Talk:Grammatical gender/Archive 2

This is an archive of past discussions about Grammatical gender. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.

Archive 1

Archive 2

Archive 3

Archive 4

I have heard that in Arabic the phrase "I love you" has different forms according to gender. It would be nice if someome could include that example. :-) FilipeS 12:53, 19 November 2006 (UTC)

Doesn't sound too particular. Although I'm not sure if there's a grammatical issue. Many languages has speech distinctions depending on male and female speech (such as Japanese). 惑乱 分からん 15:07, 24 January 2007 (UTC)

I know that Arabic makes a gender distinction in the second person, between anta "you (male)" and anti "you (female)"; so yes, the forms would be different. This site gives the following for "I love you" in Egyptian colloquial Arabic: ana uħibbuk - to a man, ana baħibbik - to a woman. (When direct objects, pronouns are suffixed to the end of the verb; -uk vs -ik). FWIW, cum grano salis. --RJCraig 18:59, 24 January 2007 (UTC)

While it's true there's a difference in Egyptian Arabic, the data above isn't correct. I'm not an expert on Egyptian, but it would be baħibbak to a man and baħibbik to a woman. The first form above is mainly Standard Arabic, where you have uħibbu-ka to a man and uħibbu-ki to a woman. I've marked the morpheme break: the difference is in the object suffix which is marked for gender (as it is in most dialects). Drmaik 05:11, 25 January 2007 (UTC)

Could you write it in Arabic characters, as well? FilipeS 16:54, 25 January 2007 (UTC)

Does the sentence "I love you" have different versions according to gender in Japanese? In Indo-European languages, it doesn't. FilipeS 15:38, 24 January 2007 (UTC)

I think that could depend on the way it's phrased, see Gender differences in spoken Japanese. 惑乱 分からん 17:20, 24 January 2007 (UTC)

Oh, thanks a lot, guys! I'll add those to the article. :-) FilipeS 22:15, 24 January 2007 (UTC)

I see above the following:

• 3) Alexandra Aikhenvald considers “gender” to refer to a system of three or less distinctions (“always including masculine and feminine”),

I haven't seen Aikhenvald's analysis, but gender even in the familiar Indo-European sense need not always include a masculine and a feminine gender. In some languages the masculine and feminine genders have merged into a common gender. In Dutch this appears to be a change currently in progress. Thnidu 19:26, 28 February 2007 (UTC)

Would FilipeS please explain in terms that even I can understand how one Indo-Pacific language's classification of penis-like things as masculine, and squat things as feminine, means that the distinction between the way that English he, she, it refer to the gender of realities, not of words, and the way German der, die, das refer to the gender of words, disregarding the gender/sex of the realities that the words refer to, is of no significance. (As an aside, English sometimes personifies objects, such as ships, the Moon, Spring, and then refers to them perhaps as "she", in line with the virtual reality of the personification; but when it does not personify them, it refers to them as "it", in line with the unpersonified reality.) Lima 13:16, 16 April 2007 (UTC)

As the argument goes, gender is special in English, because it is determined solely from semantics, rather than from morphology or some other criterion. Except that that's not true, on two accounts:

• It's not true that gender in English is always based on strictly semantic criteria. Two counterexamples are the use of "she" for ships, and the assignment of gender to indefinite pronouns like "someone" or "anyone". Granted, both cases are the subject of some controversy, but the fact that one can designate a ship as "she", or "someone" as "he" (without knowing the real gender of the person or people) shows quite clearly that the allegedly natural basis of gender in English is overstated. Just as in any other language where gender is morphologically marked, there are ambiguous cases, where gender must be assigned through some convention.
• It's also not true that having a gender classification based solely on semantics disqualifies a language from having grammatical gender. As shown by the example from the Indo-Pacific languages (where there is a supposedly perfect correspondence between semantics and classification), irregularities in gender classification, while common, are not a requirement. FilipeS 13:55, 16 April 2007 (UTC)

I am sorry, I still don't get it. If I write in English: "Spring is coming with her train of green-clad elves and dancing flowers", I am not attributing grammatical gender to the word "spring": I am only picturing Spring as a female person and using the pronoun that fits that picture. The same holds for the use of "she" to refer to a ship. When a ship is not personified, it is referred to as "it". Surely you don't disbelieve this. If you do, try reading, just for instance, Titanic - A special exhibit from Encyclopaedia Britannica. The use of "he" for an undefined personal subject is a traditional English usage now under attack on the grounds that it gave preference to the male, which in fact it did. The one-time favourite Strunk and White, while preferring this usage, admitted that it "was unquestionably biased to begin with". The usage did not attribute gender to the word "someone": it treated a "someone" (not the word, but the reality) as, prima facie, male, or it subordinated, for ease of discourse (whether this was excusable or not is another question), any possible female to the possible male. Using this as a basis for saying English has grammatical gender, deprives "grammatical" of any specific meaning that I can grasp. One might as well say that Turkish has grammatical gender because, though it has no gender-specific pronouns, it can distinguish between words/realities such as "son" and "daughter": "Baba oğluna karşı, oğul babasına karşı, anne kızına karşı, kız annesine karşı." In that case, every language has grammatical gender, since all, I suppose, have words that serve to distinguish males and females. Please excuse my difficulty in understanding. Lima 17:05, 16 April 2007 (UTC)

'The usage did not attribute gender to the word "someone"'

Oh, yes it did. Not a natural gender, obviously, but a conventional, grammatical gender. "Someone", in traditional English usage, is masculine. That's exactly how it works in other languages. FilipeS 18:25, 16 April 2007 (UTC)

Are you saying that English attributes gender to the word "someone"? Does it not rather attribute gender to the person referred to by that word? Those who think of a vague "someone" as at least primarily male (as was traditional) use "he"; those who refuse to think in that way use "he or she"; and every English-speaker, when speaking of someone in an all-female class, will use "she" and has always used "she". To me it seems clear that the word has no gender whatever. It is much the same as when you picture the spring or a ship as female persons, and then refer to them as "she". Surely you will not say that Turkish-speakers attribute gender to the words "oğul" and "baba" and to the words "kız" and "anne", when they do no more than recognize that the first two refer to males and the second two to females. English does the same, using "he"/"she"/"it" only according as the speaker pictures the reality behind the word. This is in contrast to most Indo-European and Semitic languages, whose speakers make terms (adjectives, pronouns, in some languages verbs) agree with a word's grammatical gender, regardless of the sex/gender of the reality behind the word. I regret that my lack of understanding would force me to add "{{fact}}" to what I think is your affirmation that the word "someone" has a grammatical gender in English. Lima 20:15, 16 April 2007 (UTC)

Yes, I am saying that English assigns a gender to the word, not the person referred to by "someone". How could you assign a gender to the person -- or persons! -- when you don't know who he is?

The difference between Turkish and English is that Turkish never had grammatical gender; Old English did, and modern English still has traces of it. This is all the article is saying. FilipeS 20:21, 16 April 2007 (UTC)

Turkish never had grammatical gender; English no longer has. Which gender do you say English assigns to the word "someone"? Feminine? As in "At yesterday's meeting of the Women's League, someone left her handbag behind"? Isn't it obvious that "her" refers to the person, not the word? Common? As in "At yesterday's meeting of Episcopal Church bishops, someone left his or her (her or his?) prayer book behind"? The pronoun you use really refers not to any supposed gender of the English word "someone" (which has none), but to the gender/sex that you attribute - and in this sense do indeed "assign" - to the person that you presume is meant by the word. Lima 04:29, 17 April 2007 (UTC)

Yes Lima, I agree. See my comments below. Analysis needs to take into account three separate things: referent (exophor), antecedent (cataphor) and pronoun (or whatever the anaphor). English does not attribute gender to words. However, the gender distinctions between pronouns are more subtle than simply refering to natural gender. In fact, even these pronouns do not have gender in English, rather they are just most often (but not always) used to communicate natural gender.

• [Australian English] She'll be right mate. (It'll be OK.)

The non-sexist prescriptivists might want to forbid this. But *it'll be right doesn't sound as nice as it'll be alright and that's much more formal than the friendly, culturally affirming she'll be right.

• [Australian English] It's a bitch. (This is very difficult.)

This has nothing to do with female dogs or unpleasant women, although it may well have come about by difficulties in coping with pregnant sheepdogs. What it communicates in colloquial Australian is intimacy between speaker and hearer, informality and associations of common participation in the traditions of Australian culture. English does not demand she for "grammatical agreement" with bitch. However, use of unexpected pronouns often carries overt or covert associations.

• The coal-miner downed tools and headed for the lift. Another day, another dollar! Time for a beer with mates, a good night's sleep. One day closer to the weekend! The smell of drying sweat filled the lift as it launched groundwards. The doors openned, and out she stepped.

I shouldn't leave such lyric prose lying about under the GNU license, but she'll be right.

G'day y'all. Alastair Haines 05:31, 17 April 2007 (UTC)

Thanks. I agree. The expression "reality", which I used, is not the most apt. Nor is "reality or mental construct", which is the best I can think of immediately. Others know better than I do how to find a good expression. In my poor opinion, attribution of grammatical gender to English words is on a par with or even worse than attempts to interpret English grammar in terms of, say, Latin grammar by attributing to English words a dative and an ablative case (marked by prepositions). Lima 07:28, 17 April 2007 (UTC)

I think it's less dogmatic and more honest to accept that the question of whether English has grammatical gender or not is open. Anything else would be POV-pushing. As for the gender of "someone", please read what I wrote. Then I'll get back to you. I see many people trying desperately to convince themselves that English is somehow special. Well, sorry, but as far as grammatical gender is concerned, the only special thing special about it is how little of it is has -- quantity, not quality, is the difference. FilipeS 16:30, 17 April 2007 (UTC)

Suppressing arguments for one side is not the way to leave the question open. Lima 04:40, 20 April 2007 (UTC)

Peace friends, I think it is an interesting question, not one any of us would die for. There are arguments both ways, and sadly some are motivated by political conviction. Vive la difference is what I think regarding real life male-female differences, but the English language is a separate issue to me. If English has gender it is more like the languages that discern animate and inanimate, personal and impersonal. I can agree with both of you. You're both doing great work. Alastair Haines 06:19, 20 April 2007 (UTC)

Filipe, sorry. Alastair, thanks. Lima 08:53, 20 April 2007 (UTC)

Hello again. Perhaps it would be a good idea to make a full disclosure. Some time ago, I made a substantial rewrite of this article. In it, I emphasized the difference between English and languages with "true" grammatical gender, arguing that what makes gender grammatical is the agreement shown in other parts of speech, besides nouns or pronouns. In this, I was influenced by the version of the article which preceded my rewrite.

But the more I thought about it, the less this stand made sense to me. One thing that helped me change my mind was when I also made a (less important) rewrite at the Grammatical number article. The two things, grammatical gender and grammatical number, are morphologically very similar. They're just used for different purposes. And no one makes a big deal about whether grammatical number is expressed on nouns, verbs, adjectives, or whatever. All that people ask is whether it's encoded in the morphology of the language in some manner. So I convinced myself that the sharp dividing line between English and other languages was artificial.

Now, I will readily admit that English is special, as far as grammatical gender is concerned. But that's only because there's so little gender marking left in the modern language. As I said before, it's a difference of quantity, not quality. Nevertheless, differences of quantity also matter. If a language only inflects one word for gender (say "male"/"female"), then I think everyone will agree that it isn't a language with grammatical gender. And modern English is indeed very close to being such a language; but it's not quite there.

