It's a lot of work to convey no additional information.

Complexity/difficulty in a language is usually kept to a minimum, and complexity in one area of a language tends to be balanced by simplicity elsewhere, pretty much wherever the speakers can get away with it without sacrificing the function of conveying information. For example, while Latin gives English-speakers a bewildering array of decisions to make about which suffix goes on each word, a Latin-speaker would be shocked by the number of words (s)he needed to put together, in the right order, to say the same stuff in English without all those suffixes. But each is only as complicated, in its own way, as it needs to be to get the information conveyed, and whatever complications it doesn't need to help convey information, it doesn't have or use.

...except that one of them has genders (and the other once did). How did those ever get started in any language, and how can they survive for more than a few generations, when they require extra work for the speaker but don't contain any additional information?

I was crap at French for this reason....ok this was one of many reasons. Le chat or La chat... why the gender?

Efficiency strange subject. Language lacks. Avg. Eff. lack sev. ways.

One - style. Ease listener comprehension superior ease data communication eff.

Two - Evolution. Like nature, non-optimized, gets there. Artifacts, whale pelvis, present.

Also the gender isn't logically set out which makes learning French as a second language to English a bit tricky, or at least not obvious. I understand it wasn't planned.
For example:
Le soutien-gorge.
EDIT: La barbe as well, my French is poor so may not be right.

I have no idea how genders got started, but there's no reason for them not to continue. Nearly every European language has them and people are clearly managing.

OT, I realize, but David Sedaris wrote a hilarious piece in which he described his diffculties learning the language while in France, and especially the problem of assigning gender to articles. He was particularly bemused that the word for "penis" was feminine, while the one for "vagina" was masculine. He eventually found an easy way around this when he realized the plural article, "les," was not specific to gender, and thereafter would only refer to items in pairs. His boyfriend was very puzzled as to why they had two identical toaster ovens.

NineBerry wrote:But the same principle applies. Because all speakers agree on the gender of objects, there is additional information added which helps communication be increasing redundancy.

Let me give you an example.

I can say in German

"Als ich den Kugelschreiber in die Tasche steckte, erinnerte ich mich daran, wie ich sie gekauft hatte."

Translation:

"When I put the pencil into the bag, I remembered the moment I had bought it".

Bought what? The pencil or the bag?

It is clear in the German language.

"Als ich den Kugelschreiber in die Tasche steckte, erinnerte ich mich daran, wie ich ihn gekauft hatte." ("he", the pencil is female, i.e. I am talking about the pencil)
"Als ich den Kugelschreiber in die Tasche steckte, erinnerte ich mich daran, wie ich sie gekauft hatte." ("she", the bag is female, i.e. I am talking about the bag)

It's the same in Spanish. In Italian, though, both "bag" and "pencil" are female (borsa / matita), unless one uses "sacchetto" (m) for bag.

I'm sure other examples can be found where gender is not helpful, in all languages that have them. So the additional information does not always help.

I agree that gender is more of a nuisance for language learners.

Delvo wrote:It's a lot of work to convey no additional information.

wrong. It does provide lots of information!

Basically, what you get is a kind of randomized algorithm that introduces - as already noted - a small extra level of redundancy. The extra level of redundancy given by a gender system, although not 100% helpful, is apparently helpful enough to compensate for the extra learning it requires. (Extra redundancy isn't of course the only thing it provides - it also provides some of the benefits of having two distinct third persons (generally called proximate and obviate) without requiring all the complications that such a system brings along. This extra distinct third-person system won't work all the time - you'll run into contexts where all the things you need to distinguish are of the same gender, but it's a sort of randomized solution, and it works often enough to be worth maintaining, for less of a ~cost than developing a proximate-obviate system of native north American style would entail for most languages that currently do have gender systems)

Randomized solutions that only help some of the time really are a surprisingly efficient and cheap method of going about this kind of thing!

What makes the benefits even more obvious is the fact that we don't actually hear every sound the speaker we listen to makes (nor do we read every letter in every word on the page or screen in front of us!) - and having extra clues here and there makes it way more likely that the utterance is reconstituted into the right string of morphemes in our mind.

