Avoiding Selection Bias
[This post has been renamed from “Desilencing”, pending changing what I call the action.]
edit 2019-06-25: this post is tone-deaf about ways people who experience the more common and dramatically stronger silencing forces of prejudice would see it. The insight here appears to me to be valid as an incremental change in an environment with low but nonzero hostility; it’s not as immediately relevant when a very large change is needed. I have changed my vote on this post to a downvote.
I often find that I filter my urges to give feedback, especially negative feedback, in public.
For example, when downvoting someone, I often feel an urge to say why. But then I hesitate because I worry that they will feel insulted, and attack me for my trouble of explaining myself.
This fear is not unfounded. sometimes when I say why, people do in fact challenge it.
But if I was on a discussion board with a bunch of slightly different myselves, and I never gave the other mes feedback, I would never get any feedback from them.
So, some semi-random fraction of the time, I say the thing anyway, in a short message with little overhead for me. I’m taking some risk, because then I say things that might get me in a fight. But people get more detailed feedback, instead of simply being ghosted or downvoted away because I’m scared of the fact that it’s unsafe for me to be straight with them.
So when I say I’m “de-silencing” myself—this is what I mean.
I call it “de-silencing” because I do it to break the attractors that silencing forces on me create. Some non-negligible portion of the time, those forces do specifically intend to silence people. And this technique would not work if someone was specifically out to get me.
This post is itself a de-silencing post: I’m not putting as much effort into it as I think would be necessary to ensure it gets a good reception.
Praise in public, criticise in private is something that I have heard and seems reasonable.
In most circumstances this is how I try to approach things. One issue is that the private messaging system on LW2.0 is currently wonky (it works, but not in a way anyone would really want to use).
Another is that I feel like idealized LW culture would hopefully do some kind of “make people feel safe to politely criticize others.” And part of that means that there should be public instances of people criticizing each other in a clear/constructive tone, and people taking the feedback to heart, and people who respond badly getting some pushback and people who respond well getting some kudos and people who try to take advantage of them system (either using “constructive criticism” as a thin veneer to attack people, or using “I accept your feedback!” as a way to dodge responsibility) getting called out… hopefully with an end result of “it actually feels safe to give and receive feedback.”
We’re not there yet (at least, not in all directions across all people on the site), but hopefully can be. In the meanwhile I think Lahwran’s policy is a reasonable middle ground
It seems like a reasonable and important experiment to try.
Watch this (all of it): https://vimeo.com/38247060
I think that the benefit of criticizing publicly is that it allows your criticism to in turn be criticized.
Let us say that Alice writes a post. Bob finds the material too <adjective> to interest him. If he messages Alice privately, that is the extent of the feedback. If, however, he comments as such, Carol, Daniel, and Eve may all chime in saying that they found the material the right about of <adjective> to be interesting. The vocal minority inspires feedback from the silent majority, who might not have independently thought to give Alice feedback (because they didn’t realize their views weren’t universal, because they didn’t believe they had actionable feedback, because they didn’t have something they disagreed with).
Like all systems of voluntary feedback, there’s a great deal of self-selection involved; moving from private to public feedback just affects who selects themselves. But I think it can do so in a valuable direction.
If you have a policy of always giving feedback, then people who are disproportionately sensitive to negative reinforcement (somewhat me) will avoid interacting in the first place.
If you have a policy of always responding to feedback, then people who are disproportionately sensitive to negative reinforcement will avoid giving feedback.
De-silencing is what I call my policy of sometimes just doing the thing even though someone might give feedback or meta-feedback, even though it’s risking being painful. I made this policy because I decided that saying things moves toward a better attractor.
Note that I am describing my own behavior, not prescribing it. I predict others will like and adopt this, but I also expect a bunch of people to hate it and not do it, some of them being people who don’t feel the “place yourself as an instance of people who make this decision” thing as being important enough to be worth the pain.
Private criticism from strangers can be particularly stressful if you have lots of readers, because there’s an implicit request that you respond to it (and then maybe get in a stressful argument, etc). Public criticism you can avoid responding to if you don’t have the emotional energy for it, and if you do respond it might be seen by everyone else who wants to make the same criticism.