So, I rewrote the article again, de-emphasizing what seemed to me to be an exaggeration of the difference between English and languages with "true" grammatical gender. Regards. FilipeS 22:06, 20 April 2007 (UTC)

I do not think that sex-related pronouns are a reason for saying that a language has grammatical gender. Did written Chinese suddenly become a language with grammatical gender through the twentieth-century invention of a special character that, by combining the character for the genderless third-person pronoun with the character for "woman", allowed the writer to indicate that the referent is female? I don't think this changed the grammar of the language. I know some give a very wide meaning to "grammar", but I think most people understand "grammar" to refer basically to relations between words, not to relations between words and referents (semantics). English pronouns link not to words, but to referents (personalized or not). Their use is a matter of semantics, not grammar. In contrast, Portuguese pronouns usually point to the gender of words, and so their use is usually (even if not always) a matter of grammar, not semantics, so that we can attribute grammatical gender to Portuguese pronouns, though not to English. FilipeS has a different point of view, but I believe mine is not unreasonable. Lima 08:19, 23 April 2007 (UTC)

I'm not sure the two of you disagree, at least I'm finding it hard to see where you disagree. I agree with what both of you say. Maybe I'm contradicting myself in a way I don't understand, and maybe I've missed something. Yes, grammatical gender is about morphology of words, not natural gender of referents. The two are related to one another in strikingly different ways in different languages. In Greek, autos, aute, auto are clearly inflected for gender. In English he, she, it are actually separate words, which inflect for case, not gender. In fact, he and she have different etymologies. Even though it's probably no accident they look similar now, nowhere does English use a prefix s as a morpheme like it does with suffix s. Distinct words for man and woman, like Mann and Frau do not give German grammatical gender. German inflections do that, which makes Mann and Frau, all the more clearly, distinct words. My vote is that English does not have grammatical gender (which makes it relevant to this article only as one of the many languages that do not have grammatical gender). This is not to say that English cannot indicate natural gender, any language can do that by using merely two adjectives -- male and female. In English, though, the third person pronouns carry most of the weight of expressing natural gender -- a simple and effective strategy, but with interesting issues. Are there any other languages where expressing natural gender depends largely on distinction between third person pronouns? Alastair Haines 12:15, 23 April 2007 (UTC)

I agree with Alastair. Does Filipe agree with him? Lima 12:26, 23 April 2007 (UTC)

You both make good points. Certainly, gender marking is nowhere near as pervasive in English as it is in a typical language with grammatical gender. The current version of the article acknowledges this, and notice that I did not object when another editor removed English and Afrikaans from the list of languages with masculine, feminine, and neuter grammatical genders. If someone asked me, point blank, whether there is grammatical gender in English, I guess I might very well answer "Basically, no". Whatever there is is meaningless compared to other languages such as German, or Portuguese, or Russian.

But the fact that English once did have gender, and still carries traces of it, should not be simply swept under the rug. Just as I would generally describe Spanish as a language "basically with masculine and feminine genders only", but might add, as the article does, that it has also a few determiners with neuter inflections, which can be regarded as a trace of the Latin neuter. Keeping a historical perspective of languages seems important.

In short, I am leaning more now towards the position that English is indeed not a "grammatical gender" language, but I still insist on making clear that it isn't "gender-free", either. It's not like Finnish or Hungarian, where there's only one word for "he" and "she". The gender of Old English has left a mark. Finally, it should also be understood that what makes English fail to be a grammatical gender language is not a qualitative difference, such as describing "reality" as opposed to... (what, exactly?)... but rather the inescapable fact that there are only a minute handful of gender inflections in English.

It seems we are nearing a consensus. Would you like to rewrite the article's discussion of gender in English?... FilipeS 19:22, 23 April 2007 (UTC)

Would Alastair like to? Better him than me. Or why not Filipe himself? Lima 04:03, 24 April 2007 (UTC)

I will work on this, I have recently researched a little Old and Middle English. I will seek a good source or sources for current academic consensus on the subject. It seems likely that between the three of us we can come up with something that is informative and NPOV. Alastair Haines 04:21, 24 April 2007 (UTC)

First I want to register here how impressed I am with both Grzegorj and FilipeS. Both acting in good faith to give Wiki the best articles possible. Both showed great humility and team playing, while having very impressive skills and knowledge. Truth is the winner when we dare to settle our differences by investigating new information. Thanks for your fine example, both of you.

Regarding this article, I think great progress has been made, and we are getting close to knowing what we need to say at this entry. I want to make two suggestions and offer to help with finding sources and contributing.

My first suggestion is more of a comment. I think there are four main groups of people who will use this article:

1. Editors working on grammar topics who simply want to cite an internal reference.
2. Web surfing students who may want an easy introduction to the subject.
3. Non-technical native English speakers who want to check up on the technical side of sex/gender/grammar/non-sexist debates.
4. Technically knowledgable editors who want to keep the article up to date with current journal publications.

I've listed these groups in an order for a reason. I think we need to work towards satisfying each of them in this order. So we first need to ask, "is there a simple summary at the top of the article that works for internal links?" The body of the article can give students a fuller picture, which then addresses the facts in the political debates. Finally, we can address the highest level of scholarship. Actually, I think the article is close to this already, I just wanted to state it clearly.

My second suggestion addresses the topic of the article directly, in the light of what I've just mentioned. I think we need to be very clear about the difference between grammatical gender, which is about the use of words, and the way words are used to communicate real, semantic gender in referents.

Example:

• [Australian English] Nice car, mate! Yeah, she's a beauty!

In no way do Australians think the noun car is feminine, nor that real life cars are feminine, nor that women are objects in some way, like cars are. Also, Aussies don't read the Chicago Manual of Style before showing off their new cars! What such examples show is that he and she in English are not always used to communicate something about sex, personal relationship, animacy or historical gender associations with certain nouns.

I have more to say, but it is late here and I must sleep. So I'll stop here and wait for a reply. I've learned a lot from this article. The reference to language families and list of gender systems, not to mention many fine individual examples, give this article outstanding content. I hope I can help with the finishing touches to make this a cutting-edge featurable article. Cheers all! Alastair Haines 16:46, 16 April 2007 (UTC)

Hi. Thanks for your kind remarks. Unfortunately, Grzegorj has not been around for a while. I agree with your comments. Yes, the article feels a little "heavy" and technical, and would benefit from a more lightweight introduction.

You make a good point about how people may come to this article looking for information on non-sexist language and related topics. There already are several specialized articles about that issue (Gender-neutral language in English, Gender-neutral pronoun, Gender-specific pronoun, Gender-specific job title, and so on), but I can see how it might be useful if this article referred to the others early on. Suggestions are welcome.

Regards. FilipeS 19:29, 16 April 2007 (UTC)

I'm back. Here's the main thought.

• [xyz] pronoun <-- pronoun/cataphor (some languages have inflected pronouns, others have distinctions for real gender, others don't)
• [Abcdef] noun <-- antecedent/anaphor (some languages have inflected noun classes, some of these can be called "gender" classes)
• <real thing> <-- referent/exophor (some things have gender, animacy or personality, others are treated in language like they do)

What is the "essence of grammatical gender?" Probably it has to do with noun classes and inflections and agreement.

"Does English have grammatical gender?" Definitely it has "relics". Arguably, the pronoun system reveals a consistent pattern of thought among native speakers that includes: gender, animacy, personal relationship and indeterminacy/common-ness/generic-ness.

Suggestion: this article should summarize and cite sources that answer the questions above, and some other related questions.

Note: this article is not primarily about English. It is about noun classification/inflection and agreement, synchronically and diachronically. That alone is a huge, exciting and worthy topic. It does seem worth "sharing"/"splitting" the idea over a couple of articles. What should be covered here? What should be covered at Noun classes?

Alastair Haines 04:02, 17 April 2007 (UTC)

Some interesting thoughts there. Someone else also mentioned how they "feel" about objects and what gender they would be if those objects were alive. Obviously, that is really not what grammatical gender is about. It is understandably tough for native speakers of English to comprehend this unless they have studied another language, since English nouns don't have gender at all (or they are all the same gender). At least concerning Germanic and Latin languages, what it comes down to is what article or suffix is associated with each particular noun. (Please forgive me if I am overlooking something here -- I just hadn't seen anyone make this point yet and it seemed pertinent).

For example, the German word for "the" (in the nominative case) is either "der", "die", or "das" -- depending on the gender of the following noun (note that "die" is also plural). In English, on the other hand, whether you consider a car to be more like a woman or more like a man, the article does not change: it is still "the car". In actuality, the gender of a given noun often contradicts common sense. For example, you may think "girl" should be feminine but in German, it is not: "das Mädchen" is neuter because it is a diminutive. Personally, I think studying German is a fantastic way to learn about grammar, since it still employs grammatical features that have been lost over the years in other languages -- such as gender and case. It will help you understand the workings of your own language (such as why it is incorrect to say "between Johnny and I" -- this really bugs me but I digress...) 71.71.192.231 (talk) 04:04, 6 January 2008 (UTC)

I'm really grateful for your work FilipeS. I noticed you reverted a sentence that talked about reality. There is a little information about this at reference. Linguists call the object in real life, say a dog, the referent, to distinguish it from the word itself, say canis (in Latin). Not all individual words have real life referents, in fact most individual words connect words to other words.

This is important when discussing the distinction between "real-life" referent (which may or may not have real gender) and nouns (which may or may not have grammatical gender). For example, a referent like a girl in real life has real feminine gender, but in German the word for a girl Mädchen is grammatically neuter. This leads to different things in different languages. Many will occasionally break their own grammatical rules to communicate the real gender of a referent when speaking about it. So we might want to know what German speakers prefer to say:

• Das Mädchen ist nett. Sie heisst Maria. (The girl is nice. She is called Maria.)
• Das Mädchen ist nett. Es heisst Maria. (The girl is nice. It is called Maria.)

Most languages have ways to communicate real gender if it is important, even if it goes against their normal pattern.

I'm sure you know all this, but it doesn't hurt to repeat it I guess. Alastair Haines 01:18, 20 April 2007 (UTC)

Of two elementary German grammars for English-speakers, one says of "das Mädchen" (given only as an example) that both "es ist geschickt" and "sie ist geschickt" are correct, the other says of "das Fräulein" (but apparently of that alone) that it "requires the fem. pron. sie." My feeling, which I cannot document, is that in Italian the pronoun used in connection with "il donnone" would be feminine, though an adjective would certainly be masculine. In Gaelic, the quite elaborate Graiméar Gaeilge na mBráithre Críostaí, which uses the word "inscne" (gender) only of grammatical gender and the word "gnéas" (sex) of what, until recently, everybody simply called "sex" in English, states: "In the case of people, it is the person's sex, not the gender of the noun, that normally decides the gender of the pronoun ... In the case of animals and inanimate objects, the gender of the pronoun usually agrees with the gender of the noun" (pages 144-145). It lists exceptions to both rules. So there are languages apart from English in which the choice of pronoun at least can be unrelated to the grammatical gender of the noun and can depend instead on the sex/gender of the referent.

I conclude - though perhaps FilipeS would not - that only in a language in which they at least sometimes disagree with the sex/gender of the referent can pronouns, on their own, be cited as indicating the existence of grammatical gender in that language, i.e. gender of words, not merely of referents. If pronouns never disagree, we have to look to variations in adjectives, articles and verbs as grounds for affirming the existence of grammatical gender. Lima 08:53, 20 April 2007 (UTC)

Thanks for the info on German, that is very helpful isn't it. Selection of pronoun gender seems to be influenced in an almost competitive way by German users desire to conform to both grammatical pattern and sex/gender of referent. Interestingly, perhaps just as in English we would not always consider a boy a "man", perhaps German doesn't feel so compelled to accord "womanhood" to ein Mädchen. Alastair Haines 12:21, 20 April 2007 (UTC)

I don't think that those examples are very informative. Of course you can say "The girl is nice. She is called Maria" in any language with grammatical gender. That's not because the grammatical gender of the word "Mädchen" is trumped by that of its referent, but because the referent "girl" can be designated by many different synonymous words, some of which are bound to be grammatically feminine. This is why the text "The girl is nice. She is called Maria" is acceptable: the subject of the second sentence is not "Das Mädchen", but rather another, implied, word of feminine grammatical gender (such as her name, "Maria").

Notice also that while "Das Mädchen ist nett. Sie heisst Maria." is perfectly acceptable, under no circumstances would "Die Mädchen ist nett. Sie heisst Maria." be correct. The modifier that directly qualifies "Mädchen" is never allowed to make an exception for "real" gender. Anyway, this would be my analysis.

