Grammatical gender consistency across languages

December 4, 2012 12:49 PM

Are grammatical genders, as a rule, consistent across the Indo-European languages which use them?

I am thinking of romance languages generally, but information on others are by no means discouraged. Any examples of nouns (or other gendered words) that are differently gendered in different languages would be very helpful.

posted by obloquy to Writing & Language (30 answers total)

I'm sure other folk will bring a lot more examples, but I always found the sun and moon interesting. From this blog:

le soleil, la lune
il sole, la luna
el sol, la luna

French, Italian and Spanish: Sun male, Moon female.

die Sonne, der Mond

German: Sun female, Moon male.
posted by harujion at 12:54 PM on December 4, 2012 [1 favorite]

I think there's going to be a built-in disconnect if you're comparing two-gender languages (Romance languages) to three-gender languages (like German.)
posted by workerant at 12:56 PM on December 4, 2012

In German, Mädchen, which means "girl", is grammatically neutral rather than being the expected feminine. Meanwhile in French, Spanish, and Italian, the word for "girl" (fille, chica, ragazza) is feminine as one would expect.

So, no. Nouns are not necessarily congruent across Indo-European languages that use gender for inanimate nouns.

That said, in my experience gender does tend to be congruent across Romance languages. Though I'm sure there are exceptions. I'd be curious to know whether nouns adopted post-Roman Empire tend to be the same gender across Romance languages. For example I know that the French and Spanish words for "computer" are both feminine, despite not being cognates at all. Which is sort of interesting.
posted by Sara C. at 12:57 PM on December 4, 2012

Girl in French is la fille (feminine) and in German it's das Mädchen (neuter). Edit: dammit!
posted by workerant at 12:58 PM on December 4, 2012

workerant brings up a good point -- there are so many approaches to gender for non-animate nouns across Indo-European langauges that it might be hard to compare in a meaningful way. You've got langauges that don't do it at all, languages that have two categories, and languages that have more than two categories. For starters.
posted by Sara C. at 12:59 PM on December 4, 2012

Best answer: Nope! They're not consistent. There's been a bunch of interesting research done on how the gender of nouns affects cognition for speakers of different languages.

That's a link to a paper I found when googling the subject. It has many examples in which genders of nouns differ in different languages. It has a lot of great sources for prior research as well.
posted by lesli212 at 12:59 PM on December 4, 2012 [3 favorites]

The madchen one isn't a great example because it's a feminine word with a neutral diminutive attached to it.
posted by empath at 1:00 PM on December 4, 2012

Best answer: For example I know that the French and Spanish words for "computer" are both feminine, despite not being cognates at all. Which is sort of interesting.

Do you mean "L'Ordinateur"(m) and "La Computadora"(f)? Or were you thinking of something else?
posted by vacapinta at 1:03 PM on December 4, 2012

Except that "neutral diminutive" isn't an option at all in any Romance language I'm familiar with - when you add suffixes, they have to match the gender of the root word.

So again, this might just be a meaningless question in the first place. There are just too many variables in terms of what is possible to do in each language.
posted by Sara C. at 1:03 PM on December 4, 2012

vacapinta, whoops, I thought it was La Ordinateur for some reason. Which makes no sense now that I see it typed out.

Maybe nobody in this thread should listen to me.
posted by Sara C. at 1:04 PM on December 4, 2012

Response by poster: If it helps we could limit the question to those languages with no neuter grammatical gender. But vacapinta's comment might be a really good example of differing genders for a relatively new noun in two closely related languages.
posted by obloquy at 1:07 PM on December 4, 2012

In my German class (for English speakers) the Haitian guy (native French-speaker, ESL) was sitting there smugly while the teacher was going over the basics of the idea that every noun has a gender associated with it. Fast-forward about a week, when he used the wrong gender in a sentence about a table. No, says the teacher, it's 'der Tisch', not 'die Tisch'.
What? The guy kind of freaked out. But, tables are feminine!, he says. (la table, in French)

I thought it was very interesting; as an English speaker, I wasn't really trying to associate any permanent sense of gender to the object, I was just memorizing word-pairs (der Tisch), but apparently this guy had, if not an association that tables are inherently feminine, at least a strong association between the 'la' in French and the 'die' in German. Does it mean something that this was a couple of weeks into the class when this happened, i.e. if he didn't notice it immediately there must be a lot of nouns that retain gender?
posted by aimedwander at 1:22 PM on December 4, 2012

Either that or he'd been getting them wrong for weeks and you just weren't privy to it because he didn't do it aloud during class. Especially if this was a college course where you probably only meet a couple times a week. That's only about four opportunities for him to audibly make a mistake in your presence.
posted by Sara C. at 1:25 PM on December 4, 2012

Sara C.: You've got langauges that don't do it at all, languages that have two categories, and languages that have more than two categories. For starters.

