Click here to read GNET's latest report Emergent Technologies and Extremists: The DWeb as a New Internet Reality?

Videogames, Twitter and Far-Right Extremism: An Analysis of Twitter Hashtag Networks

Videogames, Twitter and Far-Right Extremism: An Analysis of Twitter Hashtag Networks
3rd May 2022 Sam Andrews
In Insights

Many social network theories emphasise the importance of network properties. Strong networks with many interlocking and reinforcing connections are essential to reinforcing beliefs. Homophily within networks arises as individuals within them  seek out and maintain strong connections with those who are like them. People seeking to join a network are often already connected to people who hold similar beliefs. Thus, as more people join the network, the strength of the ties between individuals grows stronger and homophily increases.

The homophilic nature of such networks, however, precludes novel information. Thus the Strength of Weak Ties theory notes that connections with others who are different are likely to be ‘weak’ as there are fewer connections.  But these connections bring novel ideas to the network. Thus, these connections are also ‘strong’ in that they bring value to a homophilic network they provide bridges to different ideas as well as connections to others outside the network. Studies of terrorism, extremism and radicalisation that look at network connections usually emphasise the power of these external bridging connections. A chance encounter with an extreme actor can, for instance, bring in new ways of understanding the world which might be radicalising. They also bring the possibility of networking with others who are far outside of the social network of the average person. 

This Insight provides an exploratory look at network connections between tweets on gaming, and tweets on extremist topics. As we explored in our previous Insight, concerns have been raised around whether gaming communities are vulnerable to extremism. Some experts worry that gaming communities are being used by the far-right and other extremists to incite hatred and violence. There is, however, little evidence to suggest that gaming communities are any more vulnerable to extremism than other groups. That said, the GamerGate movement is hostile to minorities, misogynistic, and shares affinities with some far-right ideas. Therefore, this particular movement could serve as a bridge between mainstream gaming communities and far-right networks. 

Using the Twitter API, we collected 18,481 tweets over a 48-hour period in December 2021. Each of these tweets contains at least one of the following Hashtags: #gamersriseup, #gamers, #gaming, and #gamergate. Using the #gaming hashtag, a user links their Tweet with others commenting on a similar topic. Thus, hashtags allow users to network with those who share similar interests and ideas. The hashtags #gaming and #gamer capture some of the most popular gaming discussions on Twitter and are used by video game publishers, streamers, and everyday gamers. Conversely, the hashtags #gamersriseup and #gamergate capture a niche section of the gaming community that might be connected to reactionary, right-wing attitudes. While this article provides a small sample of Twitter interactions and thus should be taken as exploratory and not generalisable, it nonetheless provides an overview of typical interactions over a 48-hour period on these hashtags.

The data was processed using R, isolating hashtags, nodes, and edges from the data. In this dataset, every hashtag is a node, and the connections between hashtags are edges. The dataset gives us information on how different hashtags are connected to wider gaming networks and communities. Hashtags were then compared to a dictionary of abusive, misogynistic, and far-right terms. Betweenness centrality scores were calculated for each node, which measures the connecting power of a node in the network. A high betweenness score indicates that the node functions well as a bridge between other nodes, and can be used to easily connect different nodes in the network. 

In total, 14,663 unique hashtags were extracted from the data. A total of 427,588 hashtags appear in the dataset. The most popular hashtag, #gaming, appears 15,470 times and has a betweenness score of 79,713,327. As #gaming was a search term, it is expected that this would be extremely powerful within the network. Comparatively, the next most popular hashtag that is not a search term is #twitch, referring to the popular streaming platform, appearing 4,101 times and having a betweenness score of 2,939,614. Most of the hashtags relate to video games and video gaming culture – very few of the hashtags extracted were found to correlate with extremist terms:

#gamergate 8033,579
#trump 501,117
#dlive 49266
#pussy 373,985
#escorts 220.4
#imnotracist 220
#maga 200
#whoresofthrones 180
#gamersriseup 15327
#bitchute 141
#trump2024 140
#npclife 1319
#leftists 121
#bitchutegaming 110
#racist 70
#odysee 60
#npcneedlovetoo 50

The total number of extremist hashtags in the dataset is 395, representing just 0.09% of the total hashtags in the dataset. This only includes hashtags within the dictionary mentioned above. Other hashtags which could be considered extremist were found within these networks, but these did not appear in the dictionary. However, there were very few, and many had marginal influence. The low betweenness score indicates that they are not well connected to the wider gaming networks. This indicates that – at least on Twitter, and within this dataset – extremists have low networking power within gaming communities. However, these hashtags do function as bridges between gaming communities and extremist networks. For instance, the #trump hashtags create bridges to other extremist hashtags. #Trump creates a network that links #gaming with conspiracy hashtags like #nwo, #infowars, #ccp, #flatearth and #freemasons. If support for Trump were to become more influential within the gaming community, then it is likely that it would more effectively bridge the gap between gaming and other extremist topics, like the New World Order conspiracy. 

This largely echos the findings of our previous Insights exploring extreme language in gaming communities and looking in detail at how that language is used within those communities. There is no evidence here to suggest that extremist networks are influential or welcome in gaming communities, although a larger dataset, or a dataset looking at different hashtag seeds, might reveal further or more influential relationships.

However, as with previous findings, this data highlights the challenge posed by the GamerGate movement. This movement, which draws on reactionary and far-right ideas, is unique to the video gaming community. It is a movement which plays on and plays with the hyper-masculine and misogynistic cultures that video gaming has struggled with for decades. The high betweenness of #gamergate within the network, as compared to all other extremist terms within the dataset indicates that GamerGate should not be subsumed into our generalist assumptions about right-wing and extremist mobilisation on the internet. The data indicate that GamerGate is not a bridge to other far-right movements. Rather, it remains mostly contained within gaming communities. Users who tweeted about #GamerGate did not connect it to other far-right topics or movements. Thus, unlike the Trump hashtag, users interacting with #GamerGate were not exposed to further far-right content. 

However, tweets on #GamerGate did serve as a bridge to misogynistic tweets, and tweets about women in video gaming. Users tweeting about #GamerGate also tweeted hashtags like #gamergirl, #streamergirl, #pokimane and #twitchgirls, as well as misogynistic hashtags such as #pussy and #whores. This strengthens the idea that GamerGate is part of wider networked misogyny on the internet that includes anti-Feminist, Men’s Rights and Incel movements. Thus while within this dataset, #GamerGate did not link video gaming with white supremacist, far-right and extremist politics, it can be understood to  provide a bridge to misogynistic ideas and movements.

Online misogyny is a threat that cannot be underestimated; its links to real-world harassment and violence have been repeatedly noted by feminist researchers.  States do not have a good track record in fighting this. Thus, we should be aware that the recent interest in the relationship between video games, extremism and terrorism might wane as evidence continues to be lacking. Tackling online misogyny through securitising the communities where it arises is a matter of debate. It is worth asking, should we be mobilising the counterterrorism apparatus in response? Certainly, a wider debate around male supremacism in society and how to tackle it would be welcome here.