Private criticism from friends is probably a good idea.
I seen it mainly recommended for Employer-Employee relations. So private (or no?) criticism from high status people also might be a good idea.
This is all hairy stuff.
This is something I strongly encourage, even though it creates awkward situations where two people disagree about the truth values underneath a claim.
In more detail: when I have a problem and I ask for help with it, what I’m usually *actually* asking for is new hypotheses to put into my hypothesis space. I get to assign probability to those hypotheses. “I think your p[hyp107]=0.35 is too low; it should be 0.995” is not and cannot be valid criticism; if people mention a hypothesis that I already have in my hypothesis space, they should accept when I say “yes that’s hyp223, p[hyp223]=0.05; got any other ideas or any particularly strong evidence around p[hyp223]?”
But more often, people just want to *assert* that a particular hypothesis is true, and demand that I update.
I can think of two ways that one can de-silence without risking falling into this trap: one is kind of a Bayesian version of NVC, where you present potentially novel hypotheses with “what probability do you assign to the hypothesis that [X]?”. This makes it sound like you aren’t demanding that they believe X, *and* that you aren’t assuming that they haven’t thought of X, while still bringing their attention to X in a helpful way. The second method is to just dive in there, but be completely okay with them pushing back and challenging you on X, and not feeling attacked or disrespected if they appear to reject X out of hand.
I hope that made sense?
This feels related to the “community space” attitude I mentioned here. Having the attitude that we live here, and it is OK to just say things sometimes, because people are likely to be interested in what each other think.
Interesting post, but it’s been mentioned as a problem before that posts on LW tend to get too much criticism in that people pick apart every individual point.
There’s certainly something to be said, at least for those of us who tend to try to shy away from potential confrontations, for deliberately being a little more vocal.
lahwran made a comment earlier today that he explicitly labelled “De-silencing comment”. I was baffled at the time by this label; now I understand what it meant, but still can’t convince myself that “De-silencing” is a good name for the thing; perhaps the name plus a link here might do OK, but what the name immediately suggests to me is something more like “The viewpoint I’m about to express is in danger of being suppressed, so I’m bravely fighting against oppression by saying it”—which, when attached (as I suppose it always will be) to a comment of the sort that is more than averagely likely to provoke a fight, doesn’t look good: it seems like an attempt to seize the moral high ground while being rude to someone else.
With the context of this article, of course, it looks different: the prefix “De-silencing comment:” needs to be interpreted as something more like “I know that this may sound confrontational, and for that very reason I nearly didn’t say it, but I’m trying to increase the overall amount of feedback people here get. I hope this won’t provoke a fight.”.
But a thing that strikes me here is this: the LW community really doesn’t, so far as I can tell, have the reputation of being a place where ideas don’t get frank enough criticism. I’m not convinced that the community as a whole is really taking that side of the tradeoff between “informative” and “polite”. I’d be more confident of the experiment being helpful overall if LWers who embrace confrontation were running a parallel experiment of trying to be a couple of notches nicer even at the cost of not always giving the challenging feedback that they think would be useful.
I guess the thing to do is to watch what actually happens. Do we end up having more fights? Do we end up learning more? I think it’ll be hard to tell...
But the character of LW definitely has changed. In the early days when it was gaining its reputation, it seemed to be a place where you could often find lots of highly intense, vociferous debate over various philosophical topics. I feel that nowadays a lot of that character is gone. The community seems to have embraced a much softer and more inclusive discourse style, trying to minimize the overall amount of offense and anxiety one could feel when they are trying to engage here. And indeed, a large amount of the discussion topics these days are surrounding social norms and community norms, which is a pretty big difference from the original LW culture.
I’m not completely sure how I feel about this. On the one hand, we seem to have created a place that seems much more approachable (and the changes might probably be a factor in why I began to take part in the discussions somewhat recently), and less intimidating to newcomers. So I think that part of it is good. On the other hand, some of the best thinkers were people who had a pretty aggressive conversational style (such as EY himself). I think these people have mostly switched over to private discussions among people they trust with limited public engagement. It’s unclear how much we are missing from them.
