Much of the election misinformation conversation in mainstream journalistic sources centers on a number of important, but limited, themes:
- How can the “truth” of information be policed online?
- How widely are falsehoods spread online?
- How widely are they believed?
- What are the impacts of these belief on candidate choice, come election day?
And so on. The broader theme tying all of these threads together is an insistence that truth and/or voter epistemology are important aspects of election-oriented influence campaigns. While this is true, it does not go nearly far enough. The optimal objective of an election-oriented information attack is to disrupt the get-out-the-vote (GOTV) operations of your competitor, the target of the operation.
Properly understood, an information attack is an effort by an attacker to impact the operational capabilities of their target. In such a framing, the target of an election-oriented information attack is not a population of voters, but a competing campaign.
A modern election campaign is centered on its capacity to turn out voters on election day. Targeting is part of that. So is messaging. But those tactical frames are not the end, they are means. Given the central role of GOTV in modern campaigns, its role as a core vulnerability should come as no surprise.
Computational propaganda efforts have leveraged this insight extensively: the targeting of Bernie supporters and black voters during the 2016 primary; the harassment of Hillary supporters online and in public; the promotion of #WalkAway as a core message among trolls. The effects of these attacks are sweeping and cumulative. The distinction between citizen and campaign is quite gray and the mission of campaign building is cumulative, with compounding interest. Early successes by political campaigns build opportunity for GOTV success later. Online disruptions can permanently derail a promising candidate.
In my next post I will talk about some specific examples of tactics and techniques deployed to achieve this goal.