexuality messages, and tweets of a racist nature were also frequent among the
accounts in this category. In one example, a user tweeted to comedic actor Seth Rogen after he reacted
negatively to the ‘Remake The Last Jedi’ project: “You know how I know you’re gay? You don’t want to
remake last Jedi”. A majority of the accounts in the Political Agenda category tweeted antagonistically about
“SJW” — Social Justice Warriors. Often, this would be in reference to a purported SJW “agenda” injected into
the Star Wars universe by Rian Johnson, Disney and Lucasfilm by way of CEO Kathleen Kennedy, and into
American society by liberals and left-wing activists. The SJW term emerged from the #GamerGate
controversy (Chess and Shaw, 2016), and has since entered dictionaries as a pejorative term for people who
express or promote progressive views (Ohlheiser, 2015). But, as Massanari and Chess as well as Brock point
out, the SJW term is rarely used about people who promote progressive views about e.g., the national
economy or the environment. The “social justice” in the SJW term is primarily used to refer (pejoratively) to
those who support and promote equality of gender, race and sexuality. Brock describes the perception of an
SJW by those who use the term thus: “an SJW is ... overly concerned with online reputation and obsessed
with being politically correct. Coincidentally, the SJW’s activities in this definition revolve around perceived
injustices to women and people of color” [4]. Examples include:
“@rianjohnson The more I read your posts on twitter the more the terrible
direction TLJ went in makes sense to me. You SJWs ruin everything”
“... politics is certainly a part of the reason I don’t like TLJ; Rose Tico’s
deeply moronic social justice lectures for instance. The SJWs within
Lucasfilm have made this a proxy battleground”
“... dont be fooled #SJWs dont want equality they only truly want what fits
their agenda. Look no further than TLJ. Every single male character was
either a coward, an idiot, or evil”
“The reason [Kathleen Kennedy] did not get fired after #BoycottSolo is
because Disney cant fire employees with an SJW agenda because then the
media will demonize Disney as sexist and racist. Disney/Hollywood will not
let that happen.”
Several males in this category also express anger over what they perceive to be feminist agitation in The Last
Jedi:
“... if you can’t see the TLJ was basically ‘the woman’s march in space’ than
I have to believe you also have an agenda.”
“I will never watch anything Star Wars that you or Kathleen ‘The Force is
Female’ Kennedy is associated with ever again”
“Dont let this distract you from the fact that Rian Johnson killed General
Ackbar offscreen to make room for his feminist subplot ... #Fact”
In this regard, it is also worth pointing out how the negative and positive/neutral sentiments divide across
gender identification. Of the accounts whose owners identify clearly as female, 108 tweeted positively about
The Last Jedi, whereas only five tweeted negatively. The remaining accounts either belong to owners who
identify clearly as male or whose owner does not declare any gender identity.
Discussion
7/13/2019 Bay
https://firstmonday.org/ojs/index.php/fm/rt/printerFriendly/9388/7603 8/14
During the study period, in June 2018, one of the lead actors in The Last Jedi, Kelly Marie Tran, stopped using
social media due to what she later called “online harassment” from fans who held negative views of The Last
Jedi (Tran, 2018; Chuba, 2018). This led several commentators to call out the negativity towards the film as
being driven by sexism and racism (Menta, 2018; Zimmerman, 2018; Mendelson, 2018b; Holland, 2018),
something that Proctor (2018) also addresses as mentioned earlier. Based on the findings in this study, it is
not fair to generalize and paint all of the The Last Jedi detractors as alt-right activists, racists or misogynists.
However, the findings above show that a majority of the negatively-poised users included in the study do
express such sentiments, either in The Last Jedi-related tweets or in other tweets on their accounts. These
identity-based political values combine with traditional party politics and issue-based politics to represent a
politicization of Star Wars critique which is found in more than half of the negative accounts in this study.
Whether the criticism comes from a Russian troll/bot or from a fan who feels increasingly distant from the
values presented in the new Star Wars films, the objective of their actions is a political one. Russian trolls
weaponize Star Wars criticism as an instrument of information warfare with the purpose of pushing for
political change, while it is weaponized by right-wing fans to forward a conservative agenda and for some it is
a pushback against what they perceive as a feminist/social justice onslaught.
