For example, if someone has racist beliefs which do not affect his actions in any way, his effect on other people in society would be exactly the same as it would be if he did not have racist beliefs. In other words, beliefs that are not acted on or expressed have no effect on society.
There may be some social benefit to knowing who the racists are. But if they are anonymous, that benefit is probably lost. Their speech is protected by the first amendment either way, but Reddit is not required to enforce the constitution, so they can shut down the accounts if they want to.
I would agree that unexpressed values would propagate less.
That was an extreme example used to illustrate the point I was making. Racism is also only an example; there are plenty of other beliefs that are bad for society. Personally, I suspect that sexism is probably far more prevalent on Reddit than racism, for example.
Social stigma can prevent one's beliefs from affecting one's actions, or at least reduce the effect of those beliefs on one's actions. If it is shameful to be perceived as a racist, racists will attempt to hide their racism by not acting in racist ways toward others, at least when other people are looking. This does not obviate the effects of unconscious biases or eliminate racism, but the net result for society is fewer overtly racist actions, and that's a good thing.
If racism becomes acceptable again, racists will feel freer to act on their beliefs, which will make life worse for the targets of their racism. So, there is a case to be made that if people repress their beliefs only partially, society is better off -- not as much better off as if certain people repressed their beliefs fully, but still better off.
The danger of allowing racism to flourish in online communities, especially ones as large as Reddit and Twitter, is that it re-normalizes racism and makes it seem acceptable to be a racist. It can be more than merely acceptable -- it can even be a contrarian badge of pride, justified as a rebellion against political correctness. (That kind of justification for it is reinforced when anti-harassment policies are not uniformly and consistently enforced, because it can make moderation appear to be arbitrary or politically motivated.) When that kind of thing happens, it makes society worse.
I do not agree with giving up on freedom of speech because some people will say hurtful things. Every time I have read racism or seen racists on TV it has merely exposed how ill-informed, counter-productive and just plain wrong those ideas are, and has prompted myself and others to dispute those wrong ideas. There is also the slim chance that if the dummies are called out on their wrong ideas, they may end up changing them.
Anonymous freedom of speech is a slightly different case, but I still feel that all freedom of speech is valuable and needs to be preserved.
Some would say that racism is still acceptable in USA today. Just look at Trump. (Yes not as openly as before, but it hasn't been erradicated)
There's a saying that the USA went from thinking that racism was OK, to not knowing what racism is.
However, I very strongly suspect that no matter what the level of racism on Facebook is, it would be far higher if people were able to post anonymously.
Counterexample HN and arstechnica.com
As I have previously hypothesised a lot of good users (me included) seems to be scared away by real names policies.
Pseudonyms allowed and linked to phone numbers or national ids combined with karma seems like a workable middle ground for me.
If a racist hermit is racist in the woods, does he make an offensive sound?
I have hypothesised that the reason is that without anonymity only
* (trolls with) fake accounts
* politically correct people and people who think they are pc
* and people who don't care about real life reputation anyway
post in any kind of controversial debate.
Normal knowledgeable people doesn't want to lose the dream job because of something they commented on 5 years ago.
* naive idealists who believe that truth and reason matter more than immediate public perception
This is subtly different than "people who don't care about real life reputation" in that they sincerely hope that despite any short term damage to reputation, in the long run others (or at least some subset) will come respect their honesty and rationality.
I'm probably naive and idealist, but at least on HN, I think this is a significant group. Although, am I reading correctly that you are considering HN to be "without anonymity"? In the current "persistent pseudonymous" system, what do you think constitutes a "fake account"?
No. Sorry if unclear. I categorise HN accounts as pseudonymous with optional real life names.
> In the current "persistent pseudonymous" system, what do you think constitutes a "fake account"?
What would annoy me is real sockpuppet accounts, controlling more than one account and using it to
* upvote ones own comments and postings
* astroturf
* using it as a strawman to defame opponents (i.e. by running a sock puppet account that seemingly supports some cases while posting bad arguments for that case etc.)
* etc
Personally I have nothing against using multiple accounts as long as there is a reason (i.e. someone has an account with full name and another that they use when they want to post or comment something that they do not want to reflect back on their real life persona, family and workplace.)
Honest question since two people read my post that way, I obviously was unclear, but where?
I wonder how feasible it would be for Reddit to implement subreddit-based web of trusts for users, as opposed to moderation. People could filter others (or not) individually and not have to rely on someone else to try to filter for the whole group.