Hacker News Reader

Comment by irrational | Hacker News Reader

>The neuter gender is very old indeed, arguably older than the other two genders. Through the systematic comparison of ancient languages, we believe that Proto-Indo-European, in its oldest reconstructable state, had only two gender-like categories for its nouns: animate and inanimate (that is, for moving and not-moving things). It was the inanimate that became the neuter gender in its descendant languages, while the animate would later split into two further genders. This is the system that Latin inherited.

That is super interesting to me. But I wonder how they defined moving versus non-moving things? At first I was thinking moving meant living things like animals, but a car moves. Of course, they didn't have cars/trains/etc. back then. But they did have boats (probably). Are boats moving or non-moving? And there are certain animals that don't ever move (especially in the oceans). Would they be moving or non-moving?

View on HN

I'm not sure that moving versus not moving is the right dichotomy, despite what the article says. "Animate" itself comes from the proto IE root *ani- meaning "to breathe" so it could refer to things perceived of at the time as breathing, which includes most larger animals.

But as others have noted, even in ancient times, the relationship between the gender of the word and the thing it represented in the real world was highly fossilized and often arbitrary outside of personal pronouns and a few animal names of agricultural importance (cow vs. bull).

Yes, the expressions “moving” – “not-moving” are rather misleading.

The actual distinction according to this theory (or to my favourite flavour of this theory) was: “able to act willfully” – “unable to act willfully”. The distinction was quite logical. Cp.:

(A) “Susan runs down the hill.”

(B) “Water runs down the hill.”

A speaker of Proto-Indo-European might have said, learning English: “This is strange – as if these were somehow the same acts or events. ‘Susan runs’, OK, but water does not run – it is dragged down. It does not act, something happens to it.” (And then the PIE-speaker might have speculated that anglophone people have an animistic worldview or so.)

In PIE the subject of a sentence seems to have denoted an actual agens; and there seems to have been a class of nouns for things unable to act willfully, hence unable to be the subject. So that the subject in (A) and (C) would have been a subject in the nominative, denoting an agens, whereas in (B) “water” would not have been seen as subject; the sentence would have been constructed differently and “water” would have been put in the same case as “water” in (C).

Because all over the old IE languages, neuter nouns have the same form for the nominative and the accusative, and it seems to have been rather the form for the accusative: Cp.

Latin masc. nom. sg. ‑us

Latin masc. acc. sg. ‑um

Latin neutr. nom. sg. ‑um

Latin neutr. acc. sg. ‑um

There is further evidence, too.

In principle, the distinction was quite logical, but its application is difficult and gets easily messed up over time until the old semantic distinction has become a mere formal one.

> all over the old IE languages, neuter nouns have the same form for the nominative and the accusative

The lack of distinction between neuter nominative and neuter accusative is not just true of nouns in old IE languages. It's true of all grammatical categories in virtually all modern IE languages too (with some very obscure exceptions). Consider German articles der/den vs. das/das. Or English pronouns she/her vs. it/it.

More info: https://latin.stackexchange.com/questions/18/accusative-equa...

https://linguistics.stackexchange.com/questions/9105/why-do-...

Apparently one of the obscure exceptions is the Russian word оно - but they are indeed quite rare.

Correct: According to the theory outlined above, no neutral form could have been a PIE nominative, because the PIE nominative would have been marked willfully acting entities. So the root cause would have been the distinction made for nouns (“can willfully act” – “cannot willfully act”), and this must affect all words with noun-agreement (i.e. adj., det.) or words substituting, deputizing nouns (i.e. pronouns).

It is wonderful that this formerly semantic system (if the theory is correct) has left its formal trace over millennia and a range of thousands of miles.

Such preservation was likely in so far as otherwise a new suffix-inflectional distinction had to be introduced in order to distinguish the nominative neuter from the accusative neuter; and in the history of the IE languages, suffix-inflectional distinctions are rather removed than built.

so the origin of masculine feminine in IE is weird. The issue is that Hittite, the oldest attested IE language has two genders animate and inanimate. These are very very conventional. The sea Aruna is animate, but a grindstone is inanimate! Tocharian also has evidence of the animate inanimate split, but tocharian is such a mess no one knows what to make of it. All other IE language families (besides the weirdos like Armenian) have or had 3 genders. The question is the relationship between Hittite and the rest of IE. Most indoeuropeanists think Hittite (and maybe tocharian) split before the rest of the IE languages, which after the split developed gender and other things that Hittite lacks, before splitting in turn. That said, it’s possible that masculine and feminine merged in Hittite, and the trigender system came first.

> The sea Aruna is animate, but a grindstone is inanimate!

