LINGUIST List 7.613

Wed Apr 24 1996

Disc: Grammatical Gender

Editor for this issue: Ljuba Veselinova <>


  1. Alexis Manaster Ramer, Re: 7.592, Disc: Grammatical Gender
  2., Disc: Grammatical Gender
  3. "Douglas Moore", Disc: Theory of Gender
  4. David Costa, gender in Alngonquian
  5. KNAPPENVKPMZD.kph.Uni-Mainz.DE, Re: 7.592, Disc: Grammatical Gender
  6. Karl Teeter, Grammatical gender

Message 1: Re: 7.592, Disc: Grammatical Gender

Date: Mon, 22 Apr 1996 01:13:57 EDT
From: Alexis Manaster Ramer <amrCS.Wayne.EDU>
Subject: Re: 7.592, Disc: Grammatical Gender
Here is a simple example of a rule about gender which holds good in a
number of languages, incl. Polish, Russian, German, French, Spanish,
Hebrew, and I am sure many others:
If a language has a pair of nouns X and X-y (where X-y is derived from
X by the addition of some kind of derivational morphology, not
necessarily a suffix) and where the meanings of X and X-y differ only
in that one denotes females and the other males of some category of
human beings, then absolutely positively never will you find that the
one denoting men is grammatically feminine and the one denoting women
is grammatical masculine.
There are other rules, some of them language-specific, and the point
that should I think be acceptable to everyone is that gender of nouns
denoting persons in all these languages and many others is
overwhelmingly rule-governed, even if there are occasional exceptions.
The rule just stated has no exceptions as far as I can tell, and it
alone (not to mention the other rules, some of which I have already
mentioneed in earlier postings and most of which are to be found in
the linguistic and grammatical literature) are ample reason for
calling the masc gender masc and the feminine feminine. The Greeks
were not completely crazy when they came up with these terms (and as
far as I can see my rule applies to Greek and Latin as well).
Alexis MR
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue

Message 2: Disc: Grammatical Gender

Date: Mon, 22 Apr 1996 13:26:14 -0000
From: <>
Subject: Disc: Grammatical Gender

While it is true that spoken Chinese does not distinguish gender in
third person pronouns, Modern written Chinese HAS introduced a
feminine WRITTEN form and a neutral WRITTEN form to distinguish
masculine from feminine from neuter 3P. In this case it might actully
be a case of feminism--because the use of the feminine written form of
the 3P would serve specifically to signal the presence of a woman,
whereas this presence did not exist in any written form of the 3P in
the past. All three 3P are pronounced identically, however. The
assignation of the written form would correspond more to logic (that
is, biological gender in the case of masculine and feminine; neuter in
the case of non-sexual referents).

Sean Golden
Facultat de Traduccio
Universitat Autonoma de Barcelona
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue

Message 3: Disc: Theory of Gender

Date: Tue, 23 Apr 1996 09:59:37 -0000
From: "Douglas Moore" <>
Subject: Disc: Theory of Gender
 Corrections and Clarifications

As some members of the list have pointed out to me privately, my
previous post needs some corrections and clarifications.


It appears that I've lived a bit too long away from France. Stavros
Macrakis <> points that I made the blunder of mixing
up the gender of Canada in French. It should have been around the
other way.
Canada is masculine in French, not feminine..

Also, Stavros Macrakis points out another error:
"You say:
 [French], unlike any of the other Latin languages, the gender
 construct is rather universal in scope and even applies to the
 proper names of countries. Most other languages merely refer to the
 names of countries as always having neuter or, at most, masculine
Except for Italian and Spanish, where most countries are feminine (la
Spagna, la Grecia, la Francia, ...) although some are masculine (il
Marocco, l'Egitto. In Greek (not a Romance language), countries have
gender: i Gallia (fem), to Velgio (neut)."

As for the velo and bicyclette example, we had better simply erase
that example judging from Stavros Macrakis objections.


