Oh, wow, this will be so useful — thank you! I don’t have any formal training in linguistics, so it’s always nice to confirm with an actual linguist that I’m understanding everything right.

(And, yes, there’s only so many example sentences to go around :) )

Well, if you check my thesis (and can make sense of the convoluted sentences), a big part of it was wading through the mess of terminology and also quite imprecise definitions. I think Dixon's descriptions are a good first order approximation to understand the concepts, but they lack a certain amount of rigour, at least if we want to use them meaningfully for cross-linguistic comparison. That's why I attempted a definition more centred on Lazard and Haspelmath.

It's quite impressive how much you know without formal training. I spent years learning about these subjects. However, since graduating in 2013, I haven't really done anything with linguistics.

> I think Dixon's descriptions are a good first order approximation to understand the concepts, but they lack a certain amount of rigour

Oh, I completely agree. I personally quite like Deal’s ‘ergative’ and ‘absolutive properties’ [0] as a first attempt; but I haven’t yet seen Lazard and Haspelmath, so I should look into that more.

[0] http://linguistics.berkeley.edu/~ardeal/papers/Deal-ergativi...

> It's quite impressive how much you know without formal training. I spent years learning about these subjects.

Why, thank you! But I don’t think I’m so different in this respect: I spent a good portion of my spare time last year reading everything I could find about ergativity, and had read lots about linguistics even before that. (What can I say; it’s an interesting subject!)