There are obvious situations where you would desire one style of discourse versus the other: In situations where you need to gather support from many groups with potentially competing interests and complex needs, you would definitely prefer to be as inclusive and inoffensive as possible, while limiting discussion to matters that relate to everyone. In the situation which amounts to casual debate among trusted people, it becomes a lot safer to be as frank and transparent as possible.
LW began very organically as conversations between many people who already interacted quite frequently sort of merged into a single locus. That meant that lively and open debate could happen somewhat safely at least initially, but that changed as soon as the space became more known outside of its bubble and grew.
There’s nothing that can be done to prevent that from happening, I think, but one thing that can be done is to siphon aggressive conversations into separate locations, away from the front-facing side but still possible to enter if you’re prepared.
I’m noticing a lot of anxiety in myself about your disagreement with the name. I realized that asking you for feedback on the name, and suggestions for what to change it to, would make my anxiety go away. What are your thoughts on better names for this, that still capture the essence of “I’m dropping this tidbit here because otherwise you won’t even be able to know something is wrong, and/or what it is”?
I edited the post title. But I also notice that this discussion we’re having is exactly the sort of thing that, when I anticipate, would normally lead me to not make a post at all; if that’s not “de-silencing”, what should it be?
I regret being a source of anxiety. I think the things I don’t like about “de-silencing” are (1) that it’s liable to suggest that it’s a fix for being silenced rather than for being silent (i.e., that the problem is some kind of deliberately-applied pressure from outside rather than internal anxieties etc.) and (2) that it’s rather uninformative—I mean, any time you say anything you’re making yourself less silent/silenced, and presumably the point is to pick out particular instances, namely those in which you might have avoided saying anything for fear of confrontation (or whatever).
So what might be better? Nothing leaps out at me as being a really good name. (Which shouldn’t be a big surprise; probably most things don’t have obvious really good names, and most people can’t reliably locate them even when they exist.) I might consider something cute like “avoidance avoidance” (which would need an explanatory link just as much as “de-silencing” does, but at least avoids the help-help-I’m-being-oppressed vibe), or I might call instances of the pattern something vaguer like “constructive criticism” (which doesn’t draw attention to the motivation you’ve described here, but it’s not clear to me that that’s always a problem), but I would be immensely unsurprised to find that there are much better options than those.
(Maybe there’s a useful distinction to be drawn, between (1) a name for use in a post like yours here, and (2) a name for labelling instances as they arise. I’m not sure #2 is actually needed at all; after all, most comments don’t come tagged with explanations of the motives behind them. Hmm, and to what extent do we need #1 either? “Avoiding selection bias” is your current title for the article; it’s not a name for the action of “de-silencing” but that doesn’t make it a bad title, does it?)
I don’t just mean for this to be about confrontations.
That is how I meant it, though “oppression” is maybe a bit stronger than I meant—I mean that people in a fairly common social attractor in mindspace tend to try to make anyone who disagrees with them into an argument target. I do this to prevent their credible threat of dismissing me if I try to tell them nicely from completely shutting me up.
That’s not what I mean, though; the place I got this name is that I often notice that I want to give feedback about a way someone has done something that excluded me, and they won’t be happy with me complaining. this policy is an attempt to make sure that some of the time, they still find out. I have found that it applies a bit more generally, and I execute this any time I worry that someone will be blackholed or ghosted or in some other way simply abandoned if they aren’t told what’s happening.
Yeah, I agree that there are things that I don’t often have the energy for that would make things come across better, and possibly things I do have the energy for that I don’t understand well enough to do easily/correctly. There may even be free actions I could take. These are all probably true of others. I’m interested in any comments you can make about how someone who’s constantly background on guard against rejection can still minimize rudeness.
(Keep in mind that I’m not just oblivious.)
I approved strongly of your instance of de-silencing earlier today, and felt gratitude for it, and silently upvoted.