I don’t get the point of gender (and gender modification) in any language. Was there ever any good reason besides making things more complicated than necessary?
Didn’t the ancients have better things to do? Like thinking about survival?
Why did any ancient consider it terribly important to decide that a tree was a macho man and a leaf was a girly girl?
Yes. All languages have some degree of redundancy, and gender (or, more generally, noun classes, of which gender is one type) is one of the ways of accomplishing that. It provides a way of error correction – with a phrase like “una casa pequeña”, the fact that all three words are together indicates that “una” and “pequeña” are modifying “casa”, but the -a at the end of each of them provides extra assurance that they go together. This is admittedly more important in languages with no fixed word order such as Latin and Sanskrit. The presence of gender in languages like Spanish and Hindi is just a holdover from their parent languages.
With Indo-European languages, proto-Indo-European originally had two genders: “masculine” and “other”. The “other” gender eventually split into “neuter” and “feminine”, but which nouns were assigned to each class was only partially due to their inherent gender, and was also a function of what sounds were in the words, and random chance.
Sort of related, how, in languages with gender, do new words get assigned gender? Who decides if, say, “computer” is masc, fem or neuter? I know France has some sort of linguistic oversight academy that acts to keep the language “pure” or whatever, but does it decide how to classify new words too?
Anecdotal, but… I noticed in Brazil that newly adopted words tend to be masculine by default, with words like o shopping (the mall), o show (the on-stage event - concert, usually), o Internet, o videogame.
As for words like “computer…” Of course, computer is an old word - it just means “one who computes.” But, I’m not sure how they decide whether it should be masculine or feminine. If all I were familiar with was Portuguese, I’d say that the same principal as above applied - the word for “computer” in Portuguese is o computador. However, in Spanish, for some reason, the word for computer is feminine: la computadora.
In the instance of German, gender seems to get assigned mostly by analogy to existent word with the same ending. For example, Computer is masculine because German nouns ending on -er usually are masculine.
Well, both of those are neuter so… 
Remember that it’s not a question of some ancient genius sitting down and creating a language from scratch, and then teaching it to his family. Languages have murky roots in the grunts of pre-man, and although as yBeayf writes we have a decent idea of how things were in proto-Indo-European we can’t really know how the elements in that language arose. And Indo-European is only one group of languages.
Some lingual elements die out, and others arise, speakers of Finnish might wonder why the Indo-European ancients bothered with articles (caveat, I don’t know if the lack of articles in Finnish is ancient or not). The purpose of a word in a sentence might be decided by the word order, or word endings.
Hmm, I seem to be ranting. But languages are slowly evolving “living” things, and we don’t really know how most of the elements arose. As a native speaker of a language with gendered nouns I find it a completely natural thing. Words sound wrong if you try to change the gender. How this arose I don’t know. Maybe it started out with names, but instead of spellings (Billy for a boy, and Billie for a girl) they used words that were different in pronounciation. Then it spread to animals, buck, doe, fawn. And then … Well, maybe someone’s already mapped this out for real for Indo-European, so I’ll stop theorizing.
Although my Mac OS X uses “el ordinador,” which I’m assuming is Spanish Spanish. At least it used to in an early iteration of Mac OS X. I’m not sure if it finally says “la computadora” which I prefer, having dedicated my learning to Mexican Spanish. I think I’ve “el computador” while out and about, but then I’ve also seen “una computer” so I just stick to the simplicity of English when I can.
This one is feminine. (Of course, it may be like “lettuce” which has regional differences).
The gender of computers (insert witty joke here) in some languages may have something to do with the fact that most of the original human computers were women.
This is interesting purely from a linguistics point of view, but doesn’t the use of gender to accomplish this betray something about the thought processes of peoples who developed early languages? It sounds like this is an after-the-fact analysis of how this structure has become useful and therefore persisted, rather than how it arose.
Has there been psychological work to tie into linguistics to understand why language arises with gender assigned to inanimate objects?
One cite that I can give is Women, Fire, and Dangerous Things, by George Lakoff – the title comes from a gender classification in Dyirbal, where one gender groups women, fire and danger. (But I must confess to only reading bits of Lakoff’s seminal work).
In the Linguists: Why does a language’s grammar simplify over time? thread, some of us referenced The Unfolding of Language : An Evolutionary Tour of Mankind’s Greatest Invention by Guy Deutscher.
Deutscher also touches on the origins and evolution of gender in languages.
No, there never was a reason. Human language never developed along logical lines … it’s always been an organic thing. Once communication was fostered in early humans, the advantage over non-speaking competitors was fully conferred. No further refinements or corrections were evolutionarily necessary.
However, do note that for native speakers, grammatical gender is absolutely no hindrance to learning a language. It doesn’t even so much as slow down the rates at which children learn fluency.
Steven Pinker’s books (esp. The Language Inctinct) discuss this topic at length.
Further good reading on this topic from this well-cited paper:
Come to think of it, you’re right. I think that’s because the Portuguese word for net, a rede, is also feminine.
My bad.
For foreign words, it appears to be popular consent decides it is. And there’s no consistency, either; it’s " le battledress" but “la B.D.”
There’s been a few papers that have looked into this by using made up words to figure out what people’s intuitions were, and they were remarkably regular across speakers. There’s something about native-speaker intuition that comes into play. While gender assignments to inanimate objects may seem haphazard, some linguists think there’s more at play. One article (and, dang, I hope I have this right) is Schwichtenberg, Beate; Schiller, Niels O., “Semantic Gender Assignment Regularities in German.” Brain and Language, 2004, 90, 1-3, July-Sept, 326-337. From the abstract: “Gender assignment relates to a native speaker’s knowledge of the structure of the gender system of his/her language, allowing the speaker to select the appropriate gender for each noun… Participants presented with a category … & a pair of gender-marked pseudo-words … preferentially selected the pseudo-word preceded by the gender-marked determiner “associated” with the category… This finding suggests that semantic regularities might be part of the gender assignment system of native speakers.”
Disclaimer: This is not my area of expertise, I mostly do sociophonetics, but we had a long discussion about it in my historical linguistics class, which got me interested in it a little bit.
Gender is one type? I had the impression that gender means pretty much exactly the same thing as noun classes, for example if a language classifies nouns based on whether they are living, or little furry animals, or have the color green, that each of these would be called a gender.
That’s kind of splitting hairs … but AFAIK you are essentially correct. However, plenty of researchers would rather just avoid the term “gender” altogether to avoid confusion between grammatical gender and natural (sexual) gender.
I suspect (but have no learned cites to offer) that “grammatical gender” originated in assonance: that it sounded better to say anima bona but mundo bono. Latin’s five declensions show this sort of function: 1st declension has -a- stems, 2nd -o-, 3rd consonantal and -i-, 4th -u-, and 5th -e-. As nouns representing female forms came to be concentrated in the 1st and 5th declensions and those with male forms in the 2nd and 4th, a nascent grammatical gender began to evolve. Romance nouns tend to preserve the gender of the Latin forms, whether or not they’ve retained anything like the same endings, and with articles evolved from ille and related demonstratives compelling the need to identify grammatical gender for them.
Interestingly, Russian has both grammatical and natural gender. Nouns fall into grammatical masculine, feminine, and neuter in terms of the adjective forms that agree with them, but in forming the direct object, neuter nouns and inanimate objects described by masculine and feminine nouns take the accusative case, while direct objects identifying male and female persons and creatures take the genitive.