In a further correspondance with Macrakis I wrote:

I'm particularly struck by the examples 

C'est une femme. (it's a woman)
C'est sa femme. (it's his wife)

C'est une fille (it's a girl)
C'est sa fille (it's his daughter)

compared with the masculine variants

C'est un homme (it's a man)
C'est son mari (it's her husband)

C'est un garcon (it's a boy)
C'est son fils (it's his son)
C'est son garcon (it's his boy)

Here grammatical gender accords with sexualisation but ignoring that
aspect one sees the feminine noun playing the role of a much more
"blanket" category than the masculine terminology which is much more
context dependant. It roughly goes with the ide of the feminine being
what it is but has an attribute (un husband, a father...), whilst the
masculine is an attribute and - as word - must change for the
different things to which it is an attribute of.

Stavros Macrakis replied:

< I'm not sure what this proves [etc.. see below]

In response I wrote:

I have to preface my response with a little more clarification of my

My approach to all of this is to start from a non linguistic base
where gender is a generic systemic construct - something like an
ontological or epistemological form of typing without any meta-theory.
In brief, I'm after a new kind of mathematics from which I can
bootstrap up to a generic system with its corresponding language I
spoke of -the language that I claim will turn out to be none other
than the genetic code. The natural languages are postulated to be
specialised derivatives of this generic language. It's been a long
haul just to get to where I'm at now. To complete the story ,well, who
knows. ..

The first step in this bootstrapping, constructionist approach is to
get a formal handle on the starting point. In this regard my gender
construct is crucial and enables me to start elaborating a theory of
relative types. The two fundamental types are M and F which I
symbolise with the 'M' and '#' tokens. I'll use the tokens M and F

>From these two primitives four fundamental forms naturally emerge by
taking the four possible binary combinations MF, FM, MM and FF. I
claim that these are the four "letters" of the generic language and in
fact match up with the four G,A, T and C letters of the genetic code.
It's up to me to make this into a rigorous, formal system and prove
this match. I'm working flat out on it. It's not easy.

Strictly speaking, it is quite premature for me to do what I've done
in the mailing list, and attempt to project these very low level
formal constructs into the domain of natural language. Nevertheless,
this is no more presumptuous than what others are doing, working from
no formal constructs whatsoever except the empirical methodology of
descriptive linguistics.

Thus, throwing all caution to the wind, I jump in head first and
postulate the following.

1. Grammatical gender is a surface manifestation of this deep inner
 mechanism which is at work in all languages.
2. A language like French manifests this as a binary gender opposition
 of the purest kind, without a neuter - an opposition between pure M
 and F
3. Other languages, like German and Greek also have a neuter. In this
 case the noun typing is along the lines of MF for masculine, FM for
 feminine, and MM for neuter. 

In the case of a 2G structure like French, the gender typing is
decidably less literal than in the 3G languages. We could say that
French employs _non_literal_ gender typing whilst the others use a
_literal_ form of gender typing. The terms literal and non literal are
used here in some kind of technical sense and shouldn't be taken too
... literally. What is at work here is an algebra of literals and non

One way of thinking about the three types is to think of the literal
masculine type MF corresponding to something like "the object as
subject" The literal feminine FM would be the "subject as object" and
the literal neuter MM corresponding to "object as object".

It appears, from what has been said on the mailing list that Russian
has a fourth gender type. This may be the allusive FF type. According
to my theory, such a type could be intuitively understood as "subject
as subject"

>From an epistemological point of view, all our present day sciences
are uniquely directed at a knowledge of the object (ontologies), never
the subject. The latter form of knowledge has been coined as
"gnoseology." Curiously it has really only been the Russians and Poles
who have made a name for themselves in this latter domain. Whatever
the case may be, the construct "subject as subject" is quite an
allusive one.

Anyway, having said all this, I can at last say something in response
to your last post. 

[Stavros Macrakis]
> I'm not sure what this proves. In English, after all, "She's my
> girl" (i.e. girlfriend) works, but "He's my boy" doesn't, although
> both work in the sense "child". 

Just a cursory remark. "She's my woman" and "He's my man" are
symmetrical. A parent (of either sex) could declare "He's my boy." I
don't think that this leads anywhere though. What matters is that
modern English is practically devoid of non-sexual grammatical gender.
According to my theory, such a language is one which has somehow opted
to specialise in the domain of "objects as objects" and to hell with
the subject and all of the messy dialectical ramifications that result
from such a complicated (European) problematic. There would be a
downside of such a specialisation, but the upside might
 be what we witness today, English is the lingua franca of the
 scientific world - a world populated exclusively by "objects as
If you want to be objective, just exclude the subject - repress the
feminine, and stick to literals.

