1) Not only should the state be involved in determining what constitutes harassment and discriminator, the state ALREADY is responsible for doing that, and would be regardless of the bill in question. Furthermore, the notion that the state should not be involved is nonsense - the state is inextricably involved, and what you term as neutrality is actually just a decision in favor of one side over the other.
2) Given that the state is inextricably involved, we have no option but to consider the question of pronoun usage on its merits. We must either find complaints about pronoun usage reasonable, and move to restrict usage, or we must find them unreasonable, and move to protect usage (or in many cases continue to protect usage).
3) The notion that attempting to evaluate complaints on their reasonableness is fraught or otherwise untenable is contradicted by the fact that courts already are evaluating similar complaints on their reasonableness, and have been doing so for many years. The fact that there is ideological disagreement over pronoun use is irrelevant - there also exists ideological disagreement over many discriminator or harassment claims that courts have handled.
4) Attempting to handle complaints by nebulous notions of "intent" and "consequence" without evaluating the reasonableness of those complaints IS untenable. If we take any complaint seriously, no matter how nonsensical (attack helicopters, apples, etc.) then we will find ourselves in the nightmare scenario of empowering anyone to arbitrarily restrict others' speech.
Democracy is not a very specific term.
If with this case the definition isn't very agreeable, on what basis is to to be determined?
And would that hold up beyond opposition?
The courts determine what is and is not a reasonable complaint.
In order to determine whether a hostile work environment has been created, the judge must evaluate whether the work environment is hostile as a result of what has been done. What does 'the reasonableness of the complaint' even mean here, let alone be the priority?
I would think it's not important what gender identity truly is, being mistreated because of one's views on the matter is an act of discrimination.
If we take the court case I posted, in which the complaint was about hearing the word "n****r", and change it so that the complaint is about hearing the word "apple", is that still a hostile work environment? No, it's fundamentally unreasonable to be offended by the word apple in the same way as people are offended by the n-word. If someone showed up to court with the apple complaint, they'd be laughed out of the building. You can't conclude whether there is a hostile work environment without evaluating the reasonableness of the complaint.
the conduct is severe or pervasive enough to create a work environment that a reasonable person would consider intimidating, hostile, or abusive.
Anti-discrimination laws also prohibit harassment against individuals in retaliation for filing a discrimination charge, testifying, or participating in any way in an investigation, proceeding, or lawsuit under these laws; or opposing employment practices that they reasonably believe discriminate against individuals, in violation of these laws.
To be unlawful, the conduct must create a work environment that would be intimidating, hostile, or offensive to reasonable people.
The reasonableness of the complaint is absolutely critical here. The laws would be completely unworkable if they covered any conduct that anyone perceived to be abusive, regardless of the reasonableness of that perception.
No, I'm not saying there is no sensible way, I'm saying is there an agreeable one? I am saying why should the state be in practice of deciding what it means?
Agreeable to whom?
The state is already involved in deciding what is and is not legitimate, whether you like it or not. If Alice goes to court and says, "Bob called me 'he' and that made for a hostile work environment", the judge has to decide whether that's a legitimate complaint or not. You can argue that it's not a legitimate complaint, or you can argue that it is a legitimate complaint, but either way, the judge has to make that decision.
I didn't say you could have judges arbitrate what is reasonable here. The problem is- what is reasonable with this issue? Is it really the point to give the power of determining what gender identities and pronouns are valid to judges? At that point, why is it important what judges consider to be reasonable?
Why not have the standard that determines the difference between fair refusal to use gender pronouns and discrimination be based on what judges can determine about the intent, not what judges consider to be a reasonable gender identity? Murder isn't determined by whether or not judges think it was a reasonable act to kill the person, it's whether or not someone intentionally killed someone without being in self defence.
Judges can determine what racial epithets are unacceptable based on the relationship to intent as well. Religious belief can be based on societal institution mostly. A lot of other determinations about what is reasonable can be based simply on social convention.