Because of the limitations on the dataset and the less-than-comprehensive nature of this study,
generalization and extension of these results to the entire Star Wars fandom should happen with extreme
caution. Not all disappointed fans are Twitter users and not all disappointed fans go as far as tweeting directly
at Rian Johnson in anger. The same can be said about fans who view the film positively, of course, which is
why this study looks at a specific discourse situation as a measure of the situation. It is nonetheless
noteworthy that a majority of the negativity stems from politicized accounts which are often part of an
organized attempt to disrupt and sow discord using the The Last Jedi controversy.
Cognitive dissonance as manipulation tactic
With respect to those individuals who express right-wing political views on their Twitter accounts and who
dislike The Last Jedi, I argue that the findings show indications of classic cognitive dissonance leading to
attitude change, as described by Festinger (1962). There is a close similarity between being a “fan”, or
supporter, of a political figure and participating in pop culture fandom. Sandvoss (2013) and Wilson (2011)
show how the same psychological mechanisms are in place for both these types of engagement, and that
there is a strong link between supporting a particular, political viewpoint and the expectation of it to be
respected, represented or even amplified in pop culture fandom. When these expectations are not met,
cognitive dissonance can occur, and I argue that for many of the fans in the political agenda category, this is
in fact the case. The perceived politics of The Last Jedi are in conflict with some fans’ perception of Star Wars
as family entertainment devoid of politics, which was never the case. The realization that Star Wars takes up
a position that is left-of-center, seems to have created much dissonance in the minds of some fans of with a
right-of-center political orientation. But rather than confront the well-documented fact that Star Wars has
always leaned left, they instead change their beliefs, convincing themselves that in fact, Star Wars was never
left-of-center or feminist until Rian Johnson made The Last Jedi and Kathleen Kennedy was put in charge of
Lucasfilm, post-Disney acquisition. In a significant number of the tweets analyzed, these fans blame Johnson,
Kennedy and Disney, sometimes collectively, sometimes as individuals, for this supposed leftist/feminist turn
in the Star Wars franchise.
Disinformation and Russian influence operations
These instances of cognitive dissonance further raise the likelihood of Russian influence operators being
present. Several scholars (Rider and Peters, 2018; Fuchs, 2017; Howard, et al., 2017; Kaminska, et al.,
2017) have shown how Russian influence operators exploit precisely this type of cognitive dissonance to
persuade individual social media users that their values are under attack, cultivating and advancing
polarization and disparity. This breakdown of American’s sense of community was and continues to be the
purpose of the Russian social media influence campaign targeting the West (Giles, 2016; Nissen, 2015).
Getting the media to pay attention to such conflicts only amplifies them, further advancing the Russian
objective.
Cognitive dissonance also enables the spread of misinformation, which works to the advantage of the Russian
actors, but is independent of such foreign influence campaigns. As shown by Doty (2013) and Kata (2012) in
the case of the anti-vaccination debate, those participating in online debates, regardless of platform, will
distort facts and spread them as misinformation to support their own argument. This often happens through
the establishment of cognitive authority, such as citing first-hand knowledge, uncritically referencing dubious
source material or assuming causality where none has been established (Doty, 2013). As mentioned in the
introduction, Fantagonists have claimed that a majority of Star Wars fans dislike The Last Jedi. They base the
claim on “evidence” such as The Last Jedi not performing as well as The Force Awakens at the box office and
later on Blu-ray/DVD, the user-generated score on the Rotten Tomatoes Web site and the fact that Solo: A
Star Wars Story which opened about five months after The Last Jedi flopped at the box office, as this was
supposedly because of a fan boycott. All of these claims have been easily and thoroughly debunked or shown
as having no evidentiary basis (Mendelson, 2018a; Robles, 2018; Nash Information Services, LLC, 2018;
Lovett, 2017). Yet, fantagonists still propagate this misinformation, which is then amplified by bots and sock
puppet accounts for political purposes, just as it did during the Russian influence campaign during the 2016
U.S. presidential election.
Star Wars fandom is not the only cultural or pop culture sphere in which these influence operations are
conducted. In February 2018, NBC News published a dataset of more than 200,000 tweets collected from
accounts that Twitter deleted after ascertaining that they were Russian troll accounts and part of the influence
operation (Popken, 2018). The dataset clearly shows that the Russian trolls were not just spreading messages
7/13/2019 Bay
https://firstmonday.org/ojs/index.php/fm/rt/printerFriendly/9388/7603 9/14
of a political nature but were using engagements in pop culture and related fandoms as either an alibi to
distract from their other trolling activities, or to insert themselves in online social groups that they may not
reach through pure political messaging. By doing the latter, the trolls were able to advance their
disinformation practices and create division within the fandoms they infiltrated. Examples from the
NBC/Twitter dataset includes a number of Russian trolls that played along with the then-popular Comedy
Central TV show @Midnight’s ‘Hashtag Wars’, engaging with the online geek culture community
@Blackgirlnerds, debating European soccer teams such as Manchester United and F.C. Barcelona and yes,
also engaging with the Star Wars fan community, even then.