Concerning the word "reality", specifically. I don't like it, and would use it sparingly in the article, or avoid it altogether. Here's why. First, we already have the term "natural gender". Second, if we say that English pronouns agree with the reality (not always exactly true, by the way: what about "she"-cars and "it"-dogs?...), then what does that make of languages with grammatical gender? "Delusional"? But language is a reality, too. The difference is that English focuses on the meaning of words, while those languages focus on their morphology. Notice that languages with grammatical gender typically have a significant amount of inflection, which may mean that people are used to paying extra attention to the form of a word. Regards. FilipeS 22:14, 20 April 2007 (UTC)

That's a very interesting idea Filipe. Are you fluent in a strongly morphological language? Is it a noticeable strategy employed by people that they choose synonyms with morphological gender that reinforces natural gender when they want to make a point of it? Could you give some examples? I deal with inflected languages often, but dead ones, it's hard to guage what were natural speech patterns in those. Alastair Haines 06:19, 21 April 2007 (UTC)

No need to worry about the word "reality". It no longer appears in the article. Since a personalized referent must necessarily be pictured as either male or female, and since animals are often, indeed usually, thought of merely as objects, abstracting from their sex, which in general is not at all as obvious or important to us humans as the sex of humans is, I think it is still the referent as pictured, not the gender-less English word, that dictates the choice of the pronouns "she" and "it". Lima 08:11, 21 April 2007 (UTC)

I am fluent in Portuguese. How strongly morphological it is, I can't qualify. It does have grammatical gender (masculine versus feminine). A sentence analogous to the German example you gave above is perfectly possible and colloquial. For example, Ele chama-se Mário. É uma pessoa simpática "His name is Mário. He's a nice person." Here's a breakdon of the two sentences:

• Ele: "he"
• chama-se: "is called/named"
• É: "Is" --> The subject is omitted, as Portuguese is a null subject language.
• uma: "a", feminine singular indefinite article
• simpática: "nice", feminine singular adjective
• pessoa: "person", a feminine noun

It's quite common for the gender of an anaphor to change when a new sentence begins, "forgetting" the grammatical constraints of the previous sentence, and agreeing with meaning instead (as in the German sentence). I wouldn't even say it's against the rules. And I've just discovered a word for it, synesis! :-)

You have similar examples of this in English with grammatical number. Here's a made up example: "In 1939, Germany went to war. But their decision to attack the Soviet Union prevented the quick victory that they were expecting". O.K., my sentence may be a little clumsy, but I hope you can see the point: when you begin a new sentence, you ignore how you represented the referent "German state" in the first one, using the plural number instead of the singular which would be required by "Germany". I think this sort of construction is common in English. Finding an example with grammatical gender would of course be more problematic. FilipeS 11:09, 21 April 2007 (UTC)

I wonder if the Portuguese example is really parallel. To me it seems to fit instead "Sie heißt Maria. Sie ist ein netter Mensch" (masculine) or "... ein nettes Mädchen" (neuter) or "... eine nette Person" (feminine). In these examples, the pronoun in the second phrase must, even in German, be the same as that in the first phrase, regardless of what follows. In my poor opinion, the parallel would have to be in the form of "A rapariga é simpática. Ela chama-se Maria e a sua amiga chama-se Tomázia", thus forcing the use of the personal pronoun. To constitute the parallel, "a rapariga" should then be replaced by a grammatically masculine word referring to a girl. I cannot think immediately of any such word, but I suppose they exist, even if they are rare. Lima 12:57, 21 April 2007 (UTC)

Dêmos as boas-vindas ao membro mais novo do nosso clube. É uma rapariga de Coimbra e chama-se Maria. "Let's give our welcome to our club's newest member. She's a girl from Coimbra, and her name is Maria."

• dêmos: "let's give"
• as boas-vindas: "our welcome"
• ao = a ("to") + o ("the", masculine singular)
• membro: "member" (masculine singular noun)
• mais novo: "newest" (masculine singular adjective)
• do nosso clube: "our club's"
• É: "She is", with implied subject (not marked for gender in Portuguese)
• uma: "a" (feminine singular indefinite article)
• rapariga: "girl" (feminine singular noun)
• de Coimbra: "from Coimbra"
• e: "and"
• chama-se: "her name is" (not marked for gender in Portuguese)
• Maria: feminine singular noun.

Exactly the same, I think. FilipeS 13:53, 21 April 2007 (UTC)

Alastair, I suppose Herodotus spoke natural Greek (his own dialect of course). Well, he used a neuter pronoun, corresponding to the neuter-gender word "παιδίον", not to the (male-sex) boy who was the referent, in 1:110 - ἔλεγε ὁ Ἅρπαγος τάδε, κελεύει σε Ἀστυάγης τὸ παιδίον τοῦτο λαβόντα θεῖναι ἐς τὸ ἐρημότατον τῶν ὀρέων, ὅκως ἂν τάχιστα διαφθαρείη· καὶ τάδε τοὶ ἐκέλευσε εἰπεῖν, ἢν μὴ ἀποκτείνῃς αὐτὸ ἀλλὰ τεῷ τρόπῳ περιποιήσῃς, ὀλέθρῳ τῷ κακίστῳ σε διαχρήσεσθαι. ἐπορᾶν δὲ ἐκκείμενον τέταγμαι ἐγώ.

This is not proof of general usage in ancient Greek, but perhaps it is at least an indication.

Can anyone tell me why Wikipedia has ceased to display polytonic Greek characters? Lima 08:49, 21 April 2007 (UTC)

The same usage with regard to the word "παιδίον", when clearly referring to a boy, not a girl, is found in Lk 1:59, 2:40. No contrary usage appears anywhere in the New Testament. Again, not really proof. For that, we would need examples of usage with other neuter-gender words, perhaps other diminutives (like "Mädchen", for that matter), referring clearly to people either of male or female sex. Lima 09:19, 21 April 2007 (UTC)

Ah! Lima, now I know why I understand you so easily. Herodotus reports Harpagos speaking to Astyagese of taking the boy to the desert, but not to kill him? The dialect is uncomfortable for me, and my Greek is only modest anyway. However, I can say that in John 15:26 (second last verse, p. 302. in NA27) you will see to pneuma ... ho ... ekeinos. There are two later, similar references. You can imagine the debate that has swirled about such things. W filie mou, Xaipe! Alastair Haines 09:37, 21 April 2007 (UTC)

Sorry, I am not convinced. Am I particularly pig-headed? You omitted the start of the verse, which explains the masculine gender of "ekeinos". The verse runs (this time I will transcribe the Greek letters and ignore the accents): "Hotan elthêi o paraklêtos (masculine) hon (masculine) egô pempsô hymin para tou patros, to pneuma (neuter) tês alêtheias ho (neuter) para tou patros ekporeuetai, ekeinos (masculine) marturêsei peri emou." Isn't it obvious that "ekeinos" refers to "o paraklêtos"? "To pneuma ... ekporeuetai" is in apposition to "o paraklêtos", and does not prevent agreement between "ekeinos" and "o paraklêtos", any more than the neuter "dôron" in the following phrase would prevent grammatical agreement with the masculine "hippos" on the part of words that follow "dôron": "Ho hippos (masculine), dôron (neuter) tou patros mou, kalos (masculine) men esti, pôlêsô de auton (masculine) - The horse, a gift from my father, is handsome, but I'll sell it." (Next time, I will just write the Greek letters: for me it is far easier.)

If there has been serious debate about the grammar of this verse, I suppose it only shows that Greek does choose pronouns to agree in gender with a word, not with the word's referent, and that some were presenting the verse as contravening the rule.

If I say: "Elementary, my dear Watson", I do so fully aware that others will say the same to me on many occasions. Perhaps immediately. Lima 11:39, 21 April 2007 (UTC)

In Old English, there are examples of recurrent use of pronouns that must, from context, refer to the same antecedent. However the first agrees with the grammatical gender of the antecedent, while the later ones refer to the natural gender. A very interesting phenomenon, especially given the way English has moved since. Several commentators on this have noted that the distance between the pronouns and the antecedent are like to have played a factor. With the NT use of ekeinos for the HS, see also John 16:13-14, a much clearer example, I agree 15:26 is not decisive in isolation. Alastair Haines 14:36, 21 April 2007 (UTC)

Returning to the German and Portuguese, it seems Portugese often doesn't require an explicit subject. A direct translation into Portugese would be the same for both my German examples! How would you translate the following into Portuguese Filipe and Lima?

• This is Maria, we greet her!

Alastair Haines 15:10, 21 April 2007 (UTC)

A possible translation is "Esta é a Maria, damos-lhe as boas-vindas." I've bolded the words that are inflected according to gender. Their meaning: esta = "this", a = "the" (direct article, normally used before people's names). FilipeS 15:23, 21 April 2007 (UTC)

Obrigado Filipe. I can't understand how the "we greet her" section works. I was hoping it would have something like "her" in it somewhere, but apparantly not. Now I'd love to know how Portuguese can say "we greet her" without needing to use a feminine pronoun. ;) Can you explain the second clause for me please. Alastair Haines 17:35, 21 April 2007 (UTC)

The word for "her" in this case is the indirect object pronoun lhe, which is the same for masculine and feminine. The second clause, translated literally, is "[We] give [to] her the welcome". "[We] give [to] him the welcome" would be exactly the same. FilipeS 18:17, 21 April 2007 (UTC)

Again thank you Filipe. I am reading Portuguese personal pronouns and possessives. Very nice work Filipe. :D Alastair Haines 23:42, 21 April 2007 (UTC)

I have had an idea. The German example I gave depended on the similarities between German and English. Namely, that we can specify natural gender, when required, by using 3rd person personal pronouns.

• Out of Mary and John, she is the taller one.

However, it seems that languages that lack personal pronouns, or do not distinguish gender in personal pronouns, sometimes use a similar strategy, but with demonstrative pronouns, if these are marked for gender.

• Out of Mary and John, that [marked for feminine] is the taller one.

In languages where there are simply no grammatical gender distinctions at all. I presume the same idea could be communicated, but "long-hand".

• Out of Mary and John, the woman is the taller one.

Clearly, the example is poor, the name would simply be repeated in most cases. However, a better example of the same strategy would be the way English has to do this in the plural.

• In my family, *shes are the taller ones.
• In my family, the women are the taller ones.

So, if communicating natural gender was actually important, could Portuguese do this with demonstratives like esta or essa, for example.

• [One girl to another]: A stranger followed me home from the shops today, I didn't like the look of _____.

English would say him, Portuguese would say aquele rather than lhe if the speaker wanted to imply it might have been a male stalker? Or, are words for people nearly all marked for natural gender anyway, so the gender would be clear as early as saying stranger [marked for male]. Alastair Haines 00:36, 22 April 2007 (UTC)

Portuguese does have separate masculine and feminine pronouns in both singular and plural: ele/ela, eles/elas (nominative), o/a os/as (accusative). Only the dative (indirect object) lhe lhes fails to distinguish. Just as French lui does not indicate whether the indirect object is masculine or feminine. In the last sentence, Portuguese would use "dele" (de+ele: of him), with no need to have recourse to demonstrative pronouns. Lima 04:41, 22 April 2007 (UTC)

Ah, I think that makes it a perfect example then. If you don't care about whether it was a man or a woman, you would say it that way. E.g. A woman followed me home from the shops. Answer: So what? A man followed me home from the shops. Answer: Was he creepy? Are you OK? Have you seen him before? Do you think we should report it? OR Someone followed me home from the shops? Answer: Was it a woman or a man? If Portuguese wanted to make it unambiguous, could it use a demonstrative pronoun, where English would use a personal pronoun? Alastair Haines 07:46, 22 April 2007 (UTC)

"You would say it that way." What way? Are you referring to "A stranger followed me"? Even that is unambiguous in Portuguese. It will be either "Um desconhecido" or "Uma desconhecida". For ambiguity, you would have to choose something like "uma pessoa", a word that I think would not be used if there were any worry about the occurrence. Or were you referring proleptically to "Someone followed me"? The response to this remark would surely be: "Quem foi?" (Who was it?).