Hmm a bit provocative, all that, but it does make some sense. Maybe.
[Stavros Macrakis] > In Modern Greek, "andras" and "yineka" mean >
man/husband and woman/wife. Does that mean that male/female relations
> in Greece are more symmetric than in France? 

French is 2G and Greek is 3G. I think this is comparing apples
to oranges. Since French employs non-literal gender typing, I would
expect the natural assymetry of the F and M to be reflected in the
grammar. In the case of Greek, the feminine and masculine are not pure
M and F but compound MF and FM types and so perhaps one could expect a
symmetry based on this kind of algebra. 

As to male/female relations versus masculine/feminine oppositions, we
see in French a coincidence of these two forms. Literally projecting
the masculine/feminine construct onto male/female relations would be
unsound. However, it does provide an open invitation for anyone of
"macho" tendencies and a corresponding insensitivity to subtle forms
of reasoning to abuse this rule. .

In English - the language of literals and "objects as objects" - there
is no danger of this happening. The danger is more perverse. There the
tendency may be to see the world as literally a land of literals and
objects as objects. This kind of perversion goes way beyond the
concerns of the PC people.

> In Modern Greek, "andras" and "yineka" mean
> man/husband and woman/wife. Does that mean that male/female
> relations in Greece are more symmetric than in France? In Ancient
> Greek, kouros/koree meant boy/girl, and son/daughter; but in Modern
> Greek, kori means "daughter" and "koritsi" (a diminutive of kori)
> means girl or daughter, and agori means "boy/son", while yos means
> "son". What can you conclude from all this?

Response provided above with the usual disclaimer that this material
is provided "as is" and no responsibility is taken for any damage
resulting from its use .. etc. !

Douglas Moore

Dr Douglas Moore mailto:
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue

Message 4: gender in Alngonquian

Date: Tue, 23 Apr 1996 08:04:45 PDT
From: David Costa <>
Subject: gender in Alngonquian

This is a response to Dan Alford's posting of 19 Apr 1996.

David Costa
UC Berkeley


> Before the invasion of the Americas 500 years ago, many language
> families (most specifically Algonkian) had no Sex differentiations
> whatsoever, according to my Native consultants. That is, no distinctions
> like man/woman, boy/girl, whose distinction rested on Sex. I know this is
> difficult to accept for many scholars, and yet the people who know insist
> this is correct. Even the bastard English term 'squaw', said to have come
> from Algonkian languages, was built on the verb 'entering', and not a
> sexual gender distinction per se in Algonkian languages. 

There is absolutely no linguistic evidence to support the bizarre
notion, which seems to be advocated here, that Algonquians' languages
did not distinguish between women and men until white contact somehow
forced it on them (I would like to hear an explanation of why contact
with Europeans would do this). For starters, there is a very large
amount of completely reconstructible Proto-Algonquian vocabulary which
explicitly makes reference to the distinction between female and
male. A few such terms are *elenyiwa 'man', *e0kwe:wa 'woman' ('0' =
theta), *metemwehsa 'old woman', *we?$i- 'to get married (of a
woman)', *nekwihsa 'my son', *neta:n(ehs)a 'my daughter', *no:h0a 'my
father', *nekya 'my mother', *-i:w- 'wife', and *-a:pe-
'maleness'. Several more could be added to this list. In addition, an
entire kinship system of Proto-Algonquian can be reconstructed which
makes constant reference to male versus female: terms that distinguish
male siblings from female ones, terms which specifically indicate
siblings or in-laws of the opposite sex of oneself, and different
terms for 'grandmother' as opposed to 'grandfather' (and no term for
'grandparent'). Additionally, the loan into English 'squaw' has long
been known to represent the completely phonologically regular
development of the *e0kwe:wa 'woman' term in the Algonquian languages
of southern New England, which predictably appears as /skwa:/ in
languages such as Massachusett and Narragansett.