The infectious nature of trolling
Finally, the findings show indications of a much simpler explanation of the behavior of some negative-
sentiment fans, especially those who are not in the Political Agenda category or can be categorized as trolls,
bots or sock puppets. Cheng, et al., in one of the of the most rigorous studies of its kind, state that “trolling
is better explained as situational (i.e., as a result of the user’s environment) than as innate (i.e., an inherent
trait)” [5]. In other words, the authors show how anyone can engage in trolling behavior, even if only a few
users or user accounts can be categorized as trolls. In the present case, this explains the similarity of
negative remarks and tweets about The Last Jedi, even though the accounts and users posting them are
different in nature. Even those who do not have political agendas or are sock puppets, troll accounts or bots,
but simply disappointed fans, engage in troll-like behavior, and this was certainly the case in the tweets
collected for this study.
The main finding by Cheng, et al. is that mood and context are the main proponents of trolling behavior. The
authors show how participating in a thread which already contains troll-like comments increases the
individual’s likelihood of engaging in trolling behavior. They also provide quantitative evidence of the rather
intuitive notion that the poster’s mood, as determined by the emotional state and physical environment of the
person behind the screen, impacts the likelihood of trolling, with a negative mood leading to more troll-like
behavior. The authors also observe how anger can spill over from one online discussion to an unrelated one
because of the accumulative nature of anger over comments read online, or feelings of anger stemming from
completely unrelated incidents, such as losing a quiz on a Web site. Simply put, an isolated comment or
tweet filled with negativity or anger towards a specific subject is likely to represent emotions that are also
derived from other activities that are unrelated to the topic in the tweet or comment. Cheng, et al. also show
how most trolling behavior displayed by non-trolls occurs late in the day and early in the work week, when
energy levels and mood may be less conducive to respectful discourse.
These suggested catalysts of trolling behavior are invisible to anyone who casually enters a discussion on
Twitter, perhaps having been led there by Twitter’s recommendation and sorting algorithms. These factors will
likely not play into any decision to react upon viewing negative tweets, but rather, following Cheng, et al., the
user is more likely to join in the trolling behavior if it is already ongoing (and is predisposed, in the present
case, from being disappointed by The Last Jedi). The findings clearly show a clustering of negative tweets
around specific times, which would indicate that several instances of the mechanism described by Cheng, et
al. was in effect during the study period. This could, however, also be impacted by the potential influence by
Russian trolls, as these also share this propensity for activity peaks. Whether the peaks of negative tweeting
activity in specific moments are mostly dominated by one or the other will require further studies of the data
and represents an opportunity for further research.
Conclusion
Keeping in mind that the collected dataset of Twitter interactions with The Last Jedidirector Rian Johnson is
limited in its representation of Twitter Star Wars fandom, there are a number of statements that can be made
on the basis of the collected data.
First and foremost, the data does not support claims that a majority of fans are so dissatisfied with The Last
Jedi that they wish to boycott further Star Wars releases under Disney ownership. Whether you consider the
21.9 percent (including, political activists, bots and trolls/sock puppets), or the 10.5 percent (excluding them)
tweets expressing a negative attitude towards The Last Jedi, it is clear that a majority is satisfied, more than
satisfied or non-committal in their attitudes. It is also shown that a majority of the negative fans with clear
gender identifications identify as male, with only a miniscule fraction of negative fans identifying as female.
Approximately one in three negative fans express misogynist, anti-progressive, anti-social justice or
conservative views. When some detractors of The Last Jedi correctly claim that it is an injustice to place these
labels on all negative fans, these detractors also have to contend with the fact that the labels actually fit a
large portion of their faction.
A number of fans feel like Star Wars has been politicized by Lucasfilm and Disney, but since the political and
ethical positions presented in the new films are consistent with older films, it is more likely that the
polarization of the Trump era has politicized the fans. The divisive political discourse of the study period and
the months leading up to it, has likely primed these fans with a particular type of political messaging that is in
direct conflict with the values presented in The Last Jedi.