Which brings up another interesting question (for me): Do any PIE languages have more than three genders?

Otherwise, I'd say it's just three major groups: the ungendered (English and ... ?), the two-gendered (most Romance languages, for starters, which lost their neuter gender over the 1st Millenium generally), and the three-gendered.
posted by IAmBroom at 1:27 PM on December 4, 2012

Even for something like fenêtre (f., French) and Fenster (n., German), they actually derive from the same Latin word, fenestra (f.), and they're different genders.

On the other hand, you also have pes/pied/Fuß, dens/dent/Zahn all masculine, and nox/nuit/Nacht all feminine, but they have common Indo-European roots. But in there, too, there are counterexamples (naris/nez/Nase).
posted by one more dead town's last parade at 1:32 PM on December 4, 2012 [1 favorite]

And then you have Danish, in which nouns have two genders, but they are not masculine and feminine. They are common and neuter. Swedish is the same way.
posted by expialidocious at 1:36 PM on December 4, 2012

Wikipedia says that Polish and maybe a few other Slavic languages have more than 3 genders, resulting from subdividing the 3 classes. PIE had animate/inanimate classes, and only subsequently evolved the fem/masc distinction within animate (so later masc/fem systems are a result of losing the neuter/inanimate class, apparently.)

Once you get outside of Indo-European langauges you will find all kinds of systems for dividing up nouns ('noun classes' for every-noun-gets-one systems like Romance languages, 'noun classifiers' for the huge and more fuzzy systems with 20+ different categorizations)

I also wandered in to talk about the same research leslie212 mentioned - there's a whole list of French/Spanish contrasts used in a study on p. 381 of that pdf!
posted by heyforfour at 1:41 PM on December 4, 2012 [1 favorite]

Here's 10 nouns from Latin and their descendants in a few Romance languages, which I guess is one of the best chances that gender has to be consistent:

Latin           French      Spanish     Italian     
aer(m)          air(m)      aire(m)     aria(f)     
bilis(f)        bile(f)     bilis(f)    bile(f)
ira(f)          ire(f)      ira(f)      ira(f)      
animal(n)       animal(m)   animal(m)   animale(m,f)
bestia(f)       bête(f)     bestia(f)   bestia(f)
bracchium(n)    bras(m)     brazo(m)    braccio(m)  
ars(f)          art(m)      arte(m)     arte(f)     
dorsum(n)       dos(m)      dorso(m)    dorso(m)    
globus(m)       globe(m)    globo(m)    globo(m)
pila(f)         pile(f)     pila(f)     pila(f)

(I didn't choose these by any rigorous method — they were just the first ten words I came up with starting from a list of common English nouns — but maybe it gives some idea.)
posted by stebulus at 1:45 PM on December 4, 2012 [3 favorites]

Note that French, Spanish, and Italian do not have neuter, so neuter has to map to something in the descendant languages. Most second-declension neuter nouns (-um) are going to end up mapping to masculine because they're already o-stems anyway and are identical in everything but the nominative/vocative and accusative.
posted by one more dead town's last parade at 1:51 PM on December 4, 2012 [3 favorites]

From what I recall of reading part of this book, Germanic languages aren't consistent (as noted above, they don't have the same number of genders anyway), but there are patterns. (Something about words going from neuter to feminine as you go east, but don't quote me on that.)
posted by hoyland at 1:57 PM on December 4, 2012

Response by poster: Lots of super interesting and helpful responses; thanks! Given the mundanity of the examples in French & Spanish in the PDF ("fork", "banana", "broom", "cloud", etc.), as well as the examples given here, it seems like it's fair to say that assuming congruency between languages as a general rule is a bad idea.
posted by obloquy at 2:10 PM on December 4, 2012

As you've seen, they're not all consistent.

One that always comes to mind (since it screwed me over on a test) was the word for table.

In French:

La table (f)

In Italian:

Il tavolo (m)

I couldn't remember the gender on my Italian test and guessed "la tavola" and was wrong. Won't forget it a second time. ;)

No other French/Italian ones come to mind, but there are others that would drive me nuts while I was learning Italian. There were just enough that I would feel mostly comfortable guessing using the French gender but I'd still get tripped up.
posted by juliebug at 2:13 PM on December 4, 2012

assuming congruency between languages as a general rule is a bad idea.

Yup.

