devintus, povintus, ktp
I've read that these are not recommended, but can't recall any convincing reasons why not. To me, devintus and povintus are a bit easier to understand at first glance than estus devinta / estus povinta.
Duolingo teaches both ways. Can someone explain why devintus / povintus are not recommended?
The conditional forms don't mark tense, and I think many argue that the addition of tense doesn't add clarity, and makes a form which is aesthetically unappealing.
If the tense is a necessary part of the meaning, it can be derived from context:
Mi aĉetus loterian bileton, sed mi forgesis. -> (literally) I would buy a lottery ticket, but I forgot. -> (inferred meaning) I would have bought a lottery ticket, but I forgot (to).
Manĝu terpomojn kaj feliĉiĝu!
well, the most obvious problem seems to me the practice of attaching verb endings onto other verb endings. I'm glad they seem obvious to you, but I find them long and complicated, and unlike anything else in the Esperanto grammar. You can agglutinate words and derivational suffixes to your heart's content, but nowhere do you duplicate grammatical endings
here I expect you to reply, "but you can add 'a', 'e', and 'o' endings to particples"; but I don't think that's exactly right. Those three are all separate grammatical cases with separate rules: "-inta" forms a particple (not unlike a Latin particple); "-inte" forms an adverbial (like in Russian, but I can't recall the name of the feature; or like the ancient Greek aorist form); and "-into" forms a noun that specifically refers to the verb's actor. I agree that these are related roles, and they're made easier to recognize and learn by being related structurally, but it's not a case of coming about by the pasting together of endings.
"You can agglutinate words and derivational suffixes to your heart's content, but nowhere do you duplicate grammatical endings".
It's true that one doesn't, but I'm still not sure why not. Combining grammatical endings doesn't seem inherently different to me than combining other affixes, with the general guiding principle being: if it makes sense and doesn't violate the explicit grammar rules, it's valid Esperanto. However, it's not a huge deal for me, and I'm happy to follow the recommended conventions. Just wanted to understand them better.
"povintus" and "devintus" are correctly constructed grammatically. They derive from "estus povinta" and "estus devinta"--which isn't usually what people intend to be saying when they use these words. "I would be having been able", "I would be having ought to have" . . . that's not the intent.
So they are basically somewhat clumsy misformations, that people think mean one thing while they actually mean something else (and the something else is pretty unclear).
The participial endings aren't in the same category as the normal verb endings, although participles retain verbal characteristics.
The issue is that in almost every case the simple forms "povus" and "devus" are adequate. Or "povis" and "devis". These more complicated forms represent people's attempts to reproduce ways of expression from their usual language.