Other business: Steven Moffat's reboot of Sherlock Holmes sounds quite promising. Front Row liked it.

Mopping up:

Look, here's the rule of thumb I personally propose to work by when it comes to single pronouns, if that helps make my stance any easier to understand

It does, thanks. You see I find it rather strange to think of using gendered and non-gendered pronouns in the same conversation. I'd be all, 'Wait, are we rejecting the binary model in this conversation or not?' Using genderless pronouns only with respect to one person in a group would feel bizarrely othering.

With respect to discarding tools, I would argue that tools absolutely should be discarded if their use is actually, in a small way, causing harm.

Agreed, and I think I tried to say that at some earlier point. But of course it applies only rather indirectly to 'they' because the use of 'they' as usually plural isn't harming anyone. Before you can use this reasoning to support the use of 'they' as singular you have to establish not only that 'he' and 'she' and 'he or she' are causing harm but also that singular 'they' is the only viable alternative, which is QED and ND (nóndum démónstrátum = not yet proven).

Except it's not "other ambiguities" it's "the same ambiguity".

Basically what Shim said. Partly my fault because I used 'other ambiguities' in an unclear way. I think what I'm calling an ambiguity you're calling an instance of ambiguity, and what you're calling an ambiguity I would call a type of ambiguity. But it doesn't matter, so I'll adopt your terminology. The point is that some instances can be avoided by either of two different methods (rephrasing or use of different pronouns) and some can be avoided using only one method (rephrasing) because the other doesn't help (because the appropriate pronouns are the same anyway). The fact that one method works in both instances is not a good reason for causing the other method not to work in either instance. Nor does the fact that rephrasing works in both instances mean that rephrasing is the best method in both instances, which is what you're saying when you say both problems are caused by bad writing. It isn't bad writing to write 'The patient waited, trying not to let himself think about what he would say when they came back.' The appropriate pronouns happen to be different, and therefore using pronouns is a perfectly proper way to write the sentence. 'The patients waited, trying not to let themselves think about what they would say when they came back' is bad writing, because the appropriate pronouns are the same and therefore the use of pronouns is not a valid method for avoiding this particular instance of ambiguity. But that doesn't mean it wasn't a perfectly appropriate method for a perfectly good writer to use in the previous instance. Some tools don't work every time, but that isn't a good reason to throw them away. The distinction between singular and plural pronouns is a tool that we have. It works in some instances, such as the former of the two examples above. The fact that it doesn't work in other instances, such as the latter, is not a good reason to deny that it's a useful tool worth keeping.

What have I done? What have I done?

Calm down, dear: it's only a commercial disagreement about words.

In the cold light of day, I suppose you're right. Metaprocrastination is presumably faffing about instead of doing the faffing you were going to do instead of work.

It's all procrastination. Unless you're hell-bent on using up your prefix ration. ;)

Completely tangential: I went on the computer to do some work, and accidentally came to Ferretbrain, where I procrastinated. Then I went and talked to my housemate about procrastinating on FB. Now I'm writing a post about telling my housemates about procrastinating on FB, which must surely be metametaprocrastination?

In a similar vein, some languages distinguish relatives by maternal/paternal, which English doesn't. Losing this distinction would not be disastrous, but would make the language more ambiguous.

German makes it harder to distinguish "girlfriend"/"boyfriend" from "friend". American English does this with "girlfriend" sometimes. Welsh distinguishes "lover" from "friend", but doesn't indicate its gender and also uses it as a term of affection, leading to different ambiguities. Manx has gendered words for "girlfriend"/"boyfriend" and non-gendered "friend". Clearly, you don't need any specific set of these distinctions, but each one lost means more ambiguous language. Going back to the earlier discussion, Dan's examples are all talking about ambiguity over which friend is meant, whereas Jamie is highlighting the loss of distinction between "lover" and "friend".

Except it's not "other ambiguities" it's "the same ambiguity".