The demonstrative pronoun corresponding to "lhe" would not be, as you perhaps think, "aquele/aquela" (which corresponds instead to "ele/ela"), but "àquele/àquela", i.e. "a" (to) + "aquele/aquela". Lima 09:37, 22 April 2007 (UTC)

Yes, I remember thinking before that because so many nouns for people are marked for the sex of referent, there's little need of gender distinctions in third person personal pronouns. Portuguese is "shaped" differently to English in this regard. Your proleptic suggestion is very helpful. For dramatical purpose, one can still disguise the sex of a referent until the point at which you want to reveal it. My example worked on the situational context that a stranger was following. Deliberately contrived so the question "Who?" would not be natural, but inquiry regarding gender would be, were it not supplied. Thank you Lima and Filipe for giving me a basic feel for how different the issues are in Portuguese to the languages I'm familiar with. Alastair Haines 10:16, 22 April 2007 (UTC)

Only now have I noticed your above remark about Jn 16:13-14, which, in view of my "Elementary, my dear Watson", I cannot let pass. There too, τὸ πνεῦμα τῇς ἀληθείας is in apposition, and the word ἐκεῖνος is grammatically quite unconnected to it, just as in Jn 15:26. It refers instead to the already mentioned παράκλητος, not to a following and unrelated πνεῦμα. Note the commas in "ἐκεῖνος, τὸ πνεῦμα τῇς ἀληθείας, ὁδηγήσει ..." marking the apposition. Just like the commas in "ἐκεῖνος, δῶρον τοῦ πατρός μου, καλὸς μέν ἐστι• πωλήσω δὲ αὐτόν", referring to a previously mentioned ἵππος. The text certainly does not employ such barbaric grammar as: "[ἐκεῖνος τὸ πνεῦμα (!) τῇς ἀληθείας ὁδηγήσει] Error: {{Lang}}: text has italic markup (help)." May I be permitted to repeat: "Elementary, my dear Watson"? Lima 10:43, 22 April 2007 (UTC)

I think you are really stretching things this time. Context makes it clear that the referent of ekeinos is the third person of the Trinity. Our question is, is this person concieved of as Paraklete or Holy Spirit in the use of the pronoun. Both expressions are used in the textual context of the pronoun. Without looking at the text, grammatical agreement would lead us to expect that since Paraklete is grammatically masculine, and Spirit neuter, the masculine pronoun would be far more probable. However, on the one hand parakletos is mentioned in verse 7. I count 70 words between parakletos (v7) and ekeinos (v13). On the other hand, to pneuma the aletheias is directly in apposition to the pronoun, with no intervening words. Given the actual text then, we find prefered interpretation of ekeinos as Paraclete over HS to be impossible. Having eliminated the impossible, what remains — cataphoric reference of masculine ekeinos to HS — must be the truth.

Liddle-Scott gives hippos, hê, mare, Opp.C.1.162, Hippiatr.14, BGU21iii 8 (iv A.D.). Which makes your classical reference most apposite. The gift was a stallion, which hippos and doron could not tell us, but ekeinos can, because it is not bound to grammatical agreement with either.

Perhaps I was misleading when I said debate has swirled regarding the gender of the HS. The debate that has swirled is a modern phenomenon, of the last three decades. Divergent views are not found in the ancient writers. The Gnostics were happy to address the HS as mother, but they were happy to write their own Gospels also. The modern debate is similar in that those who champion a genderless or feminine HS do not feel constrained to do so on the basis of the NT text.

My classical Greek is not good enough to compare with my knowledge of Koine, or the dialect of it that the NT may represent. The dual and optative had all but vanished in Koine. A number of other morphological structures had been streamlined. My early training in Classical Greek still leads me occasionally to become muddled and misread some vocabulary items which shifted in meaning over 400 years. It could very well be that classical Greek would have been much more particular than Koine, and Plato might naturally read John just as you do. However, in the 2nd century scrutiny of John, prompted by the debates that anathematized the Gnostics; "proto"-Christological and Trinitarian doctrine formed, which relied on these references to establish the personhood, not the gender of the HS. We two are far too distant from the language, to spend our energies defending one dialect against another. I count your comments as a free lesson in Classical Greek, and will hearafter hold my peace regarding Koine interpretations. Alastair Haines 13:25, 22 April 2007 (UTC)

Perhaps, after all, it is not as elementary as I thought, and still think. In this discourse of Jesus (Jn 15:26-16:15, 17 verses), what is the name that Jesus gives to the subject he is speaking of? Is it not ὁ παράκλητος, referred to repeatedly as ἐκεῖνος (Jn 15:26; 16:7, 8, 13, 14)? Only in two verses is the more specific τὸ πνεῦμα τῆς ἀληθείας added parenthetically, to explain which παράκλητος is in question, out of the many that the word could refer to. You mention verse 7, but forgot the masculine ἐκεῖνος in verse 8 and its implied application also to verses 9-11. Then in verse 13, after just one short verse without any reference to ὁ παράκλητος /ἐκεῖνος, we again have the repeated ἐκεῖνος before the mention of the explanatory τὸ πνεῦμα τῆς ἀληθείας. Surely that ἐκεῖνος, like the preceding ones, refers to ὁ παράκλητος. Any other interpretation is surely "really stretching things". By the way, my formal studies of classical Greek ended 52 years ago. For many years I have read more Koine Greek than classical. Personally, I do not think of Koine as a separate language: it is far closer to classical Attic than is, for instance, the language even of Herodotus, not to speak of that of Pindar, and yet their dialects are classified, along with Attic, as ancient Greek. And my comment on the passage in John is not based on rules of ancient Greek, but simply on rules of what I would consider elementary logic.

Of course, if you think the horse my father gave me was a mare, you are free to use καλή and αὐτήν. But remember that my comment below on Portuguese applies also to all forms of Greek, even the modern form. Lima 14:10, 22 April 2007 (UTC)

Ah! Since you are a Koine man, BDAG:302 ekeinos is ambiguous! It could be read to support either of us. On the one hand he says "resumptive", on the other he says "immediately preceding". Parakletos is not the most immediate reference in either context, however I fully understand the nicety of the reading you propose. The NRSV passes over translation of ekeinos in 15:26, so no gender is communicated in their translation (it is a gender debate sensitive translation). However, even they translate he in c. 16. I think we have come close to exhausting the possibilities of this discussion. But I certainly have learned a lot in the course of it. Thank you Lima. Alastair Haines 13:48, 23 April 2007 (UTC)

Apologies for not being able to follow your reasoning. I do not have the Bauer-Danker Lexicon. (I had to look up "BDAG" to find out what the abbreviation meant.) I don't think we need it. Maybe translations, which I hardly ever consult, have made it unclear to you. But the text itself seems absolutely clear. Why do you say that the word "parakletos" "is not the most immediate reference in either context"? What else can you point to? I think we were most recently talking about 16:13. The "ekeinos" there, I would have thought, is obviously the same as the immediately preceding "ekeinos" of verses 8-11 (mentioned explicitly in verse 8, but verses 9-11 continue to speak of the activity of that "ekeinos"). The "ekeinos" of verse 8 is no less obviously the same person referred to as "auton" at the end of verse 7. And that "auton" refers again no less obviously to the "parakletos" mentioned just a few words earlier in the middle of verse 7. I see absolutely nothing else between that mention of the word "parakletos" and the word "ekeinos" in verse 13 to which the latter word could refer. Instead I see a series of pronouns that put the link beyond doubt. So what do you see between "parakletos" and the "ekeinos" of verse 13? Do you perhaps have a different text of Jn 16:7-13 from what I have? Lima 15:38, 23 April 2007 (UTC)

But to return to Portuguese (and similar languages). In such languages, the feminine form, the form that has the marking -a, excludes the masculine, but the masculine (seen as unmarked, as the singular is seen as unmarked because of lacking the -s that marks the plural) does not exclude the feminine (is inclusive). "Denise Carvalho foi o segundo juiz afastado do cargo" (Denise Carvalho was the second judge to be removed) does not indicate whether the first judge removed was male or female. "Denise Carvalho foi a segunda juiza afastada do cargo" indicates that the first judge removed was female also. Though we use the masculine form when we say: "O brasileiro trabalha mais do que se pensa", we are saying that all Brazilians, women as well as men, do more work than is thought. Feminists would be unable to impose on Portuguese what they call "inclusive language" and which perhaps should instead be called "exclusive language", since it insists on interpreting words like "man" only in an exclusive sense. Just a curiosity! Lima 13:20, 22 April 2007 (UTC)

Ah yes, Lima! When given the opportunity, and it is rare. I share my personal musing that traditional English generic he is sexism against men (I'm being careful not to chew my tongue while it is in my cheek). She has been exclusive of men, where he has not excluded women. Males have no pronoun reserved for their exclusive use. Do I understand you correctly, Portuguese cannot be reformed in this by teaching people to treat masculines as exclusive, as people teach in contemporary English, because there are no alternatives except the exceedingly cumbersome brasileiro e brasileira? Alastair Haines 13:41, 22 April 2007 (UTC)

Interesting exchange! :-)

Indeed, in Portuguese Fui seguida até casa por um desconhecido ("A stranger followed me home") could, in principle, mean either a male "stranger" or a female "stranger". But in such a context the normal thing to do would be to say desconhecido if the stranger was a man, and desconhecida if it was a woman. Add to that the fact that it's more often men than women who stalk others (though not always), and desconhecido would definitely be interpreted as a male stranger. It's natural for people to want to know whether the stranger was male or female; as you pointed out, Alastair, the two situations carry very different degrees of risk.

If for some reason the speaker did not know the gender of the person who followed her (say, if it was a car following her from a distance), then she would more likely say Alguém me seguiu até casa ("Someone followed me home"). The word for "someone" is not inflected for gender.

For dramatical purpose, one can still disguise the sex of a referent until the point at which you want to reveal it.

Some authors have written love poems where it's unclear whether the addressee is male or female. You can imagine what a challenge that can be in a language like Portuguese! ;-) FilipeS 17:13, 22 April 2007 (UTC)

ROFL! Yes, from the little I know, it does indeed sound like a challenge. But what people will do for love! :D Alastair Haines 18:07, 22 April 2007 (UTC)

Can one really say that the words "iru" and "aru" are examples of Japanese gender? Isn't that just words with an animate/inanimate semantic meaning (or is the claim that Japanese has genders because these two words are referring to objects)? 惑乱 分からん * \)/ (\ (< \) (2 /) /)/ * 14:11, 20 April 2007 (UTC)

I don't know! I look forward to what others say. You understand our questions well. Do animate and inanimate or personal/impersonal count as "gender" categories? Are gender categories better understood as a description of noun classes? How much do pronouns tell us about grammatical gender in a language? Who has written the best books on these subjects? Are there any good online references? Arigato gozaimas. Alastair Haines 15:35, 20 April 2007 (UTC)

I think "animate" and "inanimate" could calssify as "grammatical gender", but Japanese looks strange to me, since it isn't really a fusional language. In, for instance, Russian and German, whether a noun is masculine, feminine, or neutrum, it affects adjectives and other words in a sentence, as well. In Japanese, "aru" and "iru" don't affect anything else in the sentence, so it doesn't really strike me as a grammatical issue. 惑乱 分からん * \)/ (\ (< \) (2 /) /)/ * 17:43, 20 April 2007 (UTC)

I don't think it fits, since the animate/inanimate distinction isn't a noun class, it's generally a natural distinction. Also, apparently, according to this discussion, "grammatical gender" should apparently be a prominent feature in a language, not an example of a minimal pair. 惑乱 分からん * \)/ (\ (< \) (2 /) /)/ * 18:00, 20 April 2007 (UTC)

If anyone is interested, here's a link to a site that shows what English looked like when it had noun classes/gender marking on nouns. It's a wonderful thing to have information like this available on the internet. Old English Alastair Haines 19:14, 20 April 2007 (UTC)

Yeah, that's what I mean. Old English is clearly a language with grammatical gender. Note how an (often "arbitrarily") assigned gender affects nouns, adjectives and pronouns. (This is also reasonably similar to Icelandic and German). The pdf is nice, but I'd like to see modern English translations on all words, many I can't recognize, though I'm rather familiar with Germanic languages. 惑乱 分からん * \)/ (\ (< \) (2 /) /)/ * 19:46, 20 April 2007 (UTC)

So, in Old English/ Icelandic/ German, every noun has an assigned gender which you must know beforehand. (In Russian, a noun's gender could usually be inferred from it's final sound). In Japanese, the difference is only for a pair of words. By the way, you could usually get around the problem entirely in Japanese, by using the word des(u) (colloquial form: da) instead of aru/iru. 惑乱 分からん * \)/ (\ (< \) (2 /) /)/ * 20:03, 20 April 2007 (UTC)

Thanks for that friend, I'll try it next time I can afford a Bento Box. ;) Alastair Haines 21:11, 20 April 2007 (UTC)

Thank you. This is what I suspected: the iru/aru distinction is optional. That means it's not grammatical gender. True grammatical gender must be mandatorily expressed for the large majority of nouns. FilipeS 22:16, 20 April 2007 (UTC)

This Wiki team-work thing is kinda happenin'. :D Yeah, I agree with you, btw, Filipe. Alastair Haines 09:42, 21 April 2007 (UTC)

That was how I interpreted it, although there possibly might be an "exist"/"be" distinction or similar for the words ("iru" corresponds to "aru", but "desu" carries different connotations). My Japanese is very limited, but it would appear that it isn't a matter of true grammatical gender, but something else. 惑乱 分からん * \)/ (\ (< \) (2 /) /)/ * 11:37, 21 April 2007 (UTC)

Hello again. Here are two tests you can make, to find out whether a certain affix is representing grammatical gender.