There's a few things that contribute to us not hearing every sound.
1) Apparently the brain actually does skip over a lot of them for some reason - possibly it's too taxing to actually parse every sound/read every symbol. Some randomly built system that often enough provides enough extra hints by reducing the number of possible words uttered into a smaller set is a good enough solution to be rather prevalent once it appears.
2) Something armchair linguists don't think much about is background noise.
3) We actually don't utter our words the same way every time - mistakes in production occurs all the time, to varying degrees. Again, redundancy reduces the number of options we subconsciously need to consider and thereby improves the situation. (Otoh, we also do gender-related mistakes, which whenever we actually do hear enough of the noun don't cause much of a problem, but if there's too much interference or whatever obfuscating the noun itself, a gender error might prove fatal to recovering the noun. I would love to know how the brain actually searches through its internal lexicon and such when doing this!)

Complexity/difficulty in a language is usually kept to a minimum, and complexity in one area of a language tends to be balanced by simplicity elsewhere, pretty much wherever the speakers can get away with it without sacrificing the function of conveying information. For example, while Latin gives English-speakers a bewildering array of decisions to make about which suffix goes on each word, a Latin-speaker would be shocked by the number of words (s)he needed to put together, in the right order, to say the same stuff in English without all those suffixes. But each is only as complicated, in its own way, as it needs to be to get the information conveyed, and whatever complications it doesn't need to help convey information, it doesn't have or use.

And most languages mandatorily mark information that one could do just as well without, most languages have mandatory redundancies of various kinds. The claim that complexity/difficulty is kept at a minimum is not, properly speaking entirely true, but a rather popular misunderstanding of how things work out. Languages with little redundancy tend to get more to ensure that speech acts don't elicit 'excuse me, what did you say?' every single time - and often this will initially happen by periphrastic means, and won't therefore show up in morphological tables or whatnot, so most people that like to talk about how difficult different languages are by means of comparing morphological tables won't notice it. Grammaticalization later turn such things into morphology (just look at French, where the orthography of course obfuscated the fact that it's halfways going on being a polysynthetic language!)

...except that one of them has genders (and the other once did). How did those ever get started in any language, and how can they survive for more than a few generations, when they require extra work for the speaker but don't contain any additional information?

I hope my post showed how the assumption that they 'don't contain any additional information' is grossly mistaken, and that the instances where the degree of helpfulness seems low or nonexistent it still might be beneficial, and that furthermore, it's best to look at it as a randomized algorithm - doesn't work all the time, but often enough to be a worthwhile solution.

Polysynthesis is a typological class which is kind of fuzzily defined. Generally speaking, they're languages where the morphology generates monstrously large verbs, and not by compounding, but by actually stacking multiple affixes on top of each other.

Canonical examples of polysynthesis can be found among the Eskimo-aleut, Iroquian, Algonquian, Salishan and Uto-aztecan languages. (IIRC there's examples in Siberia, South and Central America, the Caucasus and Australia as well, but I don't recall any of those without looking them up).

Exactly where the line goes between polysynthesis and mere agglutination or fusionality is not defined in any universally agreed upon terms, and whether such a distinction is relevant or worthwhile is not entirely clear.

NineBerry wrote:But the same principle applies. Because all speakers agree on the gender of objects, there is additional information added which helps communication be increasing redundancy.

Let me give you an example.

I can say in German

"Als ich den Kugelschreiber in die Tasche steckte, erinnerte ich mich daran, wie ich sie gekauft hatte."

Translation:

"When I put the pencil into the bag, I remembered the moment I had bought it".

Bought what? The pencil or the bag?

It is clear in the German language.

"Als ich den Kugelschreiber in die Tasche steckte, erinnerte ich mich daran, wie ich ihn gekauft hatte." ("he", the pencil is female, i.e. I am talking about the pencil)
"Als ich den Kugelschreiber in die Tasche steckte, erinnerte ich mich daran, wie ich sie gekauft hatte." ("she", the bag is female, i.e. I am talking about the bag)

Surely, this only works when the two things are of different genders. There will still be this ambiguity; it's chance whether they're the same gender or not.