I think it's only the same from one point of view, as in, it's an ambiguity about who is who. I believe what Jamie's saying is roughly:
* "they" usually means the third person plural, just as "she" and "he" are feminine and masculine 3S respectively.
* using "they" as a neutral 3S creates the possibility of an ambiguity that previously did not exist (it does not exist, for example, if we use "it" as the neutral pronoun).
* the fact that it is possible to envisage situations where "she", "he" or "they" are ambiguous is irrelevant, because the problem being raised is [the creation of a new possible ambiguity that makes language more ambiguous overall], not [that pronouns are ambiguous]

Dan's examples are of situations where a pronoun that always refers to a specific group (she, he, they) is ambiguous where more than one example of that group exists. The same is true if we are discussing a number of cats, or boxes, or anything.

"The box is on the table, next to the cat and the box. When you go in, put the box on top of the box."
(Yes, it's a rubbish example, I know)
Anything that has more than one potential referent becomes ambiguous. However, if (for whatever reason) we decide to refer to cats as "boxes", we can point to at least one situation where it creates new ambiguity:
"The box is on the table, next to the box and the box. When you go in, put the box on top of the box."

What have I done? What have I done?

The fact that other ambiguities sometimes occur is irrelevant. The two variants of my sentence that you offer, Dan, only show that English lacks certain tools. It has only one singular third-person feminine pronoun and only one plural third-person pronoun. Therefore it cannot use pronouns to resolve the ambiguities in those sentences, and they have to be resolved by more extensive re-writing.

Except it's not "other ambiguities" it's "the same ambiguity".

The ambiguity occurs because it is always possible for two things to be legitimately describable by the same pronoun. This problem cannot be avoided by the use of gendered pronouns, or plural-identifying pronouns, or racially-coded pronouns, or different pronouns for tall and short people. The SAME PROBLEM arises.

Trying to solve the problem of "two things legitimately being described by the same pronoun" by creating arbitrary categories of pronouns to distinguish them is unhelpful, because it only solves the specific case where a singular individual is interacting with a plural group. In the same way, gendered pronouns do not help resolve ambiguity except in the situation where you have exactly one man talking to exactly one woman. In each case the ambiguity arises from bad sentence structure.

I'm not sure that has a great deal of weight if it's a conversation in which everyone is being referred to using the same set of genderless pronouns

When would you do that though?

Look, here's the rule of thumb I personally propose to work by when it comes to single pronouns, if that helps make my stance any easier to understand:

Subject is being talked about in the abstract rather than being an actual person, gender not relevant to discussion: "They".

Subject is being talked about in the abstract rather than being an actual person, gender is relevant: "He" or "she" as appropriate for the hypothetical example being talked about.

Subject is an actual human being who identifies with a particular gender: "He" or "she", whichever they're identifying with.

Subject is an actual human being who does not identify with a particular gender, has expressed preference with respect to pronouns: Use the pronouns they prefer.

Subject is an actual human being who does not identify with a particular gender, has not expressed preference wrt pronouns, would be socially awkward to ask which pronoun they prefer: "They".

Subject is an actual human being whose gender identity is being kept mysterious for a particular reason, perhaps because they are a character in a story and a plot twist needs to be kept under wraps: "They".

Obviously that's not going to be a universal rule, no language issue ever is, but it's a million light years away from the "six different people and one transperson" example and heading rapidly in the opposite direction.

With respect to discarding tools, I would argue that tools absolutely should be discarded if their use is actually, in a small way, causing harm. Using "he" as a gender-neutral term is arguably just such a tool; you could argue that "he or she" is similarly troubling because it reinforces the gender binary (and it puts he first, as though "he" were the default and "she" were the alternative).

Just to throw something into the ring: German actually gets by pretty well with the same pronoun for you, she and they (sie). The difference is usually shown up in writing by capital 'S' for you and a different verb-ending for 'she' - but in speech and at the start of sentences, Sie is ambiguous between they and you. However, they seem to get by just fine.