Is this supposed to be weird? A grindstone won't do anything unless you put considerable effort into making it move. But you can go look at the sea and watch it constantly crashing into the shore, to say nothing of rising and falling with the tides.

Farsi has hints of this. I don't think any linguist would argue that Farsi is a gendered language (it does not even have gendered pronouns), but there are two instances where one could argue remnants of an animate/inanimate gender system remains:

1. Farsi has two suffixes that make a noun plural (I am going to write things semi-phonetically): huh and un. Huh can be used for absolutely everything. Un can only be used for living things, e.g.: derakht-un (trees), daneshamouz-un (students), baradar-un (brothers, and its similarity to brethren has caught my eye).

2. You can use third person singular verb conjugation for third person plural subjects, but only if the subject is non-living.

> You can use third person singular verb conjugation for third person plural subjects, but only if the subject is non-living.

This does seem likely to be a relic of something in the past, since nearly the same thing is true in classical Greek - neuter plural subjects agree with singular verbs, not plural verbs. (You weren't clear about whether singular conjugation is required for nonliving subjects in Farsi. I believe singular agreement for neuter plural subjects is mandatory in Greek.)

Only Attiki syntax in Ancient Greek had this as a special form I believe and it was not limited to inanimate subjects. The classic example of this is τα παιδία παίζει , the children are playing.

I didn't mean to say it was limited to inanimate subjects. It's for neuter subjects. But we've been talking about a relationship between Indo-European neuter gender and inanimacy, so this still seems plausibly related.

This is a new development in farsi. ha is from the middle person iha, which ultimately derives from the indo-iranian plus -s (same as in english). s->h is characteristic of indo iranian in general and iranian in particular. an in middle person (which comes from another indo-iranian plural an) means this particular thing in middle person. In farsi this converted into general vs. living.

Japanese has this distinction of animate/inanimate although I don't think(?) it's connected in any way to Proto-Indo-European. Anyway in that language it's subtle and weird. Living humans and animals are animate and paperclips are inanimate. Dead bodies - inanimate - unless you're talking respectfully about a deceased relative. Dolls are inanimate sometimes, but I've also seen them use the animate version (haunted dolls?). Camels can, according to [1], be either depending on whether you look at them as an animal or a mode of transport.

[1] https://japanese.stackexchange.com/a/59531

The grammatical gender of nouns is mostly arbitrary. A famous example is "key", which is feminine in Italian and Spanish but masculine in German. This is also the case in Latin with its neuter. My Latin is very rusty, but IIRC from high school, there are a lot of nouns that refer to "neuter" things, but are actually masculine or feminine, for example "vis" (= strength) is feminine.

A lot of grammar is not arbitrary but just has a very rich history that is not apparent and sometimes not even traceable anymore.

Schlüssel basically is a combination of schließen (close) and the ending -el. Here this is the third meaning of -el listed on wiktionary: "suffix in agent and instrumental nouns". (https://en.wiktionary.org/wiki/-el#German). As far as I can see all of the words that have this version of the -el suffix are masculine.

Languages are full of those rules that just are not apparent anymore, because they are so deeply layered and often apply to only a few words that they are not useful to learn.

well noun gender does seem disconnected from the nature of the thing only in modern languages, IMO. because of how each word evolved. in antique languages, grammatical gender is closely tied to the word's morphology, especially to the word's ending in Latin. in essence there is not really gender in languages, just some group of nouns which behaves very grammatically similarly. they just happened to be suspiciously congruent to the genders in words indicating things which do have gender in those languages which were studied when linguists invented the term of grammatical gender.

"Key" in Italian and Spanish comes more or less directly from Latin 'clavis' (fem), I guess, they are being romance languages; in German however it may have a different root (aber https://www.dwds.de/wb/Schl%C3%BCssel#etymwb-1 says something about Latin origin too). Most likely its form was more similar to other masculine words when its gender stabilized.

The arbitrarity is among naturally gender-less things, as I observed. People, professions, animals have the "right" grammatical gender overwhelmingly. There was also a process in Romance languages. Neuter words tended to become masculine because many of them behaves more like masc than fem in most cases. thats why some woman body parts are grammatically masculine.

> aber https://www.dwds.de/wb/Schl%C3%BCssel#etymwb-1 says something about Latin origin too)

It doesn't say that "Schlüssel" has a Latin origin (see also https://en.wiktionary.org/wiki/Schl%C3%BCssel for an etymology). It says that some terms related to Latin "clavis" have corresponding German terms formed by somewhat analogous derivation from "Schlüssel". Most prominently "Schlüsselbein", literally "key bone", from "clavicula" (clavicle).

danke for the interpretation. i hoped someone clarifies it for me :)

Some languages do have non arbitrators genders. In Dravidian languages male and female nouns usually only refer to humans. In one dravian language, kvi, the feminine was lost and so the only masculine things were adult human males, leading to amusing fact that early ethnographers didn’t realize a major god was actually a goddess!

