Ideolect, coined by my collaborator on this project Jeremiah Milbauer, is a play on ideology and idiolect, the speech habits peculiar to a particular person. We examine the dynamic between ideological communities and the language they use, with the overarching claim that users and communities have their own dialects and vocabulary. Consequently, we can meaningfully reconstruct ideological differences between users and communities on issues. I overlay a structuralist interpretation over this, arguing that user interactions and preferential attachment play a big role in determining participation, affiliation and the formation of ideolects. And that we can’t reasonably reconstruct inter- and intra-community dynamics without accounting for the digital spaces they inhabit.
We use data from Reddit, thanks to Jason Baumgartner’s Pushshift.io archives. Reddit is an interesting place to study this phenonmenon for serveral reasons:
Subreddits are self-regulated communities with norms created and enforced by users within it.
Subreddits are self-declared common interest groups. While the topic of interest may be fuzzy and broad-ranging, there is a component of a user’s identitiy or interest that is attached to the subreddit they join and enggae with.
I argue that subreddits are knowledge aand sense making communities. A bunch of users get together to bond over a common interest, but that sustained bond translates into a form of epistemic reliance on the faceless community that helps them make sense of the questions and events they post within it.
For my MA thesis, I’m trying to examine far-right and conspiracy-related subreddits. I have a host of questions I want to answer there, and I’m trying not to get ahead of myself. I’ll keep posting fieldnotes here and random musings.