How about (3): the person you are dealing with is a transperson who has gone through a ton of shit to be generally recognised by society as the gender they identify with, and they want people to respect that by using appropriately gendered language to refer to them?

I'm not sure that has a great deal of weight if it's a conversation in which everyone is being referred to using the same set of genderless pronouns. It would be bizarre if in a group of half a dozen people you were using an invented genderless pronoun for everyone except the trans person, and I'm not sure that the person concerned would find that helpful. The same with 'they': would you actually go out of your way to say 'he' in referring to a trans man in a sentence in which you would say 'they' if he were a cis man? Maybe; it's complex, as Dan said, and I don't want to deny that it's complex, but I'm not sure that 'I want to be referred to in a gendered way', even if it has weight in particular cases, is fully comparable to 'I want to be referred to neutrally' because it isn't a case of 'I want to be categorized as X' versus 'I want to be categorized as Y', it's a case of 'I want to be categorized as X' versus 'I don't want to be categorized'.

But this reminds me that I haven't answered this, sorry:

You seem to keep sliding into arguing against the sweeping away of all gendered pronouns when nobody's actually advocating putting "he" or "she" in the naughty box, just being conscious as to when it is and isn't appropriate to use them.

Well, I think some people are advocating that. Not here, but in the world at large there are people saying that's the way to go. And when I appear to be engaging with that position, it's because I'm trying to explore the consequences of it. Because it may be that the answer to 'are invented genderless pronouns helpful' is one thing if what you want to do with them is merely to accommodate particular people who don't want to be gendered but another thing if what you want is to stop language routinely gendering people.

... that's just plain sloppy writing

How is that more ambiguous than..?

Chaps, I'm afraid you've missed my point, which may be my fault. The point is this:

Ambiguity can occur in language. One avoids it by using the tools the language provides. It's undesirable to discard those tools. The distinction between singular and plural pronouns is a useful tool in English. It's undesirable to discard it because it deprives us of one of the various ways of avoiding ambiguity.

The fact that other ambiguities sometimes occur is irrelevant. The two variants of my sentence that you offer, Dan, only show that English lacks certain tools. It has only one singular third-person feminine pronoun and only one plural third-person pronoun. Therefore it cannot use pronouns to resolve the ambiguities in those sentences, and they have to be resolved by more extensive re-writing. That is not an argument for failing to use tools that English has got. English has a tool for distinguishing singular from plural, and it is the fact that 'they' is generally plural. The fact that we have no tool for distinguishing between 'she (whom we mentioned just now)' and 'she (whom we mentioned somewhat earlier, before the last woman we mentioned)' is not a justification for not using the tool that we have got for distinguishing between 'she (who is one person)' and 'they (who are several)'. It's an invalid argument. It's like saying 'we haven't got any mangoes so we don't need apples'. If we had mangoes, I'd eat the mangoes, but we haven't. We have got apples, and I don't see why the lack of mangoes should stop me eating the apples.

I can't off the top of my head think of any reason why anyone would object to being referred to by a genderless pronoun except (1) someone who objects to genderless pronouns, or the particular one being used, in principle, which may be fair enough but is nothing to do with respecting that particular person's identity, or (2) someone who for reasons of prejudice doesn't want to be referred to using language that's associated with genderqueer people, which is not a reason I'd be inclined to respect all that much.

How about (3): the person you are dealing with is a transperson who has gone through a ton of shit to be generally recognised by society as the gender they identify with, and they want people to respect that by using appropriately gendered language to refer to them?

Genderless pronouns:

I think "popular" is a very relative term here.

Yes, certainly. I was use it as relative to the group of people who use invented pronouns of any kind. Sorry if that wasn't clear.

