The necessary and utterly predictable outcomes of this system: to create runaway viral dissemination patterns for tweets, and to facilitate and encourage interaction among massive groups of people connected only by follower networks. More often than not, these are exactly the outcomes we want. People use Twitter to expand the reach of their content and to interact with a much larger audience than would ever be possible otherwise.
Occasionally - and again, utterly predictably - this system creates outcomes that we frown upon. A message disseminates virally that we think probably shouldn't have got so much attention. The audience we interact with is one we don't want to interact with. Someone gets criticized, shamed, or dogpiled well out of proportion to what we think was appropriate.
Individualizing a systematic problem
When this happens, the sensible approach would be to recognize that the magnitude and proportion problems that emerge from Twitter are intinsic to the system itself. Twitter storms a-brew not because any particular user wanted it to happen or made it happen, but because the entire logic of the platform's architecture and operation was designed to guarantee this outcome. It is a mathematically complex system whose behavior is much greater than the sum of its parts.
What often happens, however, is just the opposite. Instead of recognizing that there is something about Twitter itself that foments these outcomes, and developing a systematic understanding of how discourse works, we try to personalize it. We begin with the assumption that surely there is something that we as individual users could have done to have prevented the latest dogpile or shaming spectacle - and from there, we backfill ad hoc theories of influence or conspiracy meant to provide a chain of causality. In this way, we find a way to indict individual actors not only for their own (usually quite trivial) contributions to the Twitter storm, but for the storm itself, for its severity and magnitude.
Much of this second tendency, of course, is motivated by interpersonal / tribal rivalries and attempts - naively, or cynically - to lay blame at the feet of one's internet enemies. In these cases, there's rarely any serious or even token attempt to grapple with basic questions of causality or the minimal limits that we would, as a matter of common sense, place on burdens of responsibility; the goal, implict or explicit, is just to tie as much blame to one's opponent(s) as possible. The motives here are usually fairly transparent, and the tenuous see-the-patterns and connect-the-dots analyses not particularly difficult to poke holes in.
Only socialism can own the trolls
What I find more interesting is the way that this second tendency, in its attempt to individualize a problem that is largely systematic, replicates the ethic of consumer activism so central to liberal capitalist thought.
As Marx has taught us, many of the most notorious and destructive problems we experience under capitalism - the consolidation of power into fewer and fewer hands, the exploitation and immiseration of workers, the alienation of people from each other and from their labor, the commoditization of things we think should not be commoditized, the lies and distortions of bourgeois ideology, the proliferation of corruption and conflicts of interest, etcetera etcetera etcetera - these are all intrinsic to capitalism itself.
They should not, that is to say, be understood as mere incidental problems introduced into the system by individual actors, problems that we could eliminate or moderate if people would just behave themselves. Corruption in the financial sector, for example, is not something that simply comes from a few bad apples who haven't been arrested or educated out of their bad behavior; it some from things like the profit motive and our basically unlimited ability to create sinister financial instruments that work around any extant regulation. Homophobia at Chic-fil-A doesn't come from the personal failure of conscientious liberals to boycott the company out of existence; it comes (in part) from an economic system that foments bigotry as a way of dividing the working class.
Bourgeois ideology fundamentally rejects these systematic ways of thinking about power and oppression - whether on Twitter, or in our economy. Instead, capitalism fetishizes the John Galtian power of the individual to "stop the engine of the world" through sheer force of will and piety. If we can all just learn to be responsible consumers, and responsible managers, and responsible oligarchs, and - yes - responsible tweeters, we can make the system work.
The leftist solution to these kinds of problems is to acknowledge the minimal limitations of human agency and culpability, and to refuse to let individuals become scapegoats for problems intrinsic to the system itself. In my view, a great way to get rid of the meanness and bigotry we often see on social media is to change the toxic culture that it comes from - and that means, first and foremost, challenging capitalism. I also don't think that people should lose jobs or have their basic livelihoods threatened by things they say on social media, which is why I propose a universal welfare state that places into the hands of workers control of the means of production. These are admittedly extraordinarily ambitious goals, but I think they're also infinitely more plausible than this hilarious liberal fantasy that one day the trolls and haters will start behaving themselves.