1. Does it appear, mandatorily, on large classes of words (hundreds to thousands of them)?
2. In a sentence, if you change one word from one "class" to a different one (such as replacing "man" with "woman", or "person" with "chair"), does that mean you'll have to change other words which qualify that noun, too?

If the answer to both questions is "No", then it's probably not grammatical gender. FilipeS 11:49, 21 April 2007 (UTC)

Well, the test doesn't fit this Japanese distinction. "Iru" and "aru" are not affixes, they are verbs (free morphemes, I think). Second, if you change "person" to "chair" in Japanese, you change "iru" to "aru" (but nothing else), but I'd interpret it rather as a semantical meaning of "aru"/"iru" than grammatical gender. 2nd, fwiw, both Japanese language and Japanese grammar explicitly state that Japanese has no grammatical gender. 惑乱 分からん * \)/ (\ (< \) (2 /) /)/ * 12:00, 21 April 2007 (UTC)

If it's simply a matter of using different verbs when referring to animate and inanimate beings, then it certainly seems it's not grammatical gender. It doesn't even seem to be an inflection. FilipeS 12:07, 21 April 2007 (UTC)

(Using different verbs is exactly what it is.) Yeah, I'd think so, too. But I'm willing to change my mind, if any Ph.d. linguist would claim otherwise. 惑乱 分からん * \)/ (\ (< \) (2 /) /)/ * 12:27, 21 April 2007 (UTC)

Thanks again. Can you tell me whether iru/aru are clitics? It seems to me that speakers of Indo-European languages may have a tendency to mistake clitic free particles for affixes... FilipeS 14:10, 21 April 2007 (UTC)

Sorry, might've been mistaken, but according to Copula#Japanese, desu/da is a copula, while aru and iru are existantial verbs Copula#Existential_usage, as such, they're not completely equal. They don't seem to be any kind of clitics or affixes, though, basically regular verbs just as "be", "exist", "eat" etc. 惑乱 分からん * \)/ (\ (< \) (2 /) /)/ * 15:22, 21 April 2007 (UTC)

Yes, it seems to be as you say. Arigato! :-) FilipeS 18:19, 21 April 2007 (UTC)

Hmmm, upon reading Japanese grammar, I see some other examples where the animate/inanimate distinction would matter, for instance the pluralizing suffix -tachi, although rare to begin with, can't really be used for inanimate objects, and the main reflexive pronoun jibun must refer to an animate subject. I am not an expert on Japanese grammar, though, and it's quite different from the grammatical systems of Indo-European languages I know about, so I'm welcoming a discussion here from someone more knowledgeable on the subject. I put Japanese back among languages without noun classes, but I might have been hasty (although I still think it doesn't have grammatical gender). 惑乱 分からん * \)/ (\ (< \) (2 /) /)/ * 18:57, 21 April 2007 (UTC)

In fact there are various views on -tachi. One is that Japanese has no true suffixes and that tachi is presented as such only from a Western language viewpoint and that tachi is a noun meaning "and others". Likewise there is a view that Japanese (and other Asian languages) do not have pronouns in the Western sense despite the fact that words are presented as pronouns among people of Western language backgrounds. The words referred to thusly as pronouns are stated to in fact also be nouns. This is one reason given why Japanese "pronouns" often change through history though this word category is supposed to be universally stable. Some googling should find decent treatment of these points. — Hippietrail 21:39, 21 April 2007 (UTC)

Yeah, I read Japanese_grammar#Pronouns. So, what do you think about Japanese having or not having noun classes? 惑乱 分からん * \)/ (\ (< \) (2 /) /)/ * 23:46, 21 April 2007 (UTC)

惑乱 分からん * \)/ (\ (< \) (2 /) /)/ *, that iru/aru distinction seems to have some similarities with the ser/estar distinction in Spanish and Portuguese. See Copula and Romance copula. FilipeS 13:38, 28 April 2007 (UTC)

This is rather a late reply, but animacy and gender are different. Japanese has no gender. Gender is a classification of nouns, while animacy is a classification of the referents of nouns. In English, when to use who and which is nothing to do with nouns; it depends on the referents.

• a ruler who listens to his people
• a ruler which can measure in inches and centimeters

On the other hand, the same referent can be referred to by nouns of different genders. For example, in French, un lis (a lily) is masculine while une fleur (a flower) is feminine. A lily is inanimate anyway, whichever noun you use. - TAKASUGI Shinji (talk) 03:36, 17 June 2008 (UTC)

Filipe, you've been around longest, I think. What do you think about archiving some of this talk page? Alastair Haines 14:33, 22 April 2007 (UTC)

Sounds like a good idea, but I don't know how. I've never archived a talk page before... FilipeS 16:14, 22 April 2007 (UTC)

It's very easy, just make a link to a new page. Like this:

Talk:Grammatical gender/Archive 1

Wiki recommend that style of naming for archives.

Now cut and paste anything you like to the archive. :) Alastair Haines 18:12, 22 April 2007 (UTC)

Done. Thanks for the help. :-) FilipeS 20:10, 22 April 2007 (UTC)

In the sense in which the word is used here, shouldn't "referent" be changed to "referend". What knowledge I have of Latin suggests that active-voice "referent" means whatever it is that does the referring (here the pronoun), and that passive-voice "referend" would be the word to use of what is referred to (what in non-technical language I was calling the "reality" until others here advanced my education. Lima 09:01, 23 April 2007 (UTC)

That might be good Latin, but it's not the word that linguists use to refer to, err, referents. Referentially, Drmaik 09:11, 23 April 2007 (UTC)

The word is widely used in Linguistics, Lima. However, I want to know why. All I can think of is examples like, one who receives is a recipient, and one who receives communion is a communicant, one who participates is a participant, so one who recieves reference is a referent. These do not answer your question, though. The usage is whatever it is, but the scientific mind wants to know why. So does mine, I'm researching forthwith. My guess is it is a word-formation rule, like -tion, -ness, -ly, where one part of speech is turned into another, not a morpholical inflection as such. Will let you know. Alastair Haines 12:26, 23 April 2007 (UTC)

Eureka! (Australians love this word Lima, they think we made it up! LoL)

You were right, and so was I. It depends on Latin, but only for the vowel preceding nt.

“

-ent and -ant in English are allomorphs of one morpheme meaning "X does V", where V is the V affixed to; which of the two forms appears when affixed to a given verb depends on a diacritic on the verb, a diacritic which specifies what conjugation class the Latin etymon belonged to.

”

Sean M. Burke, 'English -ent and -ant', 25 March 1996.

This is not peer-reviewed, but I think he's right. Alastair Haines 12:41, 23 April 2007 (UTC)

I am not questioning the fact of the choice of the word "referent", only its rightness.

The quotation concerns the vowel before the -nt (a if derived from a first-conjugation Latin verb; e otherwise - but I think this rule has been disturbed by the passage of the words through French to reach English) not the consonant after the n (t or d).

As you say, with the ending t the meaning should be active: a "participant" is one who participates. So, logically, referent should be someone/something that refers.

"Confirmand" is a word that I do not find in my dictionary, but I know it is used in some circles to mean someone who is to be confirmed (passive voice). Perhaps there are other words of this kind. Or is it just a Latinism?

Only now have I looked up "referent" in a dictionary, and I have found that it gives both the active meaning ("something that refers, especially a linguistic item in its capacity of referring to a meaning") and the passive meaning ("something referred to"). I suppose we must just accept the twofold use as one of many illogicalities of the English language. Lima 13:03, 23 April 2007 (UTC)

Just thought of many more examples: respond-ent, co-respondent, command-ant and, I suspect, adjut-ant. One is reciprocal, the others clearly active, commandant strikingly so. Perhaps referent is an illogical *anti-deponent form as you say. Perhaps the lack of logic is eased a little by being a case of word formation rather than inflection. Alastair Haines 14:02, 23 April 2007 (UTC)

OK, first paragraph complete. Comments? Alastair Haines 06:50, 24 April 2007 (UTC)

Alastair begins his rewriting of "Grammatical gender in English" by mentioning one person's opinion about English having "covert" grammatical gender. The only (part of a) source quoted that I can consult seems (am I wrong?) to indicate that this opinion is not generally accepted. Perhaps mention of Whorf's view should be left until much later.

On reading the section in the new context, I think that a necessary preliminary step is to clarify what will be understood, in this article, by "grammatical gender". How can we seriously discuss whether English (present-day English, that is, since there is no doubt about earlier forms of the language) has grammatical gender without first indicating what we mean by grammatical gender? If we have to conclude that grammatical gender is understood in more than one way, it is still only then that we can speak meaningfully about whether English has grammatical gender in sense A and whether it has grammatical gender in sense B.

The present introduction is problematic. It begins by giving the English he/she distinction as an example of grammatical gender. But it goes on to speak of this as a matter of semantics, of indicating the sex of the referent, what it calls "the gender of the subject").

Sorry for complicating matters. But perhaps we need to clarify the introduction before rewriting the "Grammatical gender in English" section. I admit that other opinions may be much better than mine. Lima 07:16, 24 April 2007 (UTC)

That's very helpful Lima, thank you. Most importantly, I agree the introduction needs work, too. But I'll stick to the current section for the moment.

I think the issue is how to present the discussion of English, equally valid approaches would seem to be:

• History of English language
• History of Analysis of English language
• Logical treatment of issues

Personally, I like any of these when done well. Perhaps I have a preference for the last (though it is not the approach of the four sentences I've provided so far). By logical treatment I mean:

• Definitions relevant to following discussion -- e.g. semantics/sytax grammatical/natural
• Articulation of controversial issue or issues -- e.g. does Modern English have gender in some meaningful way
• Definitions of logically possible views -- e.g. yes, no, some, or something more complicated
• History of development of each school of thought -- e.g. Whorf and others - yes. Current popular concensus - no. Many - some. A few specialists - more complicated (these are probably right, but distinctions are really only relevant to academic discourse).
• History of dialogue between schools of thought -- don't know much about this yet, but it's bound to be intricate.

Although in many cases I would use this approach, I think some of it needs to be covered in other sections of the article; other parts of it go beyond the scope of the article. For example, the English section should not carry the burden of defining semantic/syntactic grammatical/natural. These are relevant to consideration of any language. On the other hand, Development of the schools of thought regarding gender in English, and dialogue between them seem, to me, to lie beyond the scope of the present article. In fact, my User page has a link to Gender in English which I've been wanting to write up at some stage.

That leaves articulation of issue and definition of possible views. Whorf seemed a very concise way of introducing the subject. I deliberately left out things he said that are clearly wrong. Covert is a non-technical word that can concievably cover all possible views. Since Whorf has a tremendous reputation (despite generally being considered wrong), he's a good place to start historically. As far as I'm concerned, the only word he gets to say is covert and a paper by an academic writing outside his field is offered, just in case anyone wants to push Whorf too hard.

What does seem worthy of report is the fine scholarship of Quirk et. al., who present something most native speakers can confirm from their own use, and presents a lot of data in a concise form. It also has the advantage of being descriptive. These features exist. How do we explain them?

Sentence three simply introduces the Encarta quote, which is provided to say all most readers will be interested in -- a concise statement of the issue and the current consensus.

My aim in sub paragraphs under the first would be to address a few thorny issues, either by quoting good contemporary analysis, or by quotes of opposing opinions where no consensus has been reached. I'm sure we'll find examples of both. I hope to be able to find three or four such examples, which would be selected so they demonstrate why the Encarta comment is a simplification, because there are issues which are still tricky and maybe intractable.

