Any Non-Free Speech Platform Will Become Left
The intolerant minority uses politeness as a tool for political conformity
The psychological profile of political ideologues is a driving force of changes in public sentiment and policy in the broadest sense. An intolerant minority has a disproportionate political influence if it is able to get others to change their behavior and language. The far left in the United States, especially those who can be called "woke", consists of a particularly intolerant group of people. This group views the world as entirely unjust. They view all statistical disparities as a reflection of historical or current oppression. They view speech as harm, going so far as to call minor faux pas and seemingly harmless comments "microaggressions." In their language they consistently conflate violence and speech. They view areas with restricted discourse as safe spaces.
Although many moderate liberals are not as extreme as the woke, they learn from them and adopt their values. Transgenderism was once strange and isolated to minorities but now it is mainstream. Concern for the well-being of transgender people is important for liberals. Although the idea of transgenderism may seem strange to some liberals, they will conform out of a sense of decency and politeness. Progressives view their ideology as merely being polite and avoiding harm. Social signaling in small ways has an important impact because liberals value creating new social norms and abandoning the old oppressive norms. They insist on "normalizing" things that are frequently not normal. Conservatives have their own causes but they see this sort of thing as a bit bizarre. Many conservatives have social causes they support but they refrain from discussing them due to liberal social pressure. While conservatives may not like liberals so much, many liberals really do not like conservatives. There is an old expression that "Conservatives think liberals are stupid. Liberals think conservatives are evil." This explains the trend of liberals frequently attributing bad motive to conservatives. They do this because they believe it and the more Machiavellian do it because it works to silence opponents with the threat of public shame.
The harm avoidance foundation for the far left is largely restricted to certain groups. There are certain harms that are legitimate and certain that are not. Being mocked for being white or a man is not a legitimate harm no matter how emotionally upset you are. Being upset for being mocked as a black woman is a legitimate harm no matter how slight the transgression. Blasphemy like "Oh my God" is not a cause for concern even if it is upsetting to religious people because religious people are the historical oppressors. Furthermore, religious people no longer try very hard to enforce these religious norms. The battle is mostly lost. Railing on the religious for being ignorant looks like bullying now. The zeitgeist shifts leftward.
Technology platforms like Facebook, Twitter, Google, Youtube, Reddit and so forth are composed of workers with college degrees and are more likely to be liberal. Many are not in the far left but they are sympathetic to the ideas or hold a more moderate form of the ideas. There are people within those organizations that have to determine what sort of things are to be regulated. As an organization, you do not want your website used for terrorist activities or filled with spam. This makes being a completely free speech platform difficult. What qualifies as spam or inciting violence? The platform may not want to be legally responsible for this so some moderation is necessary. Many platforms attempt to maintain some level of decency as well. Some attacks are clearly over the line like racial slurs. If you are a free speech platform and you do not ban anyone, you are the place where people go to do the most offensive things.
What is to be regulated and what is offensive? Why are racial slurs so harmful? They are needlessly mean to people and upsetting. They do not serve much else of a function. However, regulation extends further because the moderators view some speech as harmful in a similar way. The example that comes to mind most quickly is that of misgendering. Why is this so harmful? It is harmful because it causes psychological anguish and so it needs to be regulated. If the regulators don't feel this way, then social pressure will quickly be put on them. The left is more likely to organize and censor. If the people aren't censored via technology, they will censor them by trying to get them fired. Publicly available information has made this easier and businesses have demonstrated that they are willing to conform.
It is for these reasons that I speculate that "Any platform not specifically free speech, will become left." It is in the same vein as O'Sullivan's law that "All organizations that are not actually right-wing will over time become left-wing." If you are not committed to the idea of free speech, the leftists strong harm foundation and public pressure will drive the moderators to take a left-wing approach. The acceptable discourse will move leftward and the right-wing will be exiled to ghettos. Normal moderate people will not look at these ghettos and so they will be described as bastions of hate. There will be neutral websites and there will be right wing websites. Due to network effects, these websites will always be less popular and the right will be restrained in their ability to make influence.
Fantastic post.
In general this seems accurate, but I think in one case you are wrong about conservatives' desires to censor being less effective. Censorship of pornography is a typically conservative cause, but one shared by enough moderates to be followed by major technology platforms, so that many of them now either ban it entirely or hide it behind warnings and registration requirements (e.g. Twitter uses the same sorts of warnings about pornography that it does about right-wing politics in images & links).
An obvious corollary of the point that you make here is that, for censorship on some platform not to exert a political bias to the left, the platform must be difficult to censor overall. (Email and cryptocurrency are examples of this, as is the overall structure of the internet; IIUC Mastodon is trying to do this too but being more or less a Twitter clone they lack the popularity to make it work.) In practice, since some degree of censorship is necessary to prevent the platform from being filled with unpleasant (e.g. spam) or illegal (e.g. child pornography) content, this would seem to mean either that the design of the platform would necessarily give users the option to see unpopular content (as with email, where even spam is easily visible to those so inclined), or that users can choose between a variety of different moderation regimes (as with Mastodon hosts or early Reddit's subreddits); moreover, the platform must be decentralized, as a single host is a single point of failure for censorship (cf. Reddit). Thus a potential way to try to address this problem might be to encourage this sort of structure in developing large-scale platforms for online discussion. (This assumes that most important discussion will take place on large-scale platforms, but reversal of this trend currently seems unlikely.)