I've partly responded to your wider point below, at least as regards 'they', but I'll say here that I completely agree that the multiplicity of different sets of invented pronouns is unhelpful, and that more fundamentally the use of invented pronouns does drastically obstruct communication when not all the people involved in the conversation are familiar with them. It's a bit of a catch-22, because what it boils down to is that they can't be easily understood unless they're familiar and they can't become familiar because they aren't easily understood. It may be that 'they' is the least bad solution, despite the problems with it, or it may be that the problem is too big and pronouns are just going to have to stay gendered, or it may be that the circle is sufficiently vicious that it has to be broken in a way that will initially cause as many problems as it promises to solve in the long run. Like I say, I don't know what I think, which is why I'm investigating.

Although of course one also needs to respect the identities of people who *do* want to be gendered by language.

I'm not sure that one does, at least to the same extent. Nobody complains about the fact that there aren't plural pronouns that can distinguish all-male groups from all-female groups from mixed groups. No speakers of Finnish, so far as I'm aware, feel oppressed or incorrectly classified as a result of the fact that their language has no gendered pronouns. I can't off the top of my head think of any reason why anyone would object to being referred to by a genderless pronoun except (1) someone who objects to genderless pronouns, or the particular one being used, in principle, which may be fair enough but is nothing to do with respecting that particular person's identity, or (2) someone who for reasons of prejudice doesn't want to be referred to using language that's associated with genderqueer people, which is not a reason I'd be inclined to respect all that much. But perhaps there's something I haven't spotted here. It sometimes happens!

'The doctors said they would be back at two. The patient waited, trying not to let themself think about what they would say when they came back. They weren't even sure whether they understood what they'd already said to them.'

But if you are using "they" - or, really, any word - that frequently then that's just plain sloppy writing. In particular, if you changed even one of the theys in the last sentence to "the doctors" or "the patient" you'd lose the ambiguity.

Which is pretty much key to me. I can't think of any situation in which there was a confusion between multiple "theys" where you couldn't just happily substitute in the person's name, or the person's job, or the person's current status (like "the patient"), or indeed any other non-pronoun indicator in order to provide extra clarification if you feel it's required.

Of course this argument can also be used against genderless pronouns, or at least against their use in stead of gendered ones, because the distinction between 'he' and 'she' is also helpful in composing passages involving several individuals of different genders.

Why would you be using a genderless pronoun to refer to a specific individual who's identifiably part of the gender in question (and has no objection to being identified as such)? You seem to keep sliding into arguing against the sweeping away of all gendered pronouns when nobody's actually advocating putting "he" or "she" in the naughty box, just being conscious as to when it is and isn't appropriate to use them.

How is that more ambiguous than:

"The doctor said she would be back at two. The patient waited, trying not to let herself think about what she would say when she came back. She wasn't even sure she understood what she'd already said to her."

Or for that matter

"The doctors said they would be back at two. The patients waited trying not to let themselves think about what they would say when they came back. They weren't even sure they understood what they'd already said to them."

The ambiguity over singularity or plurality is a red herring. What's ambiguous in this case is the fact that the same pronoun can be used to describe several different parties in an interaction. In every case it is perfectly clear what is plural (the doctors in example one, the doctors and the patients in example three) and what is singular (the patient in example one, the doctor and patient in example two).

I'd add that Wolf Hall provides a tremendous example of how this ambiguity has *nothing whatsoever* to do with the singular "they", because the pronoun which causes all the trouble in Wolf Hall is "he".

Allowing for a singular "they" does mean that you can't be certain without extra context whether the term "they" refers to one person or more than one person, but that in itself is unimportant.

Of course you could illustrate the same point by going back to the racially coded pronouns from earlier. The author of the Person Paper might legitimately say:

If we alter the structure of our language to remove "racist" pronouns, this will only lead to ambiguities as you will be unable to tell whether you are talking about a black or a white. Instead of simple constructions like:

"The doctor said whe would be back at two. The patient waited, trying not to let blerself think about what ble would say say when whe came back."

We will have monstrosities such as

"The doctor said he would be back at two. The patient waited, trying no to let himself think about what he would say when he came back."

Preposterous!