I am very far from being a Whorf supporter, but he is painfully relevant to debate regarding gender in English. If we quote him saying something harmless and, in fact, helpful I think it will aid a reader's feeling that we're aware of issues and neutral. To say gender is natural not grammatical, is a form of saying that it is covert. It is not regarding the *covertness of gender with which people take issue with Whorf, it is that he argued that this covert gender was grammatical.

Anyway, I'm very open to changing approach, and moving things around. Though I'd like to get a bit more down before we make a final decision. The only thing I'd be touchy about is deleting sourced information, cause it takes time to locate it, and the article doesn't have many cited sources as it stands. Can we let it grow a little? In fact, I can hardly really move Whorf to a later point, until there is a later point. ;) Alastair Haines 09:28, 24 April 2007 (UTC)

Apologies of my lack of understanding of the Whorf "covert" expression. Unless its significance is spelled out, this expression, it seems to me, might obfuscate rather than assist comprehension of whether English does or does not have grammatical gender (the topic of this section). My comment now is that your elaborate scheme definitely requires an article apart. Within this general article on grammatical gender, the treatment of grammatical gender in English - and it is present-day English that concerns us - should, I think, be much more succinct.

I have a feeling that, especially in the light of the scheme you have presented, I am out of my depth on the question of grammatical gender in general and should quietly withdraw. Lima 11:50, 24 April 2007 (UTC)

I think you could well be right, covert is not a particularly common adjective. If I find something better I will certainly use it. Much more importantly though, as English is only arguably a language with grammatical gender, the sub-section covering it should only be a few paragraphs, justified only because: 1. this is English language Wiki, and 2. because examples from their own language will assist many readers in grasping the issues. In principal, I'd be happy with the four sentences I've provided. Had you considered them adequate I'd have been willing to stop right here, job done. I would very much regret the loss of your involvement Lima. What would you say to you writing the English section here, while I get on with the Gender in English article I was planning to produce anyway? Alastair Haines 12:14, 24 April 2007 (UTC)

I agree with the idea of moving the detailed discussion of gender in English, its historical evolution, and current analytical disputes into a separate article linked from this one. I don't really agree that "Grammatical gender in English" should be the very first section of this article, after the lead; English is a pretty terrible example for introducing people to this topic. We could choose to use English data to illustrate pronoun-antecedent agreement (for the sake of familiarity), but there are no English data available to illustrate dependent-head agreement, which is arguably the more important feature of a full grammatical gender system. I agreed with the previous consensus that English was a "borderline" case. Now the section says that English in fact has nine genders! I understand how Quirk et al. arrived at this monumental conclusion, but I think the more recent approach of the CGEL is more reasonable: treating personal pronoun agreement and who/which selection independently, as reflecting two separate natural gender distinctions (male/female/unsexed vs personal/non-personal). I have provided some primary source text in my userspace, see User:CapnPrep/Gender in English, for those who don't have access to these references. CapnPrep 12:31, 24 April 2007 (UTC)

Thanks heaps for the references! :) Please stay in touch at the Gender in English page. Alastair Haines 22:04, 24 April 2007 (UTC)

Hi. I haven't had the time yet to read the whole discussion, or the changes to the article that Alastair made, but I have been thinking of a rewrite, and I have some suggestions. Basic structure:

• Intro
• Overview
• Gender in English
• Grammatical gender in other languages

Notice that I would add a new section, between the introduction and the section on English, where there would be a short overview of what grammatical gender is, and how it can be recognized. In this overview section, I would start by giving a few examples of gender inflection of adjectives in Old English. With this change, I would also simplify the current introduction, making it shorter, more lightweight and example-based, and less technical. The section on English could point to (but not discuss thoroughly) the issues of suggested gender-bias in language, and the proposals of gender-inclusiveness. I do not think there is reason to start a new article on grammatical gender in modern English, which, as we all seem to agree now, doesn't exist. FilipeS 13:55, 24 April 2007 (UTC)

Sounds like a good structure to me Filipe. I recommend refering discussion of alleged gender bias and gender inclusiveness to any of the many pages that discuss it. Most native speakers have heard it all many times, this article is about the separate issue of grammatical gender, and you are ideal to write it, being bilingual in English and Portuguese. Alastair Haines 22:04, 24 April 2007 (UTC)

I agree fully with CapnPrep, who is far more competent than I am, in particular with a view to preparing a short section on grammatical gender and English, to be placed much lower down (perhaps even to be put as a subsection of a section?).

I agree also with Filipe (another person clearly more competent than I am), except, I think, on the appropriateness of devoting, in an article entitled "Grammatical gender", as much space as he suggests to questions of natural-gender indicators in English. I wonder if it might not be enough to reduce such matters to a simple reference to articles that discuss them. I certainly think the introduction should be lightened, reduced perhaps to little more than a definition, and followed by a section with fuller information. As the article now stands, you have to go down more than a full screen to get to the Contents box. Lima 14:20, 24 April 2007 (UTC)

Looks like we have unanimity, not simply consensus. Alastair Haines 22:04, 24 April 2007 (UTC)

O.K., I've read through the latest discussions, and I looked at Alastair's changes (I also see that he's already started a new article). A few thoughts:

1. I agree that modern English is not the best language to illustrate grammatical gender. On the other hand, how are you going to explain gender to monolingual English speakers? I still think a few carefully selected examples from English can help to ground the discussion, even if it's not a language with grammatical gender in the usual sense. And there are a few examples in English when you can, for instance, see gender agreement at work, such as the ones currently in the introduction. By the way, using English examples of quasi-gender to explain grammatical gender is not as far out as that, considering how many times I've seen English examples of quasi-cases be used to explain grammatical case, which English doesn't really have, either.
2. One thing I've been thinking is that the article could have at or near its beginning some examples of gender agreement from Old English. Something along the lines of these examples here, which currently only show up much farther down the article. That way, there would still be some sort of connection with the English language, so as not to alienate the readers from the topic in the beginning, but without the possibly misleading implication that gender can be fully explained via modern English.
3. Alastair, I appreciate your effort, but this article cannot begin with an esoterical and awfully technical definition of gender, with which most linguists apparently don't even agree! That should come way further down the article, if at all. It's hard enough trying to explain traditional grammatical gender to people whose language doesn't have it, let alone complicated and abstract notions such as Whorf's.
4. As a native speaker of a language with gender, I can't help adding that I find Whorf's analysis very unconvincing, from what I am able to understand of it. Sorry, but saying you've got nine genders just because you distinguish "who" from "which" is — there really is no other way to put it — ridiculous. That doesn't make a new gender; it makes for a footnote!
5. I don't necessarily disagree with the existence of a page for dicussing Gender in English, but such a name might be more appropriate for other topics, such as gender-bias and gender-inclusiveness in language (which I have always felt are separate issues from grammatical gender). Grammatical gender in English, if we may even speak of it, is a very marginal phenomenon — no matter how one chooses to (re)define it, I expect.
6. This article need not say much about English. It does have to dip its toe in it at some point (preferably the start), but that is to anchor the discussion, not because there is a lot to say about gender in English. IMHO, there is very little to discuss, because there's precious little to be discussed: he/she/it, her/his/its, actor/actress, "singular they" versus "he" for generic antecedents, "she"-cars, "it"-dogs, "it"-babies, and that's all folks. Compare that with trying to learn a Latin declension. FilipeS 20:46, 24 April 2007 (UTC)

I didn't put English at the top of the article! Were it up to me, I wouldn't even put English in this article, lol. I was invited to rewrite the English section, and so I wrote four sentences. Hmmm, three, one is a quote from Encarta. Move the section down! I think I suggested this a week ago. ;) Whorf has one word quoted, hard for one word to get complicated. ;) I presume you mean Quirk et. al. and Gender Classes. Those Gender Classes are real, they accurately describe the nature of natural rather than grammatical gender in English. Natural gender is more complicated than grammatical gender to be sure, not simpler. I'm content if it is now abundantly clear why gender in English is such a vexed topic.

Old English is very unfamiliar to modern speakers. I will be introducing it into discussion carefully at Gender in English. I'd recommend you rewrite your English section here, in a very general way. For instance, after discussion of noun classes and languages that have personal/impersonal distinctions. Then you can simply say English has no morphological noun classes but does make natural distinctions regarding pers/imper and sex of referent, which are reflected by lexical selection (e.g. he, she, it, who, which, that) rather than inflection. Cheers. Alastair Haines 22:04, 24 April 2007 (UTC)

Sorry to be so nagging, but shouldn't distinctions of natural gender be independent of language? Anyway, I'll get back to this when I have the time. FilipeS 23:19, 24 April 2007 (UTC)

O.K., here's my tentative rewrite. Please tell me what you think. FilipeS 23:13, 27 April 2007 (UTC)

[Intro -->]

In linguistics, grammatical gender is the systematic expression of gender in words through inflection.

Although all languages are capable of making gender distinctions through lexical means, not all languages have grammatical gender. For example, Old English had grammatical gender, but modern English is normally described as a language that lacks grammatical gender.

In languages where this grammatical category is prominent, there are often some words whose grammatical classification does not coincide with the natural gender of their referent.

Overview

To understand how grammatical gender works, consider an example from Polish. In this language, each noun is classified as either masculine, feminine, or neuter. In their nominative singular forms, Polish nouns are typically feminine if they have the ending -a, neuter when they end with -o, -e, or -ę, and masculine if they have no ending (null morpheme). So, for example encyklopedia "encyclopaedia" is a feminine word, krzesło "chair" is a neuter word, and ręcznik "towel" is a masculine word. When the adjective duży "big" is combined with these nouns, it changes form according to their grammatical gender:

gender	noun	phrase	meaning
feminine	encyklopedia	duża encyklopedia	big encyclopaedia
neuter	krzesło	duże krzesło	big chair
masculine	ręcznik	duży ręcznik	big towel

In a language with grammatical gender, every noun must be assigned to some gender class (or simply "gender"). Many languages have a three-way classification into:

• masculine grammatical gender: includes most words that refer to male human beings;
• feminine grammatical gender: includes most words that refer to female human beings;
• neuter grammatical gender: includes mostly words that do not refer to human males or females.

As the example from Polish shows, the neuter gender normally does not include all nouns that refer to non-human entities. Some of these may be masculine, and others feminine.

Languages can have only a masculine and a feminine gender. In this case, nouns which designate entities with no natural gender, such as objects or abstractions, are distributed among the two genders. In a few other languages still, most words that were formerly masculine or feminine have become indistinguishable with time, producing a new class called the common gender, which however remained distinct from the neuter gender.

• common grammatical gender: includes most words that refer to male or female human beings, but is distinct from the neuter gender.

Even though gender marking is not a significant feature of modern English, it has inherited a few gender distinctions from Old English which may be used to illustrate more clearly what grammatical gender entails. As an example, consider the English sentences below:

John said that he would pay for his own dinner.

Julia said that she would pay for her own dinner.

The gender of the subject is marked both on the personal pronouns ("he"/"she") and on the possessive adjectives ("his"/"her"). Note that this information can be considered redundant in both sentences, since the gender of their subjects is already indicated by the personal names "John" and "Julia".

We can thus formalize the notion of gender as follows: a language has grammatical gender when its nouns are subdivided into morphological classes which correlate with gender, such that:

1. A noun belongs to a single gender class — gender partitions nouns into disjoint classes. (There is sometimes a small number of exceptions whose gender is arbitrary or changes with meaning.)
2. Adjectives, and possibly verbs, have different forms for each gender class, and must be inflected to match the gender of the nouns they refer to — gender is an agreement category.

The correlation between grammatical and natural gender need not be perfect, and it often is not.

Some authors have extended the concept of "grammatical gender" to the expression of other types of natural, individual characteristics (such as animacy) through inflection, although other authors prefer the term "noun classes" when none of the inflections in a language relate to gender. For other authors still, "noun classes" are a separate concept from grammatical genders.

[--> Next, "Gender in English"]

No comment?... Well, these discussions have nonetheless made me think a lot, and reconsider some things. I've come to some interesting conclusions about grammatical gender and so-called "natural" gender. Unfortunately, they might be original research.

Anyway, I've also been looking at previous versions of the article (some of them were clearer than the present one, I am forced to admit), and found the following:

In linguistics, grammatical genders, also called noun classes, are classes of nouns reflected in the behavior of associated words; every noun must belong to one of the classes and there should be very few which belong to several classes at once. (Source of definition: Hockett, 1958, p. 231. See References section.)