There is always some kind of standard about how judges determine what is reasonable.
My argument is that objecting to the use of a preferred pronoun is reasonable as long as it's done without some kind of malice aforethought, like with murder. I don't think judges deciding what constitutes a valid gender identity is the right way to go about this.
Are you saying there is no way to formulate a standard for what is and is not a reasonable gender identity? You can't imagine any sensible way to delineate "attack helicopter" as unreasonable but "female" as reasonable?
The "I identify as an assault helicopter" argument could be an actual court case should preferred pronoun use be mandated- because who would decide what pronouns are at all valid? Unless the suggestion is to have judges decide what pronouns are valid (which I would think goes against the whole point), then forcing people to refer to others as attack helicopters would be exactly what would be happen so long as someone decides that attack helicopter is their identity. To suggest that most people would never be so unreasonable is paramount to special pleading. All it would take is one- and that's the point.
I think it would only be a matter of time before someone decided their preferred pronouns to be something absurd and unreasonable and is willing to it to take it to court. And the law would have them in the right.
You know what also be protected under such a law? Someone having their preferred pronoun being something that mismatches their identity. A cis male person could decide they want to be called she, because *****s and giggles, it's their preferred pronoun.
Judges already determine which words are unacceptable racial epithets. They determine which claims of religious belief are valid. A huge variety of laws work by having judges determine what is reasonable and what isn't. The notion that this is some sort of slippery slope betrays an incredible naivety about how the law currently works.
No, that seems to be sufficient to be some kind of discrimination. I don't think though that circumstance constitutes a particularly serious offence in my opinion. Nothing more serious than being fired perhaps. In order to be more serious, I think some kind of malicious intent would need to be there.
The most fundamental problem with comparing this word to the refusing to use certain gender pronouns is that it's an entirely optional affair to a specific word and one could live entirely consistently with their belief the word should not be taken to be offensive in at least some of the circumstances where it is without using it themselves. A person with such a belief could accommodate society's distaste with the word without having to be involved in promoting that distaste. With gender pronouns, there isn't really a say nothing option. There must be an affirmative decision to either accept a person's preferred pronoun or not to. In order to accommodate to it, one has to be actively involved in it's use.
I think there is also less grounds on which to object with the former than the later. For one simply because it's comparing one word to many. But also because of more implications that being a pronoun has versus a more optional word, other than being optional as I mentioned, related to the structure of language and such. And because not using a word doesn't have as strong of a force behind it than using specific words does- that the later is more grounded in certain motivations one could object to.
I'm not really sure I followed all of this, so I apologize if this reply misses the mark.
Suppose in our scenario Bob has a belief that n****r is the proper form of address for Alice. The use of the word captures some aspect of the way he views the world. It's not just that he disagrees with whether it's offensive, it's his way of expressing his worldview. The so-called say-nothing option is not acceptable to him. Does this excuse his behavior, or shield him from being fired?
Fundamentally though, it seems to me that what you're doing here is exactly what I contend is necessary - analyzing the merits of a request that someone speak a certain way, and balancing those merits against the merits of allowing that person to speak as they see fit. We can't simply wave our hands and declare ourselves "neutral" - we have to either choose to protect Alice's desires to be referred to in the way she chooses - be it to not be called n****r, to be referred to as "she" instead of "he", or to be called an attack helicopter, or we have to choose to protect Bob's desires to speak as he sees fit. In the case n****r, it seems that in at least some circumstances, you believe that Alice's desire trumps Bob's desire. In the case of the attack helicopter, I think we all agree that Bob's desire trumps Alice's. We have to perform a similar analysis for the case of "he" vs "she". We have to evaluate the severity of misgendering, we have to evaluate the importance free pronoun use, and decide whose side we're on.
It seems to me that Canada's laws tend to side with Alice too often. They extend far beyond the workplace, and limit what Bob can post on the internet in his private life. But we don't have to take an all-or-nothing approach. Just like it's reasonable to restrict Bob from calling Alice n****r at work, perhaps it's also reasonable to prevent him from intentionally misgendering Alice, regardless of whether Alice is transgender. On the other hand, maybe misgendering is too mild a slight to warrant protection, and instead the law should focus on more virulent anti-transgender slurs. These are the evaluations we need to make.
Now, when I said, it's about how it used, I meant that. So responding to that idea by presenting me with a circumstance where you don't tell me how it is used seems a meaningless question.
I can't judge the validity of the complaint if I don't how it was used, because I don't think usage in-and-of-itself is the issue.
Whenever Bob sees Alice in the morning, Bob says, "Good morning n****r" in a friendly tone. Alice has repeatedly requested Bob stop doing this. Bob has ignored these requests, and management has taken no action. Alice and Bob are otherwise cordial coworkers, but are not close friends. Bob is not attempting to make a joke, or otherwise be humourous. He is aware that Alice objects to the term.
Again, it's about how it is used. If the person using it is only using it in a friendly way amongst a close group of their friends, and the person complaining is just some person who observed the usage, then I don't think that's issue of discrimination. Even if we consider that the word is inescapably offensive, that doesn't mean it can't be used without offence- because we have humour, acting and talking about words themselves.
Okay, but specifically a case in which it is being used to address a particular person, and that person complains, and the user doesn't stop?
The comparison isn't completely fair. Either N****r or F****t used as a epithet can very reasonably get someone fired immediately; it's well-known that they are socially unacceptable. If one expressed that they didn't know that, they might have a chance, but even then it's not completely clear-cut, as establishing that concretely is difficult. Misgendering someone once is very different at this point. Multiple times even, unintentionally, is something else. Intent is also difficult to establish. Even once, clearly intentionally, still doesn't carry the same weight, though multiple times intentionally would be likely to get you fired. There may be a time in the future when any number of times intentionally misgendering someone will have the same gravitas as using N****r or F****t as an epithet, but I doubt it will be any time soon.
To be clear, I am absolutely not saying that calling someone n****r or f****t is the equivalent of misgendering. I just want to explore the claim that the government should never tell us how we are allowed to address each other. Since the n-word is a particularly extreme example, I think it's a useful example in that regard - not as a moral equivalent of misgendering.
We weren't talking about that example, we were talking about a racial issues comparison
Right, so if someone complained or took offense at your use of the word, you'd have to either stop, or the government would step in and stop you.
It seems to reference a broader context, in the relationship of the people involved
So it all seems like pretty clear that Mann wasn't just using the word n****r.
There are certainly other facts pertinent to the case, but the court notes, "Perhaps no single act can more quickly "alter the conditions of employment and create an abusive working environment," Meritor, 477 U.S. at 67, 106 S.Ct. at 2405, than the use of an unambiguously racial epithet such as "n****r" by a supervisor in the presence of his subordinates."
And also, it's not really important what the law is as to what the should be. Even if it's true that any use of n*****r by a nonwhite is considered discriminatory in the law, then all that changes is I have something else to disagree with because I don't think that's fair either. Uttering a single word- especially one that is perceived inconsistently with regards to offence- is such a trivial act it cannot be taken so dramatically as to constitute a serious offence in-and-of-itself.
Do you only object to the single-use case, or do you also object to the law stepping in when someone complains, and the person using the word refuses to stop?
You're changing the circumstance. Now you are talking about persisting beyond the point of complaints being made. The point is about usage no matter the circumstances.
That's always been the circumstance! Look back at when I first layed out my hypothetical.
Suppose one of my male coworkers is a bit on the effeminate side. I decide to exclusively refer to him using female pronouns, even after he corrects me and asks me to stop. I get fired for workplace harassment. Is that fair?
I said that I don't think it would always get you fired.
My point has always been that preferred pronoun use should not be generally mandated, not that there people should be protected against any consequences. Context and intent are essential
The case in question says that even a single use of the n-word is sufficient to establish a hostile work environment, even when the context and intent are explicitly unknown ("though Rodgers could not recall the precise context in which it was used.")
If Alice doesn't express that desire, whether verbally or through presentation, then Bob is blameless in using male pronouns to refer to a person apparently presenting as male. We all use male pronouns to refer to apparent men, all the time. And most of the time, those apparent men are in fact men, and have no desire to be referred to in any other way. Call me "he", and I'm certainly not going to object to it. So Alice's desire and her expression of that desire are critical components of this dispute. If Alice doesn't have the desire, there is no dispute. If Alice doesn't express the desire, there is no dispute. A state without the First Amendment could try to resolve the dispute by punishing Alice for having or expressing the desire, just as it could try to resolve it by punishing Bob for using the undesired pronouns. The First Amendment says that state isn't going to try for either resolution. Which is neutrality.
What you're describing is neutrality with respect to Alice and Bob - the government is not going to favor Alice over Bob or Bob over Alice. If the government comes down on the side of free pronoun use, then both Alice and Bob are free to use whatever pronouns they want. It is -not- neutrality with respect to free pronoun usage vs. selection of one's own pronouns. It's taking the side of free pronoun usage by following the principle of free expression. The fact that it also takes the side of free preference expression does not make it neutral, it's just another instance of non-neutrality, this time in favor of Alice's expression.
Beyond that, in practice, I think Bob cares a lot about whether Alice expresses that desire, and his objective probably is to "prevent Alice from asking me to say 'she' ", and then some. Transphobia is about hostility towards nonconformity. In the big picture, what Bob wants is for Alice to "act normal", given his own definition of "normal": he wants Alice to use "he", to change her name back to "Al", to wear pants, and so on. Ignoring that big picture, and focusing on the single narrow slice of the dispute that is Bob's pronoun usage, is going to be misleading. It's like looking at a single action in the Great Hypothetical Anglo-French War where the Britons are assaulting a French position and concluding that Switzerland's neutrality is pro-English because they're letting them do that one thing.
No, I think looking at a bigger picture only serves to muddy the waters. You're trying to invent a different scenario in which Bob has different desires to make one option seem more "neutral". The value of the simple scenario is that it lets us cut to the heart of what a neutral policy is. We have one very simple axis of dispute, where two parties want opposite and mutually exclusive things. The government then has to decide which of those things is more valuable - Bob's right to speak freely, or Alice's right to select the manner by which she is referred.
Again, so you say. I think it would depend, certainly with regards to whether you get actually fired. 'Unwelcome conduct' surely does not mean 'anything even that could be taken to be even slightly disparaging'.
Also again,
doing it respectfully would include having some awareness of how others feel about it.
I'm not suggesting it means "anything even that could be taken to be even slightly disparaging", and I'm not suggesting there can't be extenuating circumstances. But fundamentally, if I go to work and start calling my co-workers the n-word, and they complain about, I either have to stop doing it, or I'm going to get fired. If neither happens, the government is going to step in.
Perhaps it'd be helpful here to look at some case law:
First, Western-Southern argues that the district court committed clear error in finding that Mann's use of the word "n****r" contributed to a hostile work environment. In support of this argument, Western-Southern asks this court to consider the context in which that word was used. The record indicates that Mann used the word twice in Rodgers presence, though Rodgers could not recall the precise context in which it was used. The record also indicates that black employees, including Thomas and Rodgers himself, also used this word in the workplace. We find Western-Southern's argument unpersuasive. Perhaps no single act can more quickly "alter the conditions of employment and create an abusive working environment," Meritor, 477 U.S. at 67, 106 S.Ct. at 2405, than the use of an unambiguously racial epithet such as "n****r" by a supervisor in the presence of his subordinates. See Bailey v. Binyon, 583 F.Supp. 923, 927 (N.D.Ill.1984) ("The use of the word 'n****r' automatically separates the person addressed from every non-black person; this is discrimination per se."). The fact that black employees also may have spoken the term "n****r" does not mitigate the harm caused by Mann's use of that epithet; a supervisor's use of the term impacts the work environment far more severely than use by co-equals.
Isn't that a case of the government telling me how I can address other people? And further a case where the people I'm addressing have a say in how I'm allowed to address them? If they didn't complain, or didn't feel it was offensive or hostile, then there wouldn't be an issue. Isn't that in conflict with what how you've said about things should work?
If Bob wants to call Alice "he" and Alice wants to be called "she", the only speech that's relevant to that outcome is Bob's speech.
...plus Alice's desire to be called that. If Alice didn't mind being called "he", there would be no problem. This is a question of her conscience, which is a First Amendment issue per se. And for this situation to become a dispute, then Alice needs to express that desire, which is speech in the ordinary sense of the term.
Sure, but Bob doesn't care whether Alice expresses that desire. Bob's objective here is not "prevent Alice from asking me to say "she"". The only speech which is in dispute is Bob's.
Anyone mind giving the layman's cribnotes on identity? An example, in particular, would be useful. Here it seems to be separated from both (biological) sex and (cultural) expression and so I'm not sure what's left. I do have an example below but that is more mental illness than anything IMO.
So I could be biologically a male, dress and act in a feminine way and 'identify' as some other thing? What would be my impetus for doing so? Can I wake up one day and just 'identify' as a Lord Admiral and make up some definition of that and then insist that everyone refer to me as a Lord Admiral under penalty of law? By no means am I suggesting that such a silly example represents all or even most of those who have another identity that they hold deeply to, but I suppose that's the question: what is a legitimate identity that needs to be recognized, what's not and how do you determine the difference?
When I was in my teens, there was this guy that showed up to play MTG with us one night. I'm not sure where he came from or who invited him but this guy was, as far as I could tell, genuinely convinced that he was a Werewolf. I mean genuinely convinced, not joking around or trolling or anything. Being the immature jerks that we were at the time, we really let him have it behind his back. I'm not proud of that because clearly he had a mental illness and probably wasn't responsible for his weird behavior. Again, I recognize that this is a wacky example but I'm curious as to whether this law would have actually required us to acknowledge his Werewolf-ness. It was definitely part of who he thought he was as a being, both physically and culturally/socially.
Now I'm not sure about, in that above example, what the correct attitude is in dealing with such people. Maybe it's best to just pretend that they are a Werewolf and just go along to get along. But I could also see the point of view of someone who suggested not feeding into his delusions. And, either way, it certainly seems like an area around which there is legitimate room for disagreement and discussion (i.e., not a question for the law.)
Basically, after a botched circumcision, he was left without a *****. The doctors at the time suggested simply doing a gender-reassignment surgery, and raising him as a girl. Despite the surgery and being raised as female, Reimer ended up identifying as male and transitioned to living as a man. If there were no internal gender identity (as the doctors who performed the experiment believed), then this wouldn't really make sense. Gender identity is the thing that internally tells you what gender you are. Most of us don't even notice it, because we naturally feel as though we are the gender that everyone perceives us to be, and so it can be difficult to imagine what it would feel like to have a mismatch.
Private Mod Note
():
Rollback Post to RevisionRollBack
To post a comment, please login or register a new account.
1) Not only should the state be involved in determining what constitutes harassment and discriminator, the state ALREADY is responsible for doing that, and would be regardless of the bill in question. Furthermore, the notion that the state should not be involved is nonsense - the state is inextricably involved, and what you term as neutrality is actually just a decision in favor of one side over the other.
2) Given that the state is inextricably involved, we have no option but to consider the question of pronoun usage on its merits. We must either find complaints about pronoun usage reasonable, and move to restrict usage, or we must find them unreasonable, and move to protect usage (or in many cases continue to protect usage).
3) The notion that attempting to evaluate complaints on their reasonableness is fraught or otherwise untenable is contradicted by the fact that courts already are evaluating similar complaints on their reasonableness, and have been doing so for many years. The fact that there is ideological disagreement over pronoun use is irrelevant - there also exists ideological disagreement over many discriminator or harassment claims that courts have handled.
4) Attempting to handle complaints by nebulous notions of "intent" and "consequence" without evaluating the reasonableness of those complaints IS untenable. If we take any complaint seriously, no matter how nonsensical (attack helicopters, apples, etc.) then we will find ourselves in the nightmare scenario of empowering anyone to arbitrarily restrict others' speech.
The courts determine what is and is not a reasonable complaint.
If we take the court case I posted, in which the complaint was about hearing the word "n****r", and change it so that the complaint is about hearing the word "apple", is that still a hostile work environment? No, it's fundamentally unreasonable to be offended by the word apple in the same way as people are offended by the n-word. If someone showed up to court with the apple complaint, they'd be laughed out of the building. You can't conclude whether there is a hostile work environment without evaluating the reasonableness of the complaint.
https://www1.eeoc.gov//laws/types/harassment.cfm?renderforprint=1
The reasonableness of the complaint is absolutely critical here. The laws would be completely unworkable if they covered any conduct that anyone perceived to be abusive, regardless of the reasonableness of that perception.
That's not what a democracy means.
In order to determine whether a hostile work environment has been created, the judge must evaluate the reasonableness of the complaint.
Agreeable to whom?
The state is already involved in deciding what is and is not legitimate, whether you like it or not. If Alice goes to court and says, "Bob called me 'he' and that made for a hostile work environment", the judge has to decide whether that's a legitimate complaint or not. You can argue that it's not a legitimate complaint, or you can argue that it is a legitimate complaint, but either way, the judge has to make that decision.
Are you saying there is no way to formulate a standard for what is and is not a reasonable gender identity? You can't imagine any sensible way to delineate "attack helicopter" as unreasonable but "female" as reasonable?
Judges already determine which words are unacceptable racial epithets. They determine which claims of religious belief are valid. A huge variety of laws work by having judges determine what is reasonable and what isn't. The notion that this is some sort of slippery slope betrays an incredible naivety about how the law currently works.
I'm not really sure I followed all of this, so I apologize if this reply misses the mark.
Suppose in our scenario Bob has a belief that n****r is the proper form of address for Alice. The use of the word captures some aspect of the way he views the world. It's not just that he disagrees with whether it's offensive, it's his way of expressing his worldview. The so-called say-nothing option is not acceptable to him. Does this excuse his behavior, or shield him from being fired?
Fundamentally though, it seems to me that what you're doing here is exactly what I contend is necessary - analyzing the merits of a request that someone speak a certain way, and balancing those merits against the merits of allowing that person to speak as they see fit. We can't simply wave our hands and declare ourselves "neutral" - we have to either choose to protect Alice's desires to be referred to in the way she chooses - be it to not be called n****r, to be referred to as "she" instead of "he", or to be called an attack helicopter, or we have to choose to protect Bob's desires to speak as he sees fit. In the case n****r, it seems that in at least some circumstances, you believe that Alice's desire trumps Bob's desire. In the case of the attack helicopter, I think we all agree that Bob's desire trumps Alice's. We have to perform a similar analysis for the case of "he" vs "she". We have to evaluate the severity of misgendering, we have to evaluate the importance free pronoun use, and decide whose side we're on.
It seems to me that Canada's laws tend to side with Alice too often. They extend far beyond the workplace, and limit what Bob can post on the internet in his private life. But we don't have to take an all-or-nothing approach. Just like it's reasonable to restrict Bob from calling Alice n****r at work, perhaps it's also reasonable to prevent him from intentionally misgendering Alice, regardless of whether Alice is transgender. On the other hand, maybe misgendering is too mild a slight to warrant protection, and instead the law should focus on more virulent anti-transgender slurs. These are the evaluations we need to make.
Whenever Bob sees Alice in the morning, Bob says, "Good morning n****r" in a friendly tone. Alice has repeatedly requested Bob stop doing this. Bob has ignored these requests, and management has taken no action. Alice and Bob are otherwise cordial coworkers, but are not close friends. Bob is not attempting to make a joke, or otherwise be humourous. He is aware that Alice objects to the term.
Are there any other details you would need?
Okay, but specifically a case in which it is being used to address a particular person, and that person complains, and the user doesn't stop?
To be clear, I am absolutely not saying that calling someone n****r or f****t is the equivalent of misgendering. I just want to explore the claim that the government should never tell us how we are allowed to address each other. Since the n-word is a particularly extreme example, I think it's a useful example in that regard - not as a moral equivalent of misgendering.
There are certainly other facts pertinent to the case, but the court notes, "Perhaps no single act can more quickly "alter the conditions of employment and create an abusive working environment," Meritor, 477 U.S. at 67, 106 S.Ct. at 2405, than the use of an unambiguously racial epithet such as "n****r" by a supervisor in the presence of his subordinates."
Do you only object to the single-use case, or do you also object to the law stepping in when someone complains, and the person using the word refuses to stop?
That's always been the circumstance! Look back at when I first layed out my hypothetical.
The case in question says that even a single use of the n-word is sufficient to establish a hostile work environment, even when the context and intent are explicitly unknown ("though Rodgers could not recall the precise context in which it was used.")
What you're describing is neutrality with respect to Alice and Bob - the government is not going to favor Alice over Bob or Bob over Alice. If the government comes down on the side of free pronoun use, then both Alice and Bob are free to use whatever pronouns they want. It is -not- neutrality with respect to free pronoun usage vs. selection of one's own pronouns. It's taking the side of free pronoun usage by following the principle of free expression. The fact that it also takes the side of free preference expression does not make it neutral, it's just another instance of non-neutrality, this time in favor of Alice's expression.
No, I think looking at a bigger picture only serves to muddy the waters. You're trying to invent a different scenario in which Bob has different desires to make one option seem more "neutral". The value of the simple scenario is that it lets us cut to the heart of what a neutral policy is. We have one very simple axis of dispute, where two parties want opposite and mutually exclusive things. The government then has to decide which of those things is more valuable - Bob's right to speak freely, or Alice's right to select the manner by which she is referred.
I'm not suggesting it means "anything even that could be taken to be even slightly disparaging", and I'm not suggesting there can't be extenuating circumstances. But fundamentally, if I go to work and start calling my co-workers the n-word, and they complain about, I either have to stop doing it, or I'm going to get fired. If neither happens, the government is going to step in.
Perhaps it'd be helpful here to look at some case law:
http://openjurist.org/12/f3d/668/rodgers-v-western-southern-life-insurance-company
Isn't that a case of the government telling me how I can address other people? And further a case where the people I'm addressing have a say in how I'm allowed to address them? If they didn't complain, or didn't feel it was offensive or hostile, then there wouldn't be an issue. Isn't that in conflict with what how you've said about things should work?
Sure, but Bob doesn't care whether Alice expresses that desire. Bob's objective here is not "prevent Alice from asking me to say "she"". The only speech which is in dispute is Bob's.
I think a good way to understand what gender identity is is the case of David Reimer. https://en.wikipedia.org/wiki/David_Reimer
Basically, after a botched circumcision, he was left without a *****. The doctors at the time suggested simply doing a gender-reassignment surgery, and raising him as a girl. Despite the surgery and being raised as female, Reimer ended up identifying as male and transitioned to living as a man. If there were no internal gender identity (as the doctors who performed the experiment believed), then this wouldn't really make sense. Gender identity is the thing that internally tells you what gender you are. Most of us don't even notice it, because we naturally feel as though we are the gender that everyone perceives us to be, and so it can be difficult to imagine what it would feel like to have a mismatch.