The Problem of Polysynthesis

I hold many controversial opinions. One of them is polysynthetic is a bad term, not just because of it's vagueness but because of what it signals, especially in conlanging. Anyway, this argument with some people got out of hand (#selfexaminationhurts) (I said some dumb things too) so I never really got explain why it is bad beyond the vagueness.

Here's the first thing I never sent and then I'll follow up with some other ideas I've had since then:

"Anyway, my point is that even now, the languages we choose to label as polysynthetic (especially taking the large amount of morphemes approach) tend to fall on minority and especially disenfranchised groups. This wouldn't necessarily be a problem if there was actually an agreed on definition for polysynthesis. But there isn't, because whenever someone tries to come up with something, other people get angry because their language gets excluded (the biggest example of this being Baker and his exclusion of Inuit languages) or because a language they don't consider polysynthetic is included. So we are left with a category that means "lots of morphemes and if feels that way". Which then brings us back to the point that "feels that way", for whatever reason, closely aligns with "languages spoken by minority groups". So we have a category that (like all morphological typologies, mind you) doesn't tell us really much of anything about the languages classified in it, except that 1) they have long words with multiple morphemes; 2) are not placed with the other languages for some reason.

And that's crux of it. The category doesn't tell us anything that synthetic (here being agglutinative and fusional) doesn't already tell us. Yet people defend it so viciously and want their language to be in the category. Why?"

Well, a big part of it is what I call "fetishization of the exotic" (and I am guilty of it too). Polysynthesis is seen as something cool, so you want your language in it (especially for conlanging). It is seen as cool because it is different from IE (and especially English) therefore something you want to be. And that's where the underlying "racism" (for lack of a better term) comes in. It doesn't mean that the linguists/enthusiasts are being racist, but they are, because of the way the terms has been used, perpetuating stereotypes and signaling certain ideas (namely primativity/noble savage/north americanness) through the use of the term "polysynthesis". It is the "exotic" that really binds the different types of polysynthesis together, more so than head-marking, polypersonal agreement or noun-incorporation.

Why is this important? Well, the category "polysynthesis" hurts conlanging and reduces its diversity. How? First of all, since there is little if any actual tendencies that fit for polysynthesis, it isn't signaling features for the most part. Instead it signals that you want your lang to be North American-like, especially in a Salishanesque way. This is fine and all, but it further reduces the amount of languages people learn about and makes them think that polylangs actually have many binding features. It also means that they are less likely to learn about features not found in those languages. For example, I did an informal survey on switch reference (with a bunch of polylang enthusiats) and none of us could think of any conlang with a switch reference system (other than my own, in progress one). Why? I think a lot of it has to do with the fact that the primary references for polysynthesis don't have it, even though it is very common in polylangs in other parts of the world (like New Guinea) and even in the United States! In all this reduces the diversity of conlangs (I've seen one papualang (excluding my own) and none based on Australian langs, for instance) because people have an incomplete view of what "polysynthesis" really is and don't realize it.

Fetishization of the exotic aside, polysynthesis would be an okay term if it could be well defined, people agreed on a definition, didn't try so hard to fit every language into it and recognized its limits and unreasonability. It would be fine if the community used a wider variety (not just of Native American langs) of languages to act as references, showing the diversity in the term and maybe counteracting some (though not all) of the underlying marks/stereotypes within the term. But it doesn't and we don't have the self-awareness nor desire in the community to fix this. So I'm stuck ranting about it on a blog. Well the next time the inevitable "how do I polylang" or "I never see polylangs (cue 15 polylangs)" post comes up, I can link this as I try to raise awareness :p