Rumantsch Grischun is Mirror Maltese!

[pne]

I was just reading the German Wikipedia article on Grammar of Rumantsch Grischun when I came across the section on "collective plurals", in which it says:

Ein typisch rätoromanisches Phänomen ist der Kollektivplural. Er tritt auf bei männlichen Substantiven, die häufig im Plural vorkommen. Er verhält sich wie ein feminines Substantiv im Singular.

• il mail -> der Apfel
• ils mails -> die Äpfel (zählbar, nach Mengenangaben)
• la maila -> die Äpfel (nicht zählbar, allgemein)

That is, some masculine nouns form a regular (masculine) plural but also a collective plural (called "a typically Rhaeto-Romance phænomenon" here) which looks like a singular feminine noun.

Which is (as regards the grammatical genders) pretty much the opposite of what happens in Maltese, as I understand it: the examples there would be:

• it-tuffieħa (fem.) -> the apple
• it-tuffieħiet (fem.pl.) -> the apples (e.g. counted plural, after numbers: 3 tuffieħiet)
• it-tuffieħ (masc.) -> the apples (not specifically counted; apples in general)

Similarly with ħobż "bread (as a material)", ħobża "a (loaf of) bread"; ġobon "cheese", ġobna "a cheese"; ward "roses", warda "a rose"; etc.

Comments

[edricson.livejournal.com]

Well, the Rumansch is really fairly typical of Romance in general and really goes back to Latin (locus, pl. loci, coll. pl. loca as if it were neuter, but obviously assimilated to the feminine in Romance). I wonder if the Maltese construction is in some ways a Romance calque.

[pne.livejournal.com]

I wonder if the Maltese construction is in some ways a Romance calque.

Good question, and not out of the question given the influence of Sicilian and Italian.

I wonder whether other varieties of Arabic do the same thing.

[muckefuck.livejournal.com]

Nope, it's found in Classical Arabic as well, e.g. al-ḥajaru "the stones [coll.]", al-ḥajaratu "the stone", al-ḥajarātu "the stones". The second of these form is often called a "singulative" since the collective is the base form and the singular (and sound plural) are formed from it by means of suffixation.

A singulative also developed independently in the Celtic languages, e.g. derwen "oak", pl. derw or deri.

[edricson.livejournal.com]

Ah, should have remembered that, thanks.

But I'm not sure if Arabic does the later Celtic trick of pluralizing the singulatives back, so Middle Breton guez "trees; wood", singulative guezenn, Modern plural gwezennoù.

[muckefuck.livejournal.com]

Sure it does--you have an example in the Maltese pne posted. The -iet on tuffieħiet is just the Classical Arabic feminine plural ending -āt in a different guise.

[edricson.livejournal.com]

Ah yes. Thanks. Just shows I haven't picked up quite enough in my one semester of Calssical Arabic to quickly parse Maltese :).

[elgrande.livejournal.com]

But Latin didn't have a pattern of countable and uncountable plurals and I don't know of any other Romance language that has it (at least not French, Italian or Spanish), so I'm not sure it's such a typical thing.

[edricson.livejournal.com]

Umm, no, the neuter-plural-for-masculine-noun pattern is exactly a means of expressing this sort of "collective/uncountable" distinction. And you can see vestiges of Latin neuter plurals interpreted as feminine forms in Romance in the il labbro/le labbra type of plurals. Anyway, I'm pretty sure a pattern like the Rumantsch is possible in a lot of "Italian" dialects (but yeah, not in the standard language).