The presence of deliberate influence measures, i.e., bots, sock puppet and troll accounts, is further
indications of attempts to manipulate Star Wars fans as part of a political persuasion tactic. This similarity to
political influence campaigns on social media — domestic or foreign — is also underscored by the manner in
7/13/2019 Bay
https://firstmonday.org/ojs/index.php/fm/rt/printerFriendly/9388/7603 10/14
which misinformation appears and (sometimes strategically) gets propagated. The same misinformation
mechanisms as seen in the anti-vaccination controversy and the 2016 presidential election in the U.S. are
present in the debate over The Last Jedi. The three latter points are likely the most important contribution of
this small study.
However, the assertions made in this article must be considered within the limited scope of the dataset, which
may or may not limit generalizability of the findings. Another problem for replicability is the fact that Twitter
is a dynamic forum and only tweets from selected accounts are archived outside the platform’s own servers.
This means that data collected during the study period in this paper may not correspond to later searches
because users may have deleted tweets or taken down their accounts — a general problem with research
based on Twitter data.
Yet, even considering the limitations of the dataset, there are enough indications that pop culture debates on
social media are being politicized, sometimes for strategic purposes that have nothing to do with the subject
under debate. As the debate on misinformation, political communication and regulation of social media
continues, researchers studying these matters may find it beneficial to turn their attention to pop culture and
how political messaging is propagated in its fandoms.
About the author
Morten Bay, Ph.D., is a research fellow at the Center for the Digital Future at the USC Annenberg School for
Communication and Journalism.
Direct comments to: mortenbay [at] ucla [dot] edu
Acknowledgements
A majority of this paper was completed while the author was a Ph.D. candidate at the Department of
Information Studies at University of California, Los Angeles.
Notes
1. Mihaylov and Nakov, 2016, p. 1.
2. Zannettou, et al., 2018, p. 1.
3. Llewellyn, et al., 2018, p. 1.
4. Brock, 2015, p. 1.
5. Cheng, et al., 2017, p. 11.
References
Camille Bacon-Smith, 1992. Enterprising women: Television fandom and the creation of popular myth.
Philadelphia: University of Pennsylvania Press.
Francis P. Barclay, Pichandy Chinnasamy and Priyadarshni Pichandy, 2014. “Political opinion expressed in
social media and election outcomes — US presidential elections 2012,” Journal on Media & Communications,
volume 1, number 2, pp. 15–22, and at http://dl6.globalstf.org/index.php/JMC/article/view/1653, accessed
12 October 2018.
Nancy K. Baym, 1993. “Interpreting soap operas and creating community: Inside a computer–mediated fan
culture,” Journal of Folklore Research, volume 30, numbers 2–3, pp. 143–176.
Tracy L. Bealer, 2011. “Of monsters and men: Toxic masculinity and the twenty-first century vampire in the
Twilight Saga,” In: Giselle Liza Anatol (editor). Bringing light to Twilight: Perspectives on a pop culture
phenomenon. New York: Palgrave Macmillan, pp. 139–152.
doi: https://doi.org/10.1057/9780230119246_11, accessed 12 October 2018.
Alessandro Bessi and Emilio Ferrara, 2016. “Social bots distort the 2016 U.S. Presidential election online
discussion,” First Monday, volume 21, number 11, at http://firstmonday.org/article/view/7090/5653,
accessed 12 October 2018.
doi: https://doi.org/10.5210/fm.v21i11.7090, accessed 12 October 2018.
Ria Mae Borromeo and Motomichi Toyama, 2015. “Automatic vs. crowdsourced sentiment analysis,” IDEAS
’15: Proceedings of the Nineteenth International Database Engineering & Applications Symposium, pp. 90–95.
doi: https://doi.org/10.1145/2790755.2790761, accessed 12 October 2018.
André Brock, 2015. “Whiteness as digital imaginary: SJW as boundary object,” Selected Papers of Internet
Research, at https://spir.aoir.org/index.php/spir/article/view/1043, accessed 12 October 2018.
7/13/2019 Bay
https://firstmonday.org/ojs/index.php/fm/rt/printerFriendly/9388/7603 11/14
Will Brooker, 2002. Using the force: Creativity, community and Star Wars fans. New York: Continuum.
Jeffrey A. Brown, 2018. “#wheresRey: Feminism, protest, and merchandising sexism in Star Wars: The Force
Awakens,” Feminist Media Studies, volume 18, number 3, pp. 335–348.
doi: http://doi.org/10.1080/14680777.2017.1313291, accessed 12 October 2018.