Since there is this scholarly definition of grammatical gender, I shall bow down to it. It seems also (if we are to trust Encarta) that today linguists assume that any language that doesn't have grammatical gender must always be talking about natural gender. That seems silly to me, but hey, if it's the scholarly consensus... More later. FilipeS 15:36, 28 April 2007 (UTC)

In the absence of any feedback, I've gone ahead with the rewrite. I hope this version is more consensual. FilipeS 22:59, 28 April 2007 (UTC)

I don't see how that single sentence in the Encarta article can be interpreted in this way. It only describes what happened in the history of English; it doesn't make any claims about other languages or the interaction between grammatical and natural gender in general. Does anyone have access to the Corbett references? He should provide a much better idea of the scholarly consensus. CapnPrep 01:35, 29 April 2007 (UTC)

But it's also what you guys have been saying here in the talk page, isn't it? That when an English speaker points at a car and says "She's a beauty, isn't she?" that's natural gender... FilipeS 12:50, 29 April 2007 (UTC)

This article is a bit long, and it duplicates some examples that are at the Noun class article. To be more exact, I duplicated there some of the examples that were originally here, when Noun class was split from this article. One solution would be to move all the non-Indo-European examples from the "Gender across languages" section to the Noun class article, which is smaller, and leave just Indo-European itself in this one.

However, if the linguists who have studied them prefer the term "gender" to "noun class", then they should stay... FilipeS 19:23, 29 April 2007 (UTC)

Just to say that in Afroasiatic linguistics, people talk about gender rather than noun classes, whereas it's normally the opposite for Niger-Congo (with good reason). All for now. Drmaik 19:48, 29 April 2007 (UTC)

Great, but what about:

• Algonquian languages
• Athabaskan languages
• Australian aboriginal languages
• Caucasian languages
• Zande

Most of the examples in this page have something to do with gender (in strict sense), but there's also something else at work in all of them. Animacy, in particular, appears quite often. FilipeS 20:21, 29 April 2007 (UTC)

Afro-asiatic languages like Akkadian, Egyptian and Hebrew, which I study all have gender, quite similar to Indo-European. When I say gender, that is the term that is used by academics in our disciplines. We would all agree that noun-class is an excellent term when comparing languages over wider areas and longer periods of time. It applies to our own area too. However, when people learn Egyptian, they often want to read heiroglyphic texts, without thinking about contrasts between Egyptian and other languages, so we just teach them masculine and feminine, and keep it simple.

You are doing excellent work Filipe. I think what we need to clarify is the difference between noun class and gender. That should help you make your decision. All gender systems are noun class systems, but not all noun class systems are gender systems.

Many languages inflect nouns for their function in sentences -- subject, object, egative, absolute, dative, instrumental, locative, ablative, etc. I imagine this is done in some languages that do not have adjectival inflection, or gender distinction in pronouns. This would result in declentions that could not properly be associated with genders, but would still be legitimate noun "classes".

I'm afraid I do not know if such languages exist, as likely as it would seem to be that they would. However, if you can find them, these would be the perfect languages for the Noun class article. Any language with noun classes but no gender system goes there. In Grammatical gender you have all the languages where adjectival inflections and distinct pronouns have conventional patterns of agreement with the noun classes.

This actually leaves us with only one other kind of language left over -- languages without noun classes or grammatical gender. These are the "natural gender" languages. Perhaps it might be worth having a small category and template to help underline the logic. Check the Linguistic typology article, there may be a way of integrating in with their systems of classifying languages. Time Manner Place / Place Manner Time. Subject Verb Object, VSO, etc. etc. Alastair Haines 19:53, 29 April 2007 (UTC)

Thank you for your kind words, Alastair. I apologize if the harshness of my earlier criticisms to your previous work offended you. I hope you understand nonetheless that it wouldn't be very friendly towards our readers to start the article with a gender classification as abstract as Quirk's.

Right now, I'm actually feeling very tempted to move everything that doesn't fall into a traditional masculine/feminine/neuter, masculine/feminine, or common/neuter paradigm into the Noun class article (or the Animacy article, in a few cases, perhaps). But I really don't know what is more common in the linguistic literature. That's what should matter. As for languages that don't have grammatical gender, what is there to say about gender in them? Isn't it generally a straightforward matter? And, if not, how about using your new article Gender in English to practice how to discuss those other classifications, before attempting to generalize?... Regards. FilipeS 20:29, 29 April 2007 (UTC)

There is a sense in which all linguistic agreement is redundancy, but this is only true if we can assume that the grammatical link between the agreement source and target is already independently established. There are two distinct manifestations of this in the English examples in the "Overview" section:

John insisted that he would pay for his own dinner.

Jane insisted that she would pay for her own dinner.

In both sentences, the reflexive possessive is syntactically required to refer to the subject of the embedded clause: his own → he and her own → she. It is ungrammatical to say "*He would pay for her own dinner" and vice versa. In this case, the gender marking on the possessive is indeed redundant, since we don't need it to determine the pronoun-antecedent relationship; it only reinforces it.

On the other hand, there is no obligatory link between the two subjects in these sentences. It is grammatical to say "John insisted that she would pay for her own dinner" and "Jane insisted that he would …" since John and Jane can be talking about a third person. In the examples above, the gender of the pronoun is not redundant. It provides information that helps narrow down the range of possible antecedents of he and she. A natural interpretation of these examples is that John and Jane are talking about themselves, but this is not obligatory. They can still be talking about a third person, and all we know is their gender (and the fact that it's one person, and the fact that's it's not you or me) thanks to the pronoun.

The fact that some languages (Finnish, Turkish, Chinese, etc.) get by with systematic ambiguity does not mean that every other language is always redundant. There is a middle ground.

CapnPrep 12:05, 30 April 2007 (UTC)

All languages have ambiguities. However, you main point is well taken. FilipeS 12:16, 30 April 2007 (UTC)

Why did you delete the example stating that there is no gender agreement in constructed and auxiliary languages, and replace it with one claiming that it's just like in English, CapnPrep? FilipeS 19:44, 1 May 2007 (UTC)

I just rewrote this paragraph introducing the "Constructed languages" section:

In natural languages, gender inflections in nouns are normally accompanied by gender agreement in their modifiers. Even in English, where gender marking is scarce, there must be agreement between the possessive adjectives and their antecedents. "*Jane hurt his leg" and "*John broke her arm" are ungrammatical (or derogatory), if "his" and "her" refer to Jane and John, respectively. Some constructed languages, however, have gender inflection without gender agreement.

The phrasing of this text indicates a mix-up concerning possessive pronouns that appears in a few other places in the article. The choice of his in "John broke his arm" is presented as an instance of modifier agreement. But there is no modifier agreement in English. The possessive form his is indeed a modifier, but it modifies the noun arm, and there is clearly no agreement between these two terms. His is also a pronominal element, and its antecedent is John. There is agreement here. We find exactly the same situation in most constructed languages, so the contrast alluded to in the original text is non-existent!

There are two aspects of gender agreement: modifier agreement (generally marked inflectionally on the modifier) and pronoun-antecedent agreement (generally determining the choice of pronoun). With ordinary adjectives, we only have to worry about the first aspect, and with ordinary pronouns (he, she, it), we only have to worry about the second aspect. With possessives, both aspects are potentially relevant, so we have four possibilities (only talking about 3rd person singular forms here):

• pronoun-antecedent agreement but no modifier agreement

English: "John and his son/daughter" – "Jane and her son/daughter" (gender of what is being modified doesn't matter)

• modifier agreement but no pronoun-antecedent agreement

French: "John/Jane et son fils" – "John/Jane et sa fille" (gender of antecedent doesn't matter)

• pronoun-antecedent agreement and modifier agreement

German: "John und sein Sohn" – "John und seine Tochter" – "Jane und ihr Sohn" – "Jane und ihre Tochter" (gender of everything matters)

• neither kind of agreement

Mandarin: "John/Jane hàn tāde érzi/nüer" (nothing matters)

The distinction between these two aspects of gender agreement is blurred in a few places in the article, leading to confusion.

CapnPrep 20:14, 1 May 2007 (UTC)

CapnPrep, the text clearly says "if "his" and "her" refer to Jane and John, respectively". So I'm not sure what you're talking about, exactly. Obviously, the possessive doesn't have to refer to the subject of the sentence. That's beside the point. FilipeS 20:20, 1 May 2007 (UTC)

I'm not saying that the English examples are wrong, but I don't see how they are relevant in this section. Most of the constructed languages in the list have the same system as English (so we can't say that they "have gender inflection without gender agreement"). In Esperanto, for instance, you would say "Mi rompis lian brakon" (I broke his arm) and "Mi rompis ŝian brakon" (I broke her arm). This looks pretty much exactly like English: we choose li/lia to talk about a male person and ŝi/ŝia to talk about a female person, and we don't care what the gender of the modified word is (here, brako "arm", inanimate). CapnPrep 20:43, 1 May 2007 (UTC)

But you do care about the gender of the possessor of the arm. A Finnish speaker wouldn't.

Well, I don't know about Esperanto, but there's a crucial difference between Ido and English: in the former, gender-specific pronouns are completely optional. You never have to state whether you're talking about a man or a woman, if you don't wish to. In English, this is not so. FilipeS 21:18, 1 May 2007 (UTC)

--- Since when is the "his" in "John broke his arm" considered a pronoun?

See personal pronoun. FilipeS (talk) 23:43, 2 January 2008 (UTC)

You are being a victim of native language bias. Because your language has separate words for "his" and "her", you think that "his" and "her" are two necessarily separate concepts.

But in some languages there is only one word to say both "his" and "her". Examples are available in the article.

Having said this, I have convinced myself that CapnPrep had a point in his criticism. The possessive pronouns, because they really refer to two nouns (the possessor and the possessum) are not the clearest examples of gender agreement out there. Unfortunately, in English there are no better examples. FilipeS (talk) 21:13, 8 January 2008 (UTC)

I don't understand the point of that distinction. I could say the same about, say, the Spanish words niño and niña. Does that mean that they are not a masculine and a feminine noun, respectively? Of course they are! FilipeS (talk) 22:47, 8 January 2008 (UTC)

I get the impression that you two are repeating to me some linguistic definition of "grammatical gender". That's fine, but let me tell you that that's not what I understand as "grammatical gender" -- and my native language has it. To me, grammatical gender is not independent of the gender of the referent.

I don't know how standard that definition you guys are using is, but I'll come right and say that I don't like it. It makes an artificial distinction between gender in nouns and in pronouns, and gender in other parts of speech, when the two are so clearly identified morphologically. Just look at the affixes that characterize each gender: they are exactly the same, typically, regardless of whether the word is a noun, a pronoun, an adjective, or what have you. FilipeS (talk) 14:16, 9 January 2008 (UTC)

If the word is an adjective or a determiner, then it is indeed independent, in the sense that the gender of such modifiers is completely determined by that of their antecedents (but see Synesis!)

But if the word in question is a noun or a pronoun, like niño and niña, then its gender will have a high statistical correlation with the gender of the referent, exceptions like persona notwithstanding.

What is your source? FilipeS (talk) 20:13, 9 January 2008 (UTC)

My source? That's like asking me for my source after I've told you it's snowing outside and then when you didn't believe me I took you to the window and showed you the snow falling. I've just devoted a bunch of time to demonstrating this to you by example and reason and as far as I can tell the only problem is that you don't like the result. I've pointed out that grammar doesn't apply to objects (which should be patently obvious, since grammar is a linguistic concept), and I've explained so that it should be clear to anyone that sometimes agreement is with a word's gender and sometimes agreement is with a thing's gender, and the two are different concepts. If you want to go ahead and pretend they are the same concept, and dismiss the examples that demonstrate the real differences as "exceptions" that can be ignored, there's nothing I can do about it. —Largo Plazo (talk) 20:59, 9 January 2008 (UTC)

I thought so. Look, Largoplazo, your definition does not agree with the ones used by the sources used in the article, for example SIL. Unless you can come up with an alternative and authoritative source that uses your definition, I'm afraid this discussion is pointless.

I notice that your native language is one without grammatical gender. Perhaps you should be more humble about your understanding of the concept. FilipeS (talk) 16:33, 10 January 2008 (UTC)

LOL. I'm familiar with people who assume that because their native language does have a feature, that makes them experts on the subject. It's like saying a person who is deaf inherently knows more about the physiology of deafness than a hearing audiologist or otologist. Try some humility yourself. This is a technical discussion, not a contest as to who has a better command of Portuguese.

