Or maybe, Bob does not have property X, and Alice is mistaken, tendentious, projecting her own stuff onto Bob, gaslighting, straightforwardly lying, or something else. All of these and many other reasonable possibilities are consistent with the imaginary scenario as presented: "Alice tells Bob, "You have property X"."
Personally, whenever I hear "Alice" say "X" about "Bob", I never think, "oh, so Bob is X!" That Bob is X is only one of all of these possibilities about what is going on, and does not even dominate the probability space.
Bob thinks that if he admits that he has property X, then that will render him [I understand "him" to indicate the totality of his traits] (more) invisible to Alice (or onlookers). By admitting that he is X, Bob adds support in the common-knowledge context to the image of him as X. If there's common-knowledge that Bob is X, then Alice will: be emboldened to model Bob as X; treat Bob as X; expect Bob to expect Alice to treat him as X; ignore, or perhaps punish or resist, behavior by Bob not in line with X.
All of this is true not only of defensiveness but of all interactions. All exchanges of meaning, including non-verbal, carry potential updates to have some property X, Y, or Z. It is just more apparent with defensiveness.
I think this varies greatly by situation and with the specifics of X, how it's seen in that group, and how well it actually describes Bob's behaviors as seen by Alice. I don't see much generality here, and I think there are a lot of Xs that can't be usefully discussed on LW, so I'm not sure how to dig deeper.
I am curious about your use of "invisible" for what I'd have termed "low-status". Is this specific to some Xs, or just a different way of referring to the same thing?
They're completely different. Invisible means: can't be seen. X doesn't have to be good, bad, low/high status. It's any property.
Hmm. I think I'm not in the right social group to understand this. I don't share a fear of invisibility, and I don't quite see how admitting X can make one more invisible than just exhibiting behaviors of X. But again, I also suspect it's idiosyncratic to the groups in which Bob and Alice participate, and the X in question.
I wish you the best! Thanks for posting, even if it didn't quite resonate with me.
Sometimes people get defensive. That is, Alice tells Bob, "You have property X", and then Bob tenses his muscles, moves rigidly, pretends to not be bothered, and denies that he has property X, or changes the subject, or asks Alice why she thinks he's X (perhaps in a tenser / higher-pitched tone of voice) and then gives an alternate explanation of observations Alice describes. We can take it for granted that Bob's behavior is often because of something in the ballpark of, he's trying to maintain a Public Relations story to Alice or himself or onlookers (present or not); or, he's trying to avoid a bucket error.
But, given that there's stigma about defensiveness, we can expect that there are other reasons for Bob to deny that he has property X, and that those reasons will be delegitimized.
Here's one such reason: Bob thinks that if he admits that he has property X, then that will render him (more) invisible to Alice (or onlookers). By admitting that he is X, Bob adds support in the common-knowledge context to the image of him as X. If there's common-knowledge that Bob is X, then Alice will: be emboldened to model Bob as X; treat Bob as X; expect Bob to expect Alice to treat him as X; ignore, or perhaps punish or resist, behavior by Bob not in line with X. Whatever else Bob is, other than X, is implicitly marginalized; so most of Bob is invisible, which could make Bob want to scream (or deny X, or without denying X force Alice via insincere questions to expand on the proposition and add nuance until there's room for Bob to be visible).
From the inside this is like: "Oh so this is how they're implementing their decision not to listen to what I'm saying." Or: "Oh so this is how the possibilities of our interaction are collapsed away." Or: "Oh so this is how they're going to pigeonhole me so that they already know everything there is to know about interacting with me."
This is an instance of a more general problem: lopsided common-knowledge creation. A group may be prone to mistakenly act as though it is overconfident in only those beliefs which are common knowledge. For such a group, creating common knowledge of X may lead to mistaken behavior even if X is true. Maybe it helps to accompany common knowledge of X with common knowledge of other facts that might recommend against whatever mistaken behavior X alone could recommend. (I would guess that fear of lopsided common knowledge is behind a fair amount of information suppression motivated by "political" intuitions. Think of the "high decoupler" saying something like "X people are Y" and then the "high contextualizer" saying "Um um yeah but Z people are W" which is logically irrelevant but relevant as an accompanying piece of common knowledge that forces a policy of "let's not hurt groups of people (because Xs are Y and Zs are W and so on, so if we hurt people with bad properties we'd hurt everyone which is stupid)".)