We have documented users in 30 languages including the ten most-spoken on earth. See this multilingual list of academic publications that cite DiscoverText.
Just a quick video research note from the field. Contact us to join the effort to better understand how Twitter amplifies certain political actors via a variety of platform affordances.
Find out why academics trust us to support theair collaborative, multilingual, multidisciplinary research.
The New Normal: Tulsi2020
Have you ever fallen into a spiraling lake of data? That is how it feels trying to understand the debate between Hillary Rodham Clinton and Tulsi Gabbard over the “Russia” question. In the last post, we alerted readers to the heavy presence of #MAGA movement identities (user_descriptions) in the “Tulsi OR Gabbard” Twitter data which is accumulating at a remarkable rate.
Today, while we are pouring over the translated Russian and other non-English user bios, I wanted to share something new emerging from our inductive research methods.
Trump Trolls Love to Post about Tulsi Gabbard
There are a lot of things happening in this data. We are not finding one trend, a single pattern, or a clear playbook. It’s more like walking around an Iowa hog lot; no matter where you go, everything stinks. For example, we are seeing a lot of interesting activity in the accounts that are 1-day old.
The Top 20 Most Viral Tweets about Tulsi in Our Collection
Based on more than 750,000 Tweets collected from the Twitter Search API over the last three or so days, these are the most viral retweets are shown here.
That is a bunch of hot topics for a 2% candidate. So who are the most active screen_names? What are the top 10 user descriptions?
As in many of these posts, we invite the reader to reflect on these data. You can visually inspect samples from the users and hashtags cited here using the Advanced Search feature provided by Twitter. In a complex and ambiguous space, the value of manual inspection cannot be over-stated.
The accumulation of strategies and accidents, humans and machines, truths and distortions, all in one semi-regulated information system is a fact we live with. In every open democratic debate, with or without Twitter, there will be authentic and staged performances. All the world is a stage. It is worth noting, in this context, the special platform affordances of Twitter, including automated retweeting, post scheduling, private networks, and so on.
Sometimes it is precisely the lack of affordance that is notable as well. For example, the landscape of very limited authentication of identity. I am a PhD-holding recovering academic, a former 7-year customer of Gnip & Twitter, with many public presentations posted on legitimate third party web sites, yet I cannot get a blue Twitter verified check mark next to my name. Why am I unverifiable? I was a founding board member and Treasurer of a 501(c)(6) dedicated to the long term preservation of the social data ecosystem.
By enabling extreme forms of inauthentic communication, networks like Twitter seek commoditize the transmission of misleading information. There are ad tech connections creating incentives to allow this behavior.
Having spent the last two weeks documenting the extensive role of MAGA-related Twitter accounts in the Canadian election, let’s turn back to the United States and check in with the Democratic party primary where a food fight recently broke out.
Candidate Tulsi Gabbard is getting a much closer look these days. The assertion by HRC that she is a product of Russian grooming definitely got a lot of press, but most of it was opinion-based (bashing Clinton, see Van Jones on CNN for example) rather than a data-based analysis of the role that foreign actors have played promoting the Gabbard candidacy. According to a report on the event in The Hill:
“The Russians already have their “eye on somebody who’s currently in the Democratic primary and are grooming her to be the third-party candidate,” she said, in an apparent reference to Gabbard. “She’s the favorite of the Russians. They have a bunch of sites and bots and other ways of supporting her, so far,” Clinton told David Plouffe, the podcast’s host and the campaign manager for former President Obama’s 2008 campaign.”
Who is Tweeting about Tulsi Gabbard
In a new series of posts, we will inductively explore Twitter data that mentions “Tulsi” or “Gabbard” to see what appears in the digital vapor trail of her Twitter supporters and opposition. It definitely caught my attention when it was reported that Twitter CEO Jack Dorsey has maxed out his contributions to Tulsi2020.
Leaving aside her billionaire backers (Dorsey, a platform Czar, and Vladimir Putin, the other kind of Czar), who else is pushing Candidate Gabbard into the trending column on Twitter as well as the crowded Democratic debate stage in the primaries? Maybe it is these folks?
You Be the Judge
We collected over 100,000 mentions of “Tulsi” or Gabbard” produced by Twitter users with 49,082 distinct user_descriptions in under 24 hours on October 20, 2019. Using a short set of search terms and the Boolean OR operator (God OR Patriot OR Brexit OR Swamp OR Trump OR MAGA OR KAG OR QAnon OR Sharia OR Islam), we found >20% of the bios in the Tulsi Gabbard data have a clear pro-MAGA ideological orientation.
Let’s broaden this out a bit and scroll through some of the bios organized alphabetically in a spreadheeet, which is incredibly interesting if you ever try it!
None of this supports the claims that Tulsi Gabbard is an agent of the Russians. However, it does suggest we have significant and complex work ahead to guage the scope and nature of the support she receives not only from Trump supporters, but from a wider multilingual, global audience of fans and critics. We turn to that next.
In Part Eight, we drilled down into the thicket of user_descriptions that had MAGA, KAG, and Trump in them while also engaging in Canadian election discourse on Twitter. The conclusion, though bleak, bears repeating, as it speaks to the overlapping legitimate activities intermixing with actions and platform features that pose deep threats to the future of democratic discourse online.
“It is easy to spiral out of control into this data. It leads many directions, has numerous causes, each worm hole feels bottomless, and findings really cannot be reduced to a table, a visualization, an algorithm, or, apparently, eight blog posts.”
Yet that appears to be the point. It may be less about pulling a social media lever to change the outcome of a vote and more about creating currents of uncertainty and cynicism about the flow of information critical for the survival of democratic societies. If it is a worm hole by design, we need to keep asking about the technological affordances of the platform that enable and encourage the most destructive behavior.
More on Those MAGA Folks
The 83 exact duplicates in set discussed in Part Eight are dominated by #MAGA, #KAG, and #WWG1WGA. #KAG stands for “Keep America Great” and is President Trump’s 2020 campaign slogan.
#WWG1WGA is the motto “where we go one, we go all” used by many QAnon movement participants. “QAnon is a far-right conspiracy theory detailing a supposed secret plot by an alleged “deep state” against U.S. President Donald Trump and his supporters.”
Additional keyword exploration in the 11,743 items:
- MAGA turns 9,190 results. MAGA2020 returns 170 results.
- KAG returns 3,264 results; KAGA returns 51 results. KAG2020 returns 588 results; KAGA2020 returns 14 results.
- Trump returns 2,979 results. Trump2020 returns 1,625 results; “Trump 2020” returns 271 results.
- Conservative returns 1,997 results.
- WWG1WGA returns 1,466 results. “Where we go one, we go all” returns 8 results.
- Patriot returns 1,259 results.
- NRA returns 1,206 results.
- QAnon returns 674 results.
- Other recurrent keywords: christian (1,127), catholic (178), AmericaFirst (437), “America First” (111), Brexit (232), #TheGreatAwakening (107), “The Great Awakening” (23), #BuildThatWall (27), “build that wall” (124), TrudeauMustGo (158), MCGA (371).
Other points of some interest:
- “Anti” returns 216 results, and is combined with keywords such as guns, Trump, Trudeau, deep state, EU, fascist, and globalist.
- The flag emoji 🇨🇦 is in 152 user_descriptions whereas 🇺🇸 is in 1,651.
- There is an unusually large percentage of hashtagged user_descriptions.