These Proto-Algonquian etyma are found throughout the family, in forms
regularly derivable through regular sound law, and as widely dispersed
geographically as Arapaho and Cheyenne in the Great Plains, to Ojibwe
and Cree in Subarctic Canada, to Delaware, Massachusett and Micmac on
the Atlantic coast. If the Algonquian languages did not develop terms
to distinguish men from women until white contact, there would
obviously be no reason to expect such widespread agreement throughout
the family on what the terms would be in terms of both phonological
shape and semantics. In other words, the word for 'woman' found
throughout the family would be essentially random if Proto-Algonquian
had no word for the concept.

That said, it should also be reiterated that Algonquian does not
encode the distinction of masculine versus feminine as a grammatical
concept. Algonquian languages all have grammatical gender, but it is
a distinction between animate versus inanimate, which pervades all
aspects of both noun and verb morphology. However, in terms of being
able to EXPRESS male versus female, Algonquian works the same as
Hungarian, by having sex as a normal semantic concept yet not as a
morphological category. I would think the burden of proof would be on
anyone who claims that there are 'non-Western' languages which CANNOT
distinguish between female and male.

David J. Costa
Department of Linguistics
U.C. Berkeley
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue

Message 5: Re: 7.592, Disc: Grammatical Gender

Date: Wed, 24 Apr 1996 15:02:53 BST
Subject: Re: 7.592, Disc: Grammatical Gender
There is one word in german language, being grammatically neuter and
being used for adult people of either sex, it is _das Mitglied_
(member of...).

And now guess what happened to this particular word in contemporary

Its plural (_die Mitglieder_) looks like the plural of many male nouns
(like _die Arbeiter_, _die Lehrer_), so a new feminine plural _die
Mitgliederinnen_ is formed, and also a gender neutral plural using the
so-called `Binnen-I', _die MitgliederInnen_.

Another word with similar fate is _die Erstsemester_ (freshmen). 

Once the german magazine Der Spiegel documented a feminist saying
about some of her colleagues _Arschl"ocherinnen_ (from _das
Arschloch_). Sorry, I don't have the reference ready, it was on the
last page under `Hohlspiegel', but I cannot be more precise.

- J"org Knappen.

P.S. There were also attempts to analyse the word _Mitglied_ as _mit
Glied_ with new coinages for female persons, but those coinages were
always perceived as jokes (like _Mitklit_).
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue

Message 6: Grammatical gender

Date: Wed, 24 Apr 1996 10:28:43 EDT
From: Karl Teeter <>
Subject: Grammatical gender

In the recent discussion about gender, there was a communication from
Dan Moonhawk Alford on gender in Algonquian languages. The date was
April 19, and I no longer can find the printout, so I apologize. But
he did, as I recall, quote "his Algonkian friends" as saying sex
distinctions were irrelevant to the language and that the word "squaw"
came from a form meaning "entering", and, finally, that although it is
true Algonquian gender divides animate and inanimate, the animate
gender somehow represents something more mystical than the class of
living things. Dan's friends are wrong all three times, unfortunately,
and on this list we really should get our linguistic facts
straight. So a few words on Algonquian gender.

	Speakers of Algonquian languages, which have a typical
arbitrary grammatical gender class with animacy and not sex as
primary, nevertheless have no problem at all in telling boys from
girls, either linguistically or in real life. That is the first fact
to get straight. Next, (2) The reconstructed proto-Algonquian word
for "woman" is actually based on a verb "to be female": *ethkweewa
(where th stands for theta). It shouldn't be too hard to see that
this is the source of the word "squaw", which is a perfectly good word
despite derogative use in English. And finally, (3) the class of
animate nouns pretty much does cover everything in the real world
which is a living thing, so "tree" is animate but "stick" is
inanimate. We find the usual strictly grammatical arbitrary exceptions
too; thus, strawberries are animate but blueberries inanimate. There
are also extensions of gender beyond the simple distinction of being a
living or non-living thing, thus artefacts from the white world are
usually named with animate nouns: pots and kettles, stoves, knives,
etc. Nothing mystical or holy, just regular arbitrary grammatical
gender. Don't believe everything you hear or read, Dan and colleagues.
Yours, kvt
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue