In my last post, I talked about how shifted burden-of-proof and Courtier's Replies allowed one person in a dialogue to generate arbitrary amounts of work for the other. I described this as looking like a "rules-lawyery, infinite-oregano concern". After making that post, I realised that a vanishingly small number of readers will understand that reference. I'd like to try and remedy that with this post, and perhaps elaborate on the idea a little further. Once I'm done, I may confusingly forward-reference this post from that one to confound future web-archiving software.
In 2008, Wired Online put up a short commentary piece on what cookbook reviews would be like if they were subject to the same criticism as RPG books. I used to be a role-player in a previous life, and this was so on-the-nose it got linked like wildfire ((I'm not sure wildfire is actually linked all that much)) by many of my old role-play buddies. By far, my favourite excerpt is the following:
Posted: 12:48 a.m. by Goku1440 I found an awesome loophole! On page 242 it says "Add oregano to taste!" It doesn't say how much oregano, or what sort of taste! You can add as much oregano as you want! I'm going to make my friends eat infinite oregano and they'll have to do it because the recipe says so!
This is an example of rules-lawyering: being more concerned with what the rules allow you to get away with than playing the game as intended. Goku1440 has interpreted the vagueness of the recipe as a vulnerability that is open to abuse. Feeding your dinner guests infinite oregano is absurd, and hence the scenario is funny ((Jokes become way better when you explain them, right?)). In the context of a recipe in a cookbook, and in real life in general, you would never be realistically concerned that someone would force you to eat infinite oregano.
This should hopefully convey what I mean by "infinite-oregano concern": a concern that a rule, policy, or convention (or lack thereof) might be open to abuse, even though such an abusive outcome is unrealistic. I described shifted burden-of-proof as looking like one because in an actual discussion you wouldn't blithely and meticulously evaluate every unsubstantiated claim your interlocutor made; you'd go and find someone else to talk to. You wouldn't just sit there and eat the infinite oregano they were trying to feed you.
When I found myself using the phrase "infinite-oregano concern", one very clear example of such a concern came to mind. Several months ago I put forward an argument against allowing people to select their own arbitrary set of personal pronouns ((I don't think too many people are actually for this specific proposal, but it was part of a broader discussion about pronoun usability)). I won't go into it in detail, but the gist is as follows: although personal pronouns may refer to a subject, it's generally not the subject who has to use them. It might seem reasonable that we refer to people how they wish to be referred, but it's possible for those wishes to comprise an unreasonable or unworkable expectation on the part of others.
Consider the person who changes their name to "(+)--(*)-(@@@)-(*)--(+)" ((I should have checked this at the time, but there are an existing set of restrictions on what you can change your name to by deed poll, and pronounceability is the second requirement on the list.)). If this is the name you wish other people to use when referring to you, it seems unlikely that many people would find this reasonable, and even fewer would find it reasonable if you expand this out to a whole set of declined parts of speech.
My discussion partner, though amenable to this argument, proposed that "reasonableness" was a sketchy basis for issuing policy about identity politics, and my concern only became a problem when taken to extremes, which are rare and identifiable on a case-by-case basis.
In other words, it's an infinite-oregano concern.
At first blush, it seems that raising infinite-oregano concerns is pedantic and unhelpful. No-one can force you to use their exotic pronouns that look like they were lifted from the Linux command line. Why express concern for an event that will never come to pass?
I think this is the wrong question to ask. The burden-of-proof convention doesn't safeguard us against our interlocutor forcing us to verify all their claims, but against a breakdown of dialogue. I'm not worried that someone issuing a bunch of Courtier's Replies will oblige me to read a stack of antiquated theological essays; I'm worried that someone issuing a bunch of Courtier's Replies will mean I'll have to stop talking to that person.
In the case of personal pronouns, names, and forms of address, the Infinite-Oregano Concern is slightly more fiddly. "(+)--(*)-(@@@)-(*)--(+)" is clearly unfit for purpose as a name, but the deed poll restrictions prohibit a wide selection of options that probably are fit for that purpose. No doubt some people would like these restrictions to be more stringent and others would like them to be less so, but the existing restrictions provide a common standard of expectations for what people might accept or exhibit as a name. This isn't a trivial concern. You can only obtain a bank account if you have a name the bank's software can validate, and many legal procedures and ceremonies require the utterance of one's name ((Fun fact: if you're deaf-mute in the UK, you require an interpreter to make your wedding vows.)).
In the absence of any similar clear set of restrictions for whatever pronouns people might like, it's hard to coordinate on a set of standard expectations, and without those expectations, it's easy to imagine how the occasional untenable request might pose a setback to the broader discussion. I won't go so far as to claim some imposed restriction would fix this, but I will tentatively speculate that it might help.
When I first started writing this, I became mildly concerned that I was just reinventing the slippery slope argument ((For a good take on the legitimacy of slippery slopes, I recommend Scott's LW post on the subject)), but I'm now reasonably convinced I've carved out more of a negative image of the slippery slope. A slippery slope argument would run "if we allow people to use any oregano at all, they'll end up forcing us to eat infinite oregano", whereas the infinite-oregano concern would go "we have to safeguard against the risk of being forced to eat infinite oregano in order for people to feel comfortable about having any oregano at all".
The concepts are parallel enough that I don't expect "infinite-oregano concern" to catch on, but I think it's a useful pattern to consider. The next time you find someone protesting against an outcome you know to be implausible, could there be an underlying fruitful negotiation that has to fall apart to prevent that outcome from happening?