Grammar in general:

Dan, I accept what you say about 'the bits of grammar that are important' in respect of communicating comprehensibly. Every native speaker knows enough grammar to understand and be understood, and therefore the requirement of effective communication is satisfied, even if only at a sort of subsistence level. What I'm talking about is making language as expressive as possible and avoiding unnecessary loss of expressive power. The subjunctive mood in English, for example, is a very powerful and helpful tool that enables the speaker or writer to easily make useful distinctions between actual and hypothetical situations and suchlike. People who don't really know how it works demonstrably feel the need for some way of expressing the things that it expresses because they use unorthodox constructions like 'If I would have been there'. The latter aren't going to do anyone any harm, but it's perfectly reasonable and indeed helpful to observe that there's actually already a well-established way of doing that in English, and suggesting that it would be rather nice if it were more widely known.

They:

Dan, the problems with 'they' tend to arise over extended passages dealing with both individuals and groups, and in those cases it the difficulty for the reader or listener isn't so much not knowing whether the 'they' is singular or plural as not knowing whether it refers to the individual or the group. For example:

'The doctors said they would be back at two. The patient waited, trying not to let themself think about what they would say when they came back. They weren't even sure whether they understood what they'd already said to them.'

It's the same trouble as we had in Wolf Hall with 'he' in passages involving more than one man. And of course there are fairly easy ways to work around it, just as there are easy ways to avoid having trouble with 'he' in passages involving more than one man, the classic one being to use the person's name.* It would be fairly easy to re-write the passage above to be more comprehensible. So it obviously isn't an absolute road-block to the singular use of 'they'. Nonetheless, it seems to me undesirable to deliberately make language even harder to work with by using the same pronoun for both singular and plural. Because by far the easiest way to re-write the passage above to make it comprehensible is to replace every singular 'they' with a singular pronoun that's different from the plural pronoun. Arthur approves of pedantry being used to expand the range of things that language can do and that we can do with language, and I agree, and by the same token I find it difficult to be enthusiastic about things that restrict the range of things that language can do and that we can do with language.

Of course this argument can also be used against genderless pronouns, or at least against their use in stead of gendered ones, because the distinction between 'he' and 'she' is also helpful in composing passages involving several individuals of different genders. That's one reason I'm agnostic on the subject so far. But proponents would say that overcoming social injustice is worth that sort of self-restriction, and I can see that that's quite a powerful argument, if one can establish that genderless pronouns are necessary or significantly helpful in overcoming social injustice. Losing the distinction between singular and plural pronouns, on the other hand, doesn't appear to have any such justification, except indirectly, which brings me to what you said, Arthur:

Which is another argument for the use of "they", in some respects: if clear communication and the expressive capacity of the language is important, then arguably it's better to use an established and universally-understood word which more or less works in context than to use an invented word which might be perfect for the context you want but just ends up confusing people.

That's a compelling argument. The main reservation I have about it is that, as I've said, I'm not sure that 'they' can work in every case where a genderless singular pronoun is wanted, and that leaves the possibility (if we accept the need for genderless singular prounouns at all) that it might be necessary to use invented pronouns as well. In which case there's little advantage to be gained from not just using them all the time and leaving 'they' to be usually plural. But there are a lot of hypotheticals involved, and it's hard to tell.


* (As a point of little relevance but perhaps some interest, Latin deals with this very well by having four different sets of pronouns, one fairly neutral one for 'the person we all know we're talking about and there's no need to distinguish further', one for 'this person who is either physically here with us or is the person we've all been talking about just now and is therefore most immediately present in the conversation', one for 'that other person who is either physically over there away from us or is the person we referred to earlier before we talked about someone else and who is therefore more remote from the conversation', and one for 'that person who is physically over there near you but not near me or is someone you've been talking about or has something to do with you and is therefore located more in your bit of the conversation than in mine'. So Wolf Hall would be way easier to read in Latin.)

It is unfortunate that it hasn't got 'zie' and / or 'ze', since those do seem to be quite popular.