French: La mer (the sea, f)

Spanish: El mar (m, although maddeningly, sometimes f in literary/artistic/idiomatic uses)
posted by ambrosia at 2:21 PM on December 4, 2012

Interestingly, English is probably mostly genderless because it is a creole of Norman French and Saxon English, which itself was more or less a creole of Danish and Old English by the time the Normans got there.

So, twice English had large groups of people who couldn't remember each other's noun genders, and probably mumbled something like "the" instead of whatever the correct pronoun was. Eventually everyone gave up on gender.
posted by musofire at 2:27 PM on December 4, 2012 [4 favorites]

Note that French, Spanish, and Italian do not have neuter

Just a quick note to say that Spanish does in fact have a neuter form, but not in the same way as German (it doesn't refer to people or objects but concepts).
posted by ob at 2:47 PM on December 4, 2012

Funnily enough, juliebug, "la tavola" is also correct. It's just two different ways of saying table.

Sometimes genders aren't even consistent among dialects (the only ones I can compare are a few Italian dialects, though) but I really do not know what the initial impetus is for differentiation.
posted by lydhre at 4:49 PM on December 4, 2012

Just a quick note to say that Spanish does in fact have a neuter form

What I meant was that all Spanish nouns are masculine or feminine. Ello, esto, and aquello look to me like they come from illud (illō), istud (istō), and hoc illud (hōc illō) (ablative forms in parentheses). I'm not sure about eso. You could probably get away with acting as though they're masculine; they don't appear to behave substantially differently.

It appears to be an Indo-European trait that the nominative and accusative are identical for the neuter gender in all cases. I know this holds for Latin, German, and Russian (and also for inanimate masculine nouns in Russian).

It's also not surprising, given the Latin neuter o-stems collapsing into the masculine, and their being in the same declension, that masculine and neuter nouns and adjectives are similar to each other in German and Russian as well as Latin (three different language families within Indo-European).

Far more than I expected to write about the neuter today.
posted by one more dead town's last parade at 7:18 PM on December 4, 2012 [1 favorite]

Spanish neuter is not a separate set of neuter pronouns, articles, etc. It's basically a way to use certain adjectives in noun form. Using the (masculine) objective pronoun in the way you'd normally use an article.

An example I just dug up on an about.com site when I googled this (because the idea of Spanish neuter sounded weird to me after 5 years of intensive Spanish study) is

Lo importante es amar.

Which loosely translated, means, "The important thing is to love." Instead of putting a placeholder noun in there (such as "thing"), Spanish puts in a placeholder objective pronoun. Which is... random... but only a "neuter" noun group in the most abstract sense.

Also, just to be perfectly clear, the word "lo" in that sentence is masculine. So still, not really neuter compared to the way that German nouns can be neuter.
posted by Sara C. at 11:11 PM on December 4, 2012

Except that "neutral diminutive" isn't an option at all in any Romance language I'm familiar with - when you add suffixes, they have to match the gender of the root word.

The problem there is regarding "-chen" as a suffix. Mädchen is really a sort of compound word, the first half comes from "Magd," which was a feminine word and a cognate to the English word "maid". The second part is something that makes a word diminutive, but it's probably better to think of it as a word (it was a word of sorts, eons ago) that just happens to be neuter. Think of Mädchen as meaning "maid-smallness," where "smallness" is always neuter.

Interestingly, English is probably mostly genderless because it is a creole of Norman French and Saxon English, which itself was more or less a creole of Danish and Old English by the time the Normans got there.

So, twice English had large groups of people who couldn't remember each other's noun genders, and probably mumbled something like "the" instead of whatever the correct pronoun was. Eventually everyone gave up on gender.

This is silly. It's not how language and history typically work and wouldn't explain why other European languages have retained gender and other complicated features despite longer and much more frequent periods of occupation by groups of people with even harder language differences to overcome. If one believes it, it poses the question of why the residents of Britain were so stupid - people all over the world speak multiple languages fluently with no problem, and without formal education! I've spent time in places where uneducated peasants have spoken fluent German, Romani, Romanian and Hungarian (three separate branches of IE and one totally unrelated language) for centuries without any more blending going on than the borrowing of a few words.

There are quite a lot of serious theories out there why English lost gender, but that isn't one of them!

Here's an interesting one.

But to answer the original question. No. Within some branches of PIE there may be some more consistency of gender than you'd get from pure chance, but not enough to help you much. You're far better off learning some general "rules" about gender for each language (French words ending in -age are likely to be masculine, etc) than guess wrong 42% of the time.
posted by Dee Xtrovert at 11:57 PM on December 4, 2012 [2 favorites]

Farsi is an Indo-European language, and it has no gender differentiation at all.
posted by thegreatfleecircus at 7:07 PM on December 8, 2012 [1 favorite]