“mostly arbitrary”

I’ll say: el clítoris.

Latin has mentula and verpa, both feminine words referring to a male appendage.

Giacomo Cassanova had a bit on this:

“When asked by an English poet who was at the table to read the ancient couplet Discite grammatici cur mascula nomina cunnus/ Et cur femineum mentula nomen habet [Teach us, grammarians, why vagina (cunnus) is a masculine noun/ And why penis (mentula) is feminine], Giacomo answered it with a witty pentameter of his own invention: Disce quod a domino nomina servus habet [It is because the slave always takes the name of his master].”

C'mon, we all know it's an introverted dick.

If you take die Klaviatur you have it female in german as well. It's just that Schlüssel is of different semantic origin compared to clave

Klaviatur is feminine not because of the etymology of the root, but rather because of the ending -tur.

German will raise you the grammatically neuter “Mädchen”, meaning girl or maiden.

I’m reminded of the Mark Twain essay, The Awful German Language where he says:

“In German, a young lady has no sex, while a turnip has. Think what overwrought reverence that shows for the turnip, and what callous disrespect for the girl.”

Despite its age the essay* is well worth a read for native English speakers struggling to learn German.

* https://faculty.georgetown.edu/jod/texts/twain.german.html

Mark Twain was making fun - the essay isn't supposed to be taken seriously.

His conclusion would also be wrong, since the grammatical gender has nothing to do with he biological gender - it's the gender assigned to all diminutive forms, regardless of grammatical gender.

E.g. "der Vater" (the father) becomes "das Väterchen" just as well without loosing his proverbial balls in the process.

Well yes - it's clearly a joke? It seems so obvious to me that I didn't think it needed to be pointed out when I quoted it.

That's because of a systematic rule that all words ending in the "chen" morpheme are neuter.

To expand on that: -chen is usually indicating a diminutive form of another word. Some examples (also observe how that affects the vowel in the first syllable):

die Wurst, das Würstchen (the sausage, the small sausage)

der Mann, das Männchen (the man, the little man)

der Becher, das Becherchen (the cup, the small cup; this is an example in which the first vowel is phonetically close enough to the following "e" in "-chen" that no vowel change occurs (we do not even have umlauts for "i" or "e" in german)).

Sure, there's a certain logic to it.

I just think it's an unimpeachable example of the disconnect between biological and grammatical gender.

> I wonder how they defined moving versus non-moving things?

Their concept of "moving things" was what is often translated as "self-movers" in translations of works like those of Aristotle. The idea was that they were things that could move without anything else controlling them. Cars and boats don't move unless humans are controlling them. But animals do.

You may be interested to know that other languages do distinguish animacy/inanimacy:

https://en.wikipedia.org/wiki/Animacy

It isn’t always equivalent to gender though.

i'd categorize boat as inanimated. my experience is that "moving" means self-moving in this context. also languages are generally fuzzy on edge cases. a static sea creature may be inanimated in grammatical sense eventhough it's live. it depends on what it the most resembles to to the people who first name it: does it look like more of a rock or more of a snail? "oh, it turned out that rock-like thing is a snail something in a hard shell? never mind everyone call it sea-rock by now."

Germanic languages (with their retained neutral gender) are probably a good place to start looking at this.

I found this passage really provocative as well! It got me wondering what sort of conceptualization and distinction these cultures and early humans had around gender when they didn't see gender as the original way to split themselves.

It got me thinking about what sort of language a descendent of bonobos might develop, and even whether proto humans conceived of gender in a more fluid way during co-evolution of language and our biology that allowed for language. Wondering if there's anything here like the original sort of QWERTY entrenchment

Why would they need to "split themselves"? What's good about creating divisions for divisions' sake? It's certainly not necessary for human languages to have gender distinctions.

There are often stabilities in dualities, and specifically in having two strategies. "Segment" themselves was perhaps a better term, but two segments is often what I'm personally interested in

Feel free to investigate stability theory and specifically duality in dynamical systems (living cells themselves are dynamical systems, and I would venture human society could also be described as one)

https://www.researchgate.net/profile/Moshe-Farjoun/publicati...

http://home.eng.iastate.edu/~ugvaidya/papers/Vaidya_Allerton...

Reply on news.ycombinator.com