Justin Cheng, Michael Bernstein, Cristian Danescu-Niculescu-Mizil and Jure Leskovec, 2017. “Anyone can
become a troll: Causes of trolling behavior in online discussions,” CSCW ’17: Proceedings of the 2017 ACM
Conference on Computer Supported Cooperative Work and Social Computing, pp. 1,217–1,230.
doi: http://doi.org/10.1145/2998181.2998213, accessed 12 October 2018.
Shira Chess and Adrienne Shaw, 2016. “We are all fishes now: DiGRA, feminism, and GamerGate,”
Transactions of the Digital Games Research Association, volume 2, number 2, pp. 21–30, and at
http://todigra.org/index.php/todigra/article/view/39, accessed 12 October 2018.
doi: https://doi.org/10.26503/todigra.v2i2.39, accessed 12 October 2018.
Prerna Chikersal, Soujanya Poria, Erik Cambria, Alexander Gelbukh and Chng Eng Siong, 2015. “Modelling
public sentiment in Twitter: Using linguistic patterns to enhance supervised learning,” In: Alexander Gelbukh
(editor). Computational linguistics and intelligent text processing. Lecture notes in computer science, volume
9042. Cham: Springer, pp. 49–65.
doi: https://doi.org/10.1007/978-3-319-18117-2_4, accessed 12 October 2018.
Kirsten Chuba, 2018. “‘Star Wars’ actress Kelly Marie Tran leaves Instagram after harassment,” Variety (5
June), at https://variety.com/2018/biz/news/star-wars-kelly-marie-tran-leaves-social-media-harassment-
1202830892/, accessed 8 September 2018.
Jeremy Diamond, 2018. “DNI Dan Coats: Russian interference in US political system ongoing,” CNN.com (2
August), at https://www.cnn.com/2018/08/02/politics/dan-coats-russia-interference-election-
security/index.html, accessed 7 August 2018.
Randall R. Dipert, 2010. “The ethics of cyberwarfare,” Journal of Military Ethics, volume 9, number 4, pp.
384–410.
doi: http://dx.doi.org/10.1080/15027570.2010.536404, accessed 12 October 2018.
Ian Doescher, 2016. “The empire is us: The politics of ‘Rogue One’,” Politico Magazine (25 December), at
https://www.politico.com/magazine/story/2016/12/politics-of-rogue-one-214549, accessed 14 August 2018.
Colin Doty, 2013. “Believing the Internet: User comments about vaccine safety,” Selected Papers of Internet
Research, at https://spir.aoir.org/index.php/spir/article/view/1058, accessed 7 October 2017.
Leon Festinger, 1962. “Cognitive dissonance,” Scientific American, volume 207, number 4, pp. 93–107.
doi: http://dx.doi.org/10.1038/scientificamerican1062-93, accessed 12 October 2018.
Bill Fitzgerald and Kris Shaffer, 2017. “Spot a bot: Identifying automation and disinformation on social
media,” Data for Democracy (5 June), at https://medium.com/data-for-democracy/spot-a-bot-identifying-
automation-and-disinformation-on-social-media-2966ad93a203, accessed 7 August 2018.
Christian Fuchs, 2017. “Written evidence submitted by Professor Christian Fuchs, University of Westminster,
Communication and Media Research Institute & Westminster Institute for Advanced Studies,” at
http://westminsterresearch.wmin.ac.uk/20505/1/73970.pdf, accessed 27 January 2018.
Kenneth Geers, 2010. “The challenge of cyber attack deterrence,” Computer Law & Security Review, volume
26, number 3, pp. 298–303.
doi: http://dx.doi.org/10.1016/j.clsr.2010.03.003, accessed 12 October 2018.
Keir Giles, 2016. “Russia’s ‘new’ tools for confronting the West: Continuity and innovation in Moscow’s
exercise of power” (21 March), at https://www.chathamhouse.org/publication/russias-new-tools-confronting-
west, accessed 27 January 2018.
Nathaniel Gleicher, 2018. “Removing bad actors on Facebook,” FB.com (26 June), at
https://newsroom.fb.com/news/2018/06/removing-bad-actors-from-facebook/, accessed 12 October 2018.
Robert Gorwa, 2017. “Computational propaganda worldwide: Executive summary,” Oxford Internet Institute
(19 June), at https://www.oii.ox.ac.uk/blog/computational-propaganda-worldwide-executive-summary/,
accessed 27 January 2018.
Jonathan Gray, 2003. “New audiences, new textualities: Anti-fans and non-fans,” International Journal of
Cultural Studies, volume 6, number 1, pp. 64–81.
doi: http://dx.doi.org/10.1177/1367877903006001004, accessed 12 October 2018.