If after reading the first sentence of your source at SIL you still think the page is an authoritative reference, you need to think again:

Grammatical gender is a noun class system, composed of two or three classes, whose nouns that have human male and female referents tend to be in separate classes. Other nouns that are classified in the same way in the language may not be classed by any correlation with natural sex distinctions.

First, there is nothing that limits gender to two or three classes. Many of the world's language have more. Second, "gender", from the Latin "generis" meaning "type" or "kind", does not in principle mean "sex". While there are languages where two of the genders do have a partial or complete correspondence to natural sex in the case of nouns referring to sex-distinct beings, there are many where none of the genders have any relation to sex at all. (Perhaps you didn't have the humility to conceive that grammatical gender might not operate in all languages that possess it the way it operates in Portuguese.) The only reason "gender" has come to be used in English as a synonym for "sex" (as when we speak of a person's gender) is because its use to describe a feature of foreign languages English-speaking people were likely to be familiar with (Greek, Latin, French, German, Spanish, Russian) does, in those languages, have a connection to actual sex. Anyway, the page was written by someone with a clearly incomplete understanding of the topic. The page is to be dismissed out of hand and the citation should be removed from the article.

—Largo Plazo (talk) 17:08, 10 January 2008 (UTC)

I don't know how I overlooked this before. I entered this discussion by showing that English nouns don't have grammatical gender, and that the use of "his" and "her" relate to natural gender of the referent. You have been arguing with me over this ever since—and now you suddenly say that maybe I don't know what I'm talking about because, as you wrote, "I notice that your native language is one without grammatical gender." Well, that's what I said to begin with, so now I think that you don't even know what you've been arguing about all this time. Are we done?

—Largo Plazo (talk) 18:53, 10 January 2008 (UTC)

Am I misunderstanding something? The nouns actor and actress don't inflect differently because of gender. They inflect differently because actor ends with an "r" and actress ends with an "s" just as the words actor and class. 66.171.76.138 21:17, 21 May 2007 (UTC)

See Inflection. FilipeS 21:48, 21 May 2007 (UTC)

Sorry if this has been brought up already, but this is a bit unclear:

Some Slavic languages, including Russian and Czech, make grammatical distinctions between animate and inanimate nouns (in Czech only in the masculine gender; in Russian only in masculine singular, but in the plural in all genders).

Russian distinguishes between animate and inanimate nouns only in one case (the accusative). In the accusative case, animate masculine singular nouns and animate plural nouns in all genders are declined as identical to the genitive. In the other 5 cases, there is no differentiation for animacy whatsoever.

In short: maybe the entry could be changed to "in Russian, only in the masculine singular, and the plural in all genders, in the accusative case." (citation: Wade, Terence. A Comprehensive Russian Grammar, 2000, p. 68-9)

I am happy to go into more detail if anyone gets the urge.  :)

--83.167.112.146 18:02, 29 May 2007 (UTC) jkh

Why not be bold and give a shot at rewriting it? I'm sure you understand the matter better than I. ;-) FilipeS 18:09, 29 May 2007 (UTC)

P.S. Read this. It might inspire you. FilipeS 18:16, 29 May 2007 (UTC)

In any dictionary a definition pertaining to grammar will be offered as the primary definition of gender. Why should wikipedia be any different? The value of a word isn't just to be judged on how much ink has been spilled about it in modern times. This article should be called Gender, and the article currently being called Gender should be given then name Gender and Sex/Gender Studies or whatever.

But this is the primary definition and deserves the title "Gender".—The preceding unsigned comment was added by 220.253.65.161 (talk • contribs) 10:56, 8 June 2007 (UTC).

See Gender. FilipeS 12:15, 22 October 2007 (UTC)

I'm sorry, but the references to languages without pronouns corresponding exactly to English "he" and "she" in this discussion leave me very confused. I had thought, and perhaps it's just my ignorance, that grammatical gender referred to how words inflected, not to how specific the vocabulary was. "Waitress" and "waiter" refer to distinct entities (distinct in the reference frame of that particular language at a particular time, that is) while, in contrast, in Spanish there is no "meso" - no "male" form of a table - to go along with the feminine "mesa" - ie, grammatical gender has nothing to do with sex-assignment - although there could be some overlap. English is clearly "gender-free" in that we don't assign words to inflectional categories based on arbitrary gender assignment. But the fact that a language distinguishes between goose and gander, say, has nothing to do with it.

I think you've understood correctly. Yes, English does not have noun genders (even though it has some gender specific words like goose, gander, actrice, actor). Yes, Spanish has noun genders (because all nouns are either male or female and they take the corresponding articles, such as "la" and "una" for female words like mesa and "el" and "un" for male words like barco). --Gronky 21:28, 10 October 2007 (UTC)

Grammatical gender is about morphology, but it's not just about morphology. In Spanish, the words for "boy" and "girl" are niño and niña; the switch -o/-a is clearly a gender inflection. And on what grounds do you claim that "boy" and "girl", or "waiter" and "waitress", or "he" and "she", are "different entities"? Many languages have only one word for each pair (boy/girl, waiter/waitress, he/she) -- to them, it's just one concept. FilipeS 12:20, 22 October 2007 (UTC)

My Esperanto is rusty at best, but I'm not sure the summary here is entirely accurate. I think, ideally in theory, Esperanto words are gender neutral, so "patro" simple means "parent", and "patrino" means "female parent (aka mother)" and "virpatro" means "male parent (aka father). I can't remember if this is used to for humans, but I know its definitely used for animals. So for example, "bovo" is "cow", "virbovo" is then "bull" and "bovino" is then "heifer". I can't remember if this was part of Zamenhof's original plan for Esperanto, or if gender neutral was implicitly male and female was the only gender specified, and the "vir-" prefix was a later development meant either for equal rights or simply clarity. The latter would explain the odd combination of female suffix, male prefix. Anyway, I clearly remember learning this distinction in my Esperanto primer and it would be wonderful if somehow who knows Esperanto better than myself could add another line explaining the proper use of the "vir-" prefix. —Preceding unsigned comment added by 71.233.244.32 (talk • contribs) 21:02, 10 Oct 2007 (UTC)

Most Esperanto words are gender-neutral, but twenty words are unmarked masculine, such as patro (father) and frato (brother). See also Riism. - TAKASUGI Shinji (talk) 18:19, 5 January 2008 (UTC)

Can we get rid of that old non-NPOV tag please? FilipeS 12:22, 22 October 2007 (UTC)

Thanks. FilipeS (talk) 21:17, 8 January 2008 (UTC)

Other than English, which other Indo-European langauges omit the grammatical gender distinction? I've been told Bangali, also doesn't use gender distinction. Which other IE langauges don't divide grammatical gender? Le Anh-Huy (talk) 11:52, 25 November 2007 (UTC)

There's a non-exhaustive list in the article Noun class. FilipeS (talk) 23:41, 2 January 2008 (UTC)

On 16:57, 27 April 2008, FilipeS deleted this sentence

The absence of grammatical gender is unusual for an Indo-European language (such languages include Persian and Armenian, both of which lost their gender marker through the influence of the neighboring Turkish language), though common in other language families.

with the comment "Better to just scrap the whole comment, since Indo-Iranian languages don't have gender, and they make up a large branch of IE, perhaps the largest". That is both inaccurate and irrelevant, and I have restored the sentence. My reasons:

1. While Persian and Bengali have no grammatical gender, other Indo-Iranian languages do: Panjabi and Hindi-Urdu, to name just two.
2. Comparative linguistics is not a matter of "*One language, one vote". While a majority of Indo-European languages are Indo-Iranian (Ethnologue Indo-European family tree), that is just one of nine branches of Indo-European. And only one of those branches -- Armenian, which forms a branch of its own -- has no gender. Among the other seven branches,

• Welsh (Celtic)
• Greek
• Russian (Slavic)
• German (Germanic)
• Albanian
• Latvian (Baltic)
• French (Italic)

all have grammatical gender. --Thnidu (talk) 23:28, 27 April 2008 (UTC)

Be that as it may, the claim that English is unique among the IE languages for not having an active system of grammatical gender is untenable. Even within the Germanic branch, there is a clear and widespread tendency to lose grammatical gender inflections. FilipeS (talk) 13:39, 28 April 2008 (UTC)

While grammatically correct, the French examples of gender agreement Il est un grand acteur / Elle est une grande actrice strike me as barely acceptable as a native speaker, even in stilted language : it rather looks like a calque translation from English. French does not use personal pronouns in that kind of sentence, but a set phrase c'est, literally this is (plural ce sont, but colloquially the singular may be used regardless of the situation), classified in a class of "presentative" structures in modern French grammars. Actually, il est un grand acteur could even be interpreted as meaning "There is a great actor", since il est can be another of these (it is a now literary variant of il y a).

Accordingly, I modified these sentences to Lui, c'est un grand acteur / Elle, c'est une grande actrice, so as to keep an example of personal pronoun agreement. They mean "He is a great actor" and "She is a great actress" with emphasis - in this context, it points towards a contrast with other people that would not be regarded as good.

Regards, Bertrand Bellet, 17:54, 27 February 2008 (UTC)

Thank you. FilipeS (talk) 15:36, 29 February 2008 (UTC)

Shouldn't this be added as a subcategory in the noun class article? —Preceding unsigned comment added by 71.188.247.45 (talk) 17:13, 15 June 2008 (UTC)

Some folks argue that noun classes and genders are "completely different things". See the archived discussions. FilipeS (talk) 16:17, 17 June 2008 (UTC)

Some loanwords inflect according to gender, such as actor/actress, where the suffix -or denotes the masculine, and the suffix -ress denotes the feminine.

Actor doesn't inflect because it's a loanword. It came into the language in the 14th century, but wasn't so inflected until the 17th, perhaps because the profession was formerly limited to males and female actors were a novelty at the time. See http://www.etymonline.com/index.php?term=actor

I'm hard-pressed to think of any valid examples for the stated claim, which is in any event unreferenced. 192.31.106.35 (talk) 00:57, 4 September 2008 (UTC)

Just because it's a loanword doesn't mean it isn't inflected. You even make a strong case for it being inflected by stating that it once was not inflected, but that the language's own devices caused it to gain gender centuries after its borrowing. LokiClock (talk) 15:45, 25 November 2009 (UTC)

[1] contradicts Proto-Indo-European noun and related articles. So either it is a faulty source, or the articles are incorrect. Shinobu (talk) 15:13, 11 September 2008 (UTC)

Sting's song Fields of Gold contains the phrase "the sun in his jealous sky". Is that a peculiarity of this song, or would English poets typically assign a male gender to the sun? AxelBoldt (talk) 01:09, 24 December 2008 (UTC)

Whether Sting was creating his own metaphor from scratch or was intentionally echoing any earlier tradition, I couldn't say. Technically, it's off-topic here because it's an example of anthropomorphism, a separate thing from grammatical gender. —Largo Plazo (talk) 01:46, 24 December 2008 (UTC)

Throughout anchient history/mythologies, the sun was always portrayed as masculine, and the moon was portrayed as feminine. This is reflected in many laguages around the world that do assign gender to inanimate objects (e.g el sol = the sun; la luna = the moon). This is an archetype. —Preceding unsigned comment added by 74.167.201.232 (talk) 06:43, 1 June 2009 (UTC)

There are many mythologies were the sun is female and the moon masculine, including the old germanic. In germanic languages that differ between male and female gender, like Old English and Modern German, moon is masculinum and sun is femininum. --195.0.221.197 (talk) 15:43, 4 July 2009 (UTC)

And in most Slavic languages the Sun is neuter... in Latvian, it's feminine, and in Greek masculine... dnik ► 12:58, 13 October 2009 (UTC)

The article states Swedish has (or had before common gender) three genders: masculine, feminine and neuter. But why isn't the fourth listed? Reale (words that have the same prefix as mask/fem words but are inanimate). Shouldn't Swedish be listed under langs with more than three genders? ProbablyX (talk) 00:09, 18 March 2009 (UTC)

This is an archive of past discussions about Grammatical gender. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.

Archive 1

Archive 2

Archive 3

Archive 4