Suppose that I were a hard-working person, and you wanted to hire somebody who was hard-working. Now I do things which signal my hard-workingness, and you see this and hire me because of it. As a result, I got a job and you got a hard-working employee.
(Honest) signaling is about communicating the fact that you have (positive) qualities which aren't immediately obvious. To the extent that other people care about knowing whether you have such qualities, signaling is a fantastic thing, and we should all be doing it. It's only wasteful or dishonest signaling that's a problem.
I think there is one pattern that being good at something (socially accepted) is good signalling... but trying to improve at something (and thus showing that you think you are not good enough yet) is bad signalling.
I have noticed this pattern years ago, when I was trying to write sci-fi stories, and was meeting with other people who tried the same. I found somewhere on internet that many famous writers attended some kind of writer workshops. So I suggested to my friends that we should find out whether such workshop exists near us, and if not, try to create our own workshop. Most of them were horrified by that idea. When I asked why, they told me that a person either has a talent for writing, or not. The former cannot learn anything at workshops, because the true art cannot be taught; only the latter could learn to become a more skilled art-less graphomaniac. I thought such reasoning was stupid, and asked some literary critics about it: but they confirmed that they would percieve a person known to have attended such workshops as an art-less wannabe, because the true talent must be born. I refused to accept their reasoning too (because I have read a few autobiographies of famous writ...
The important thing isn't to try to not signal things (which is of course impossible) but to be aware of the nature of one's own signaling and how it can impact the exchange of information and belief.
Look at this post by Robin Hanson: Smiles Signal. Nobody is arguing that smiling is bad. But if you think about it, you realize that if smiling weren't a signaling behavior, it probably wouldn't be visible to others. Awareness of the ten thousand ways these adaptations influence our behavior is a tremendous component of what it means to be a rationalist.
One should also beware of using "signaling!" as a fully general counterargument of the form:
My opponent argues for
position X. But in doing
so, he is only signaling
high status. Therefore,
not-X.
Please observe the following distinction:
All X are not Y
is not the same as
Not all X are Y
In your case, you are claiming that no signalling behaviours are bad. You probably intended to say that at least some signalling behaviours are not bad.
There's nothing wrong with signaling (turn signals are wonderful). But signals can be faked. (What precisely does a college degree mean?) Signaling is socially problematic when effort is spent on the signal instead of what it represents. (i.e. Cheap talk)
Tribal membership declaration and dominance games occur via signaling, but it is analytically clearer to treat those topics separately from signaling itself.
It's true that not all signaling and status posturing is bad. What happens is that we recognize how destructive so many of our signaling and social status tendencies are, but then apply an overly general heuristic of, "Avoid doing anything we could refer to as status posturing or signaling."
It's certainly the case that a lot of the status behaviors are dangerously mal-adapted for our current environment (especially when it comes to our epistemic rationality), but not all of them are, or rather most of them probably aren't--it's just that the dest...
Looking "good" is still based on priors, which in anorexics, vegans, and ascetics usually involve perceptions of costs their brains subconsciously figure would be reduced if people ate less, ate less meat, or consumed less of everything.
Some vegans feel disgust when thinking of meat, even lab meat
"Disgust as embodied moral judgment"
Generally, all signaling is good from the perspective of the signaler's brain, which may be updated, like when Buddha left groups of ascetics to continue optimizing.
The thought process for signaling should usually be something like this: I recognize that although X is associated with Y, X is actually just signaling. But, is my life better off if I do X anyway?
For example, making my bed is associated with being an clean person, even though making a bed doesn't actually clear up any space. But, I find it more aesthetically pleasing to make my bed every day, even though I don't get any other benefits from it. Plus, others coming into my room may also make judgements about me based on an unmade bed.
In your example people first notice you being better (that is high status) and only then become curious about the causes. My examples were about people noticing someone practicing, or just discussing a hypothetical practice in future, with no improvement yet. That's not the same situation.
Simply said, you get -1 point for trying, +5 points for succeeding. Problem is that trying comes first, succeeding later. So there is that unpleasant phase of "already trying, not yet succeeding", which you cannot avoid (though you can keep it secret). During this phase you have low status. Only later, when the success comes, your status becomes higher than it was originally.
The high-status answer to "Where did you learn that?" is "I am just naturally good at it". Of course that works only if it is credible, which depends on the audience.
For example if I would try to get high status for my programming skills, to a totally computer-illiterate person I could say "I just naturally understand the computers; I was like this since my childhood". No details necessary. To them, any computer skills are probably magic, accessible just for special kind of people, and I just confirmed the hypothesis.
To a fellow programmer I could say: "I played with computers since I was a child; then I participated in programming competitions and won them; then I studied university, which was rather easy for me; and now I just read some tutorial on the web or google a few examples, and I get it; anyway, most of the stuff is easy if you already know a lot". I cannot pretend that one can learn programming magically without learning; but I can still move my magic more meta and pretend that it's not my programming skills per se, but my learning-programming skills which are magical. Yes, I had to learn programming, but the learning was always easy and quick -- I never failed, never got stuck, never had to ask another person for help, never doubted my success for a moment. (Which is psychologically almost as unlikely as being born with magical programming abilities, but I would expect an average programmer to believe this and to feel inferior compared to it.)
If I can take your answer literally, perhaps the word "improvising" had some bad vibe. It contains a possibility of failure, uncertainty. Also, trying is low status, but teaching institutions can have high status, so maybe people expected an answer like: "I had an internship at Google and that's where I learned that".
(Age is context-dependent. If you are 25 in a job, and you are youngest of your colleagues, they see you as a child.)
which depends on the audience.
Yeah, probably that's it. While I'm positive that among musicians just having a decent sense of rhythm and melody and improvising on the E flat minor pentatonic scale (AKA “only playing the black keys”) is lower status than having spent hundreds of hours taking piano lessons and rehearsing, I'm not at all sure whether it'd also be lower status among other people, and indeed now that I think about it, my model of non-musicians says it wouldn't.
If I can take your answer literally,
I only normally use the word “improvising”...
As I've recently been understanding signalling/status behaviors common among humans and how they can cloud reality, I've had a tendency to automatically think of these behaviors as necessarily bad. But it seems to me that signalling behaviors are pretty much a lot of what we do during our waking life. If you or I have abstract goals: become better at physics, learn to play the guitar, become fit and so forth, these goals may fundamentally be derived from evolutionary drives and therefore their implementation in real life would probably make heavy use of signalling/status urges as primary motivators. But that does not necessarily reduce the usefulness of these behaviors in achieving these abstract goals1,2.
I suppose what we need to be cautious about are inefficiencies. Signalling/status behaviors may not be the optimal way to achieve these goals. We would have to weigh the costs of actively ignoring your previous motivators and cultivating new motivators against the benefit we would gain by having motivations more aligned to our abstract goals.
Any common examples of behaviors that assist and/or thwart goal-achievement? I've got one: health. Abstract goal: We want to be healthy and fit. Status/Signalling urge: desire to look good. The urge sometimes assists, as people try to exercise to look good, which makes you healthier. Sometimes it thwarts, like in the extreme example of anorexia. Has anybody made personal trade-offs?
Note:
1) I realize that this theme is underlying in many LW posts.
2) I'm not trying to talk about whether abstract goals are more important than signalling/status goals.