Luke Holland, 2018. “Why are (some) Star Wars fans so toxic?” Guardian (7 June), at
https://www.theguardian.com/film/2018/jun/07/kelly-marie-tran-rose-why-are-some-star-wars-fans-so-
toxic, accessed 7 August 2018.
Philip N. Howard, Gillian Bolsover, Bence Kollanyi, Samantha Bradshaw and Lisa-Maria Neudert, 2017. “Junk
news and bots during the U.S. election: What were Michigan voters sharing over Twitter?” Oxford Internet
Institute, Data Memo 2017.1, at https://www.oii.ox.ac.uk/blog/junk-news-and-bots-during-the-u-s-election-
what-were-michigan-voters-sharing-over-twitter/, accessed 27 January 2018.
7/13/2019 Bay
https://firstmonday.org/ojs/index.php/fm/rt/printerFriendly/9388/7603 12/14
Henry Jenkins, 1992. Textual poachers: Television fans & participatory culture. New York: Routledge.
Derek Johnson, 2007. “Fan-tagonism. Factions, institutions, and constitutive hegemonies of fandom,” In:
Jonathan Gray, Cornel Sandvoss and C. Lee Harrington (editors). Fandom: Identities and communities in a
mediated world. New York: New York University Press, 285–300.
Monica Kaminska, John D. Gallacher, Bence Kollanyi, Taha Yasseri and Philip N. Howard, 2017. “Social media
and news sources during the 2017 UK general election,” Oxford Internet Institute, Data Memo 2017.6, at
https://www.oii.ox.ac.uk/blog/social-media-and-news-sources-during-the-2017-uk-general-election/,
accessed 12 October 2018.
Anna Kata, 2012. “Anti-vaccine activists, Web 2.0, and the postmodern paradigm — An overview of tactics
and tropes used online by the anti-vaccination movement,” Vaccine, volume 30, number 25, pp. 3,778–
3,789.
doi: http://dx.doi.org/10.1016/j.vaccine.2011.11.112, accessed 12 October 2018.
Srijan Kumar, Justin Cheng, Jure Leskovec and V. S. Subrahmanian, 2017. “An army of me: Sockpuppets in
online discussion communities,” WWW ’17: Proceedings of the 26th International Conference on World Wide
Web, pp. 857–866.
doi: http://dx.doi.org/10.1145/3038912.3052677, accessed 12 October 2018.
Elina Lange-Ionatamishvili and Sanda Svetoka, 2015. “Strategic communications and social media in the
Russia Ukraine conflict,” In: Kenneth Geers (editor). Cyber war in perspective: Russian agression against
Ukraine, NATO CCD COE Publications, at https://www.stratcomcoe.org/strategic-communications-and-social-
media-russia-ukraine-conflict, accessed 30 December 2017.
Clare Llewellyn, Laura Cram, Adrian Favero and Robin L. Hill, 2018. “For whom the bell trolls: Troll behaviour
in the Twitter Brexit debate,” arXiv (26 January), at https://arxiv.org/abs/1801.08754, accessed 12 October
2018.
Hannah Lobel, 2012. “Cyber War Inc.: The law of war implications of the private sector’s role in cyber
conflict,” Texas International Law Journal, volume 47, number 3, pp. 617–640, and at
http://www.tilj.org/content/journal/47/num3/Lobel617.pdf, accessed 12 October 2018.
Jamie Lovett, 2017. “‘Star Wars: The Last Jedi’ has record Friday-to-Friday drop for a Star Wars movie,”
Comicbook.com (24 December), at http://comicbook.com/starwars/2017/12/24/star-wars-the-last-jedi-box-
office-drop/, accessed 7 August 2018.
Adrienne Massanari, 2017. “#Gamergate and The Fappening: How Reddit’s algorithm, governance, and
culture support toxic technocultures,” New Media & Society, volume 19, number 3, pp. 329–346.
doi: https://doi.org/10.1177/1461444815608807, accessed 12 October 2018.
Adrienne L Massanari and Shira Chess, 2018. “Attack of the 50-foot social justice warrior: The discursive
construction of SJW memes as the monstrous feminine,” Feminist Media Studies, volume 18, number 4, pp.
525–542.
doi: https://doi.org/10.1080/14680777.2018.1447333, accessed 12 October 2018.
