My working theory of the eigentweet is that it's more internal than people "wanting to target this audience." I don't think it requires much if any outside influence or explicit bad faith. (I'm a moderator on a non-American political subreddit, so most of my opinions below come from my own observations and musings.)
Instead, the decay of social media happens as the platform transitions from a place where one talks with people to a place where one talks at people.
If I reply to your political thoughts by telling you off, I'm probably not trying to convince you to change your mind. Instead, I'm performing for the attention (and upvotes / retweets / kind comments) of like-minded peers.
That's obviously alienating for the person who gets attacked, but this kind of performance is also self-radicalizing. The validating reinforcement preferentially goes to the strongest attacks or defenses, favouring rhetoric (as you noted) rather than substance.
One of the few things that suppresses this cycle is exactly what HN does reasonably well: have the community be about something else, diluting political content such that there's less often a chain-reaction.
Instead, the decay of social media happens as the platform transitions from a place where one talks with people to a place where one talks at people.
If I reply to your political thoughts by telling you off, I'm probably not trying to convince you to change your mind. Instead, I'm performing for the attention (and upvotes / retweets / kind comments) of like-minded peers.
That's obviously alienating for the person who gets attacked, but this kind of performance is also self-radicalizing. The validating reinforcement preferentially goes to the strongest attacks or defenses, favouring rhetoric (as you noted) rather than substance.
One of the few things that suppresses this cycle is exactly what HN does reasonably well: have the community be about something else, diluting political content such that there's less often a chain-reaction.