Over on his Calendar, Adam Cadre has an interesting idea about English pronouns. He points out that one of the reasons that gender-neutral third person pronouns have seen poor adoption in English is that differentiated pronouns make it easier to track multiple third-person referents. All previous attempts to de-gender English pronouns have typically done so by simply neutralizing the contrast. (To be clear, both Cadre and I think this is a good goal.) If your aim is to have a gender-neutral alternative to “he” and “she”, this works fine; but if your goal is to actually take the inherent gender out of language, you run into problems with sentences like the following: (Example due to Cadre.)

John drove Yoko to the studio, where she played him her latest song.

Think about using the singular they for that:

…where they played them their latest song.

A whole lot of information is lost there.

Obviously, there are many languages that get by just fine without multiple third-person pronouns – ambiguities like this can be resolved by rephrasing if necessary. But I think that adult native speakers of contemporary English, particular adult ones, are likely to be highly resistant to a change that so modifies how you need to construct sentences; Cadre’s conclusion that this factor has hindered English adoption of ungendered pronouns in general speech rings true to me.

Cadre’s proposed solution to this, however, doesn’t quite. Given the goal of tracking multiple referents in a sentence, Cadre suggests that we simply track them by linear order, with pronouns meaning “the first (leftmost) possible referent” and “the second possible referent”. This strikes me as cognitively weird: I can’t think of any language with anything like pronominalization that is sensitive only to linear order. I’m unconvinced that such a system would actually be learnable, though I couldn’t say for certain.

(To be clear, neither Cadre nor I think that English pronouns are likely to change by decree – our best shot at a gender-neutral pronoun is to encourage the continued evolution of singular they. But it’s fun to speculate.)

There are other options, though. Ultimately, what we’re talking about is obviation – marking some third-person things as being peripheral to the discourse. Lots of languages have interesting strategies for obviation; it seems unlikely that English speakers would ever want to adopt any of the more exotic ones. But I suspect that if we actually tried to adopt Cadre’s suggestion of fi(rst) and se(cond) pronouns, they’d actually wind up being something like “proximal” and “obviate” in terms of topicality. I say this because topicality in English correlates quite strongly with word order (or at least with subjecthood, which correlates with word order), so in most cases this would look the same as Cadre’s suggestion. But I would expect examples like the following to come up:

Speaking of linguists, my friend was talking to Alan Prince the other day and fi [Prince] said that…

I’ve tried to set this up so that the first human referent (“my friend”) is very strongly non-topical. While obviously I have no intuitions about how fi should be used in English, cross-linguistic facts about obviation make me think this is the sort of pattern we would see emerge.

(Of course, I’ve never worked on obviation directly. Linguist-friends who have, any comments?)

A last note: Cadre brings up an example usage from the game of Go:

Go is one of my hobbies, and in the English-speaking go community, the player with the white stones is always referred to as “she” and the player with the black stones as “he”, for this very reason [that it makes referent tracking much easier].

Huh, that’s a good idea! I hadn’t encountered the idea of conventionalizing pronoun gender like this. Also, I feel like it has the nice effect of bringing hypothetical female referents into what could otherwise (I’d imagine) be an intensely male-dominated activity. Perhaps more conventions like this would actually be a nice intermediary step on the way to language that’s more gender-neutral.