Scott Mendelson, 2018a. “Box office: ‘The Last Jedi’ got slammed for performing like a ‘Star Wars’ movie,”
Forbes (4 January), at https://www.forbes.com/sites/scottmendelson/2018/01/04/4-ways-the-last-jedi-got-
slammed-for-performing-as-predicted/, accessed 7 August 2018.
Scott Mendelson, 2018b. “How toxic ‘Star Wars’ fans are hurting ‘Star Wars’,” Forbes (8 June), at
https://www.forbes.com/sites/scottmendelson/2018/06/08/star-wars-toxic-fans-kelly-marie-tran-last-jedi-
ghostbusters-oceans-8-incredibles/, accessed 7 August 2018.
Anna Menta, 2018. “Racist attacks against Kelly Marie Tran posted to Rose Tico’s ‘Wookieepedia’ page,”
Newsweek (19 December), at https://www.newsweek.com/racist-rose-tico-wookieepedia-page-753063,
accessed 7 August 2018.
Todor Mihaylov and Preslav Nakov, 2016. “Hunting for troll comments in news community forums,” 54th
annual meeting of the Association for Computational Linguistics, ACL 2016 — Short papers, pp. 399–405.
Nash Information Services, LLC, 2018. “Star Wars Ep. VIII: The Last Jedi (2017),” The Numbers, at
https://www.the-numbers.com/movie/Star-Wars-Ep-VIII-The-Last-Jedi#tab=video-sales , accessed 7 August
2018.
Ben Nimmo, 2015. “Anatomy of an info-war: How Russia’s propaganda machine works, and how to counter
it,” StopFake.org (15 May), at https://www.stopfake.org/en/anatomy-of-an-info-war-how-russia-s-
propaganda-machine-works-and-how-to-counter-it/, accessed 12 October 2018.
Thomas E. Nissen, 2015. “The weaponization of social media: Characteristics of contemporary conflicts,”
Royal Danish Defence College, at https://www.stratcomcoe.org/thomas-nissen-weaponization-social-media,
accessed 12 October 2018.
Michael O’Connor, 2016. “What are the politics of ‘Star Wars’?” Newsweek (14 September), at
https://www.newsweek.com/does-star-wars-endorse-political-republican-democrat-498365, accessed 14
August 2018.
7/13/2019 Bay
https://firstmonday.org/ojs/index.php/fm/rt/printerFriendly/9388/7603 13/14
Abby Ohlheiser, 2015. “Why ‘social justice warrior,’ a Gamergate insult, is now a dictionary entry,” Washington
Post (7 October), at https://www.washingtonpost.com/news/the-intersect/wp/2015/10/07/why-social-justice-
warrior-a-gamergate-insult-is-now-a-dictionary-entry/, accessed 4 August 2018.
OSoMe (Observatory on Social Media), 2018. “Botometer,” Indiana University Network Science Institute, at
https://botometer.iuni.iu.edu/#!/, accessed 7 August 2018.
Ben Popken, 2018. “Twitter deleted 200,000 Russian troll tweets. Read them here,” NBC News (14 February),
at https://www.nbcnews.com/tech/social-media/now-available-more-200-000-deleted-russian-troll-tweets-
n844731, accessed 7 August 2018.
William Proctor, 2018. “Disney’s Star Wars: Episode III — Revenge of the Crit,” Confessions of an Aca-fan, at
http://henryjenkins.org/blog/2018/2/4/disneys-star-wars-episode-iii-revenge-of-the-crit, accessed 5 August
2018.
William Proctor, 2017. “‘Bitches ain’t gonna hunt no ghosts’: Totemic nostalgia, toxic fandom and the
ghostbusters platonic,” Palabra Clave, volume 20, number 4, pp. 1,105–1,141.
doi: http://dx.doi.org/10.5294/pacla.2017.20.4.10, accessed 12 October 2018.
Sharon Rider and Michael A. Peters, 2018. “Post-truth, fake news: Viral modernity and higher education,” In:
Michael A. Peters, Sharon Rider, Mats Hyvönen and Tina Besley (editors). Post-truth, fake news: Viral
modernity & higher education. Singapore: Springer Singapore, pp. 3–12.
doi: https://doi.org/10.1007/978-981-10-8013-5_1, accessed 4 April 2018.
Marco-Francisco Robles, 2018. “How will Star Wars: The Last Jedi’s box office plunge compare to Attack of the
Clones?” Revenge of the Fans (29 January), at https://revengeofthefans.com/2018/01/29/will-star-wars-last-
jedis-box-office-plunge-compare-attack-clones/, accessed 7 August 2018.