I think "popular" is a very relative term here. I never encountered "zie", let alone any of the less widespread new pronouns, until I got to university and saw someone using it on their LJ, and even then I seem to remember it taking me ages before I thought "oh, wait, that's a gender thing isn't it?" I just thought they were doing it to be quirky and was not enough of an asshole to say "hey, what the fuck is up with that?"

I suspect that there are an awful lot of people out there who simply have never encountered one of the various invented genderless pronouns out there, and the fact that there are multiple sets makes it even worse - it fractures the usage too much and makes it less likely any one set will attain the critical mass needed to enter widespread, mainstream common usage.

Which is another argument for the use of "they", in some respects: if clear communication and the expressive capacity of the language is important, then arguably it's better to use an established and universally-understood word which more or less works in context than to use an invented word which might be perfect for the context you want but just ends up confusing people. Either the people so confused assume you misspoke/had an amusing typo and will just ignore it, or they'll say "wait, what was that xe thing back there?" and the conversation will get thoroughly derailed.

Obviously, if someone actually prefers to use a particular pronoun in relation to themselves I will try to do the decent thing and use it, but if I know they don't wish to present as male or female but I don't know which pronoun they prefer then "they" is clearly the polite choice. Or I could actually use their name.

But there's also a more mundane level of operation where the purpose of genderless pronouns is to respect the identities of actual human beings who don't want to be gendered by language

Although of course one also needs to respect the identities of people who *do* want to be gendered by language.

Which is where it all gets very complex.

(let alone more radically as a complete replacement for 'he' and 'she')

Uh, nobody's advocating that. "He" and "she" are perfectly good terms to use when talking about gendered individuals.

I am, however, entirely willing to advocate a replacement for "he or she". For my money, "he or she" looks horrendously ugly when you're talking about people in the abstract. Furthermore, almost every time I've seen "they" used in that way it's been absolutely clear from the context what the deal is, and in those few cases where the context wasn't clear that's usually because of a whole host of communication failures on the part of the writer, rather than the use of "they".

And as Andy says, it's not like I'm going off to grammarian forums to troll pedants, I'm just happy to have ammunition to use if they launch a nitpicking raid.

As to genderless pronouns, all points made so far seem very reasonable if we look at them as a big plan to make people reflect on or discard their prejudices about gender, or that kind of thing. But there's also a more mundane level of operation where the purpose of genderless pronouns is to respect the identities of actual human beings who don't want to be gendered by language. Which to some extent short-circuits some of those other concerns: to use as an example Dan's point, and I stress that it does strike me as a good and interesting point, that one may not gain very much by writing 'If someone is hungry zie should eat' rather than 'If someone is hungry he should eat' because people will still tend to imagine the hypothetical hungry person as male, that doesn't actually arise if you're sitting around a dinner-table with people and one of those people is someone who doesn't feel male or female and would feel uncomfortable being referred to by either 'he' or 'she' and you want to say to your neighbour 'I think if Robin is hungry [pronoun] should eat'.

Personally I haven't investigated nearly enough to know what I think about it all, but that's why the regendering tool is useful, because at least it helps one become more familiar with the proposition and therefore better able to judge how it works in practice. It is unfortunate that it hasn't got 'zie' and / or 'ze', since those do seem to be quite popular.

The bits of grammar that are important are the bits that literally everybody knows and nobody ever makes mistakes about except for non-native speakers. Nobody ever "accidentally" says "the mat sat on the cat" or "mat cat on sat the" instead of "the cat sat on the mat" because they've got an intuitive understanding of English grammar and they know where subjects and objects go in sentences even if they don't know what the formal definitions of subject and object are.

Interesting linguistic fact: even babies don't get word order wrong when learning a language. They don't randomly put the word 'the' in the wrong place - rather, they're more likely to just leave out the word 'the'. And if they get words wrong, they're typically mixing up, say, one noun for another, rather than a noun for a verb or suchlike.