Anastasia Salter and Bridget Blodgett, 2017. Toxic geek masculinity in media: Sexism, trolling, and identity
policing. New York: Palgrave Macmillan.
Cornel Sandvoss, 2013. “Toward an understanding of political enthusiasm as media fandom: Blogging, fan
productivity and affect in American politics,” Participations, volume 10, number 1, pp. 252–296, and at
http://www.participations.org/Volume%2010/Issue%201/12a%20Sandvoss%2010%201.pdf, accessed 7
August 2018.
Peter Spiliakos, 2018. “Star Wars and the culture-war vortex,” National Review (24 July), at
https://www.nationalreview.com/2018/07/star-wars-films-cultural-debate-social-justice-politics/, accessed 14
August 2018.
Mel Stanfill, 2018. “The unbearable whiteness of fandom and fan studies,” In: Paul Booth (editor). A
companion to media fandom and fan studies. Hoboken, N.J.: Wiley Blackwell, pp 305–318.
doi: https://doi.org/10.1002/9781119237211.ch19, accessed 12 October 2018.
Mel Stanfill, 2010. “Doing fandom, (mis)doing whiteness: Heteronormativity, racialization, and the discursive
construction of fandom,” Transformative Works and Cultures, number 8, at
https://journal.transformativeworks.org/index.php/twc/article/view/256/243, accessed 12 October 2018.
doi: http://dx.doi.org/10.3983/twc.2011.0256, accessed 12 October 2018.
Ryan Teague Beckwith, 2017. “George Lucas wrote ‘Star Wars’ as a liberal warning. Then conservatives struck
back,” Time (17 October), at http://time.com/4975813/star-wars-politics-watergate-george-lucas/, accessed
14 August 2018.
Cherie Todd, 2015. “COMMENTARY : GamerGate and resistance to the diversification of gaming culture,”
Women’s Studies Journal, volume 29, number 1, pp. 64–67, and at
http://www.wsanz.org.nz/journal/docs/WSJNZ291Todd64-67.pdf, accessed 12 October 2018.
Kelly Marie Tran, 2018. “I won’t be marginalized by online harassment,” New York Times (21 August), at
https://www.nytimes.com/2018/08/21/movies/kelly-marie-tran.html, accessed 8 September 2018.
Angela Watercutter, 2017. “Star Wars: The Last Jedi will be too inclusive for some people. Good,” Wired (15
December), at https://www.wired.com/story/star-wars-last-jedi-inclusion/, accessed 14 August 2018.
Jason Wilson, 2011. “Playing with politics: Political fans and Twitter faking in post-broadcast democracy,”
Convergence, volume 17, number 4, pp. 445–461.
doi: https://doi.org/10.1177/1354856511414348, accessed 12 October 2018.
Savvas Zannettou, Tristan Caulfield, Emiliano De Cristofaro, Michael Sirivianos, Gianluca Stringhini and
Jeremy Blackburn, 2018. “Disinformation warfare: Understanding state-sponsored trolls on Twitter and their
influence on the Web,” arXiv (28 January), at https://arxiv.org/abs/1801.09288, accessed 12 October 2018.
Asta Zelenkauskaite and Brandon Niezgoda, 2017. “‘Stop Kremlin trolls’: Ideological trolling as calling out,
rebuttal, and reactions on online news portal commenting,” First Monday, volume 22, number 5, at
http://firstmonday.org/article/view/7795/6225, accessed 12 October 2018.
doi: http://dx.doi.org/10.5210/fm.v22i15.7795, accessed 12 October 2018.
Amy Zimmerman, 2018. “The persecution of Kelly Marie Tran: How ‘Star Wars’ fandom became overrun by
alt-right trolls,” Daily Beast (5 June), at https://www.thedailybeast.com/the-persecution-of-kelly-marie-tran-
7/13/2019 Bay
https://firstmonday.org/ojs/index.php/fm/rt/printerFriendly/9388/7603 14/14
how-star-wars-fandom-became-overrun-by-alt-right-trolls, accessed 7 August 2018.
Editorial history
Received 15 August 2018; revised 8 September 2018; revised 13 September 2018; accepted 12 October
2018.
This paper is in the Public Domain.
Weaponizing the haters: The Last Jedi and the strategic politicization of pop culture through social media
manipulation
by Morten Bay.
First Monday, Volume 23, Number 11 - 5 November 2018
https://firstmonday.org/ojs/index.php/fm/rt/printerFriendly/9388/7603
doi: http://dx.doi.org/10.5210/fm.v23i11.9388
