Are there any languages that mark nouns as mass?

Question

Nouns like water, mud, furniture in English are odd with plural morphology (adding -s, as in furnitures), with numerals (three furniture(s)), and seem to have their own quantifier (much water but not much boys), and are typically referred to as mass nouns. In contrast stand so-called count nouns like chair, table, and book (e.g. chairs, three chairs, many chairs).

"Count nouns" are easily singular, a chair, or plural, chairs. When a count noun appears without the indefinite article or plural marking in English, the default assumption is that it should be interpreted "as mass": There was chair all over the highway. Interestingly, there is no equivalent to marking mass as we mark plural with -s in English, and not in any other language I'm aware of. I know there are languages that mark singular, plural, all of the dual through paucal markers, etc, and still other languages with the glorious inverse number marking systems (a single marker, depending on the lexical root, either means "plural" or "singular"). But are there any languages that mark a noun as having a mass interpretation? Such a marker would result in an odd interpretation when combined with a "count noun", and "mass nouns" might obligatorily take the mass marker.

Edit: It is true that there are languages with morphemes that, when applied to a "lexically mass" noun, describe a singular unit of the stuff, or a plurality of units of the stuff. Such data may be used to motivate the idea of a lexical category "mass" for nominals. To be more precise about my question: I am wondering if any language has a "mass marker" that marks any noun as mass, so that when added to a word expressing concepts like boy or chair, the meaning is boy-stuff or chair-stuff.

Answer 1

English has the suffix -age as one way of indicating that a noun is mass. Wiktionary defines it as "Forming nouns with the sense of collection or appurtenance." It's not a regular marker of mass, but it's a fairly productive derivation:

• A break is a single fracture or other failure; breakage is one or more such failures.
• The Jargon File lists lossage as "impl[ying] a continuing lose of which the speaker is currently a victim."
• Voltage and amperage, derived from the unit names volt and ampere, denote the mass quantities of electrical potential and current.

The Wiktionary entry lists -ery as a synonym, as in hosiery < hose or shrubbery < shrub or badassery < badass.

Answer 2

In a way, yes.

The thing about mass nouns is that the things they refer to are usually inherently plural/collective/inseparable, like flour, milk, dust, and grass. Those things are most commonly encountered or discussed as, well, masses of a given substance, rather than individual pieces of said substance.

So, it would be somewhat unusual for a language to explicitly mark the 'mass' form of a mass noun, because that would suggest that 'a drop of milk' is the run-of-the mill, everyday way of talking about milk, while 'many drops of milk as one mass' needs to be explicitly indicated because your audience couldn't assume that's what you meant. In effect, you would end up with a similar system to the singular vs. plural marking that happens for count nouns, and it would erode an existing distinction between mass nouns and count nouns.

So instead, what some languages do is mark the singulative form of a mass noun, which means basically that there is a marked form of the noun to indicate when it is a piece/part of the unmarked mass noun. In English, we can only do this in a roundabout way: a grain of rice vs. rice, but if we had a singulative marker, say *-op, we could just say *ricop and rice, in the same way as in Bari, you have lɛ-tat 'a drop of milk' and lɛ 'milk'.

A system like this means that, although the mass form of the mass noun is unmarked, mass nouns have their own specific morphology as opposed to count nouns, so are easily distinguished from count nouns (unlike in English, where mass nouns don't take any special morphemes).

Singulative marking is encountered as part of the tripartite number marking patterns found in many Nilo-Saharan and Niger-Congo languages. A tripartite number system exhibits three principles of number indication, namely pluralisation, singulativisation, and substitution of number morphemes, as discussed in this paper on Western Nilotic noun morphology by Storch (2007).

Here is an example from Anywa (Anuak) from Reh 1996 but reproduced by Storch:

              SINGULAR     PLURAL 

 ‘bell’       okoot        okóód-í     (plural formation)

              COLLECTIVE   SINGULATIVE 

 ‘butter’     búóp         búób-ò      (singulative formation)

              SINGULAR     PLURAL  

 ‘loincloth’  kèèl-ʌ́       kèèl-é      (replacement pattern)

And you can see that in this, the mass/collective noun 'butter' is unmarked for the collective form but marked for the singulative formation, with distinct morphology from the plural formation and the 'replacement pattern'.

In the example given above, and others typical of the tripartite system, for a noun like 'bell' the underlying form is a singular, with a morphologically complex plural, while the morphologically unmarked and thus underlying form of ‘butter’ is plural/collective/mass both grammatically and semantically. A singulative noun is typically a word that denotes an item that is singled out from a mass or group of similar items. Items that normally occur in pairs or larger numbers are expressed by nouns that are semantically and grammatically plural or collective and morphologically not overtly marked. For a noun like 'loincloth', with the 'replacement pattern', both forms a morphologically complex, and neither may be underlying.

Of course, the semantic distinctions made by speakers of a particular language are crucial to determining how a noun is treated for number marking. This is discussed quite a bit in this paper by Dimmendaal on number marking and noun classification in Nilo-Saharan (2000) (access via an institution). He argues that that there is a continuum between 'object' and 'substance', with the end points of this continuum being universally distinguished (i.e. the most 'county' count nouns and the most 'massy' mass nouns), but with intermediate cases being "linguistically malleable" across languages. This is not just true for, say, Nilo-Saharan languages compared to Indo-European languages - it can be true for even very closely related languages, because there is an enormous amount of variation in how these systems work (particularly in the parts of Africa relevant to the Niger-Congo and Nilo-Saharan language groupings, where there are large amounts of linguistic transfer between languages which are geographically close but genetically distant).

Dimmendaal also gives examples from some other languages which show structural similarities to the the singulative marking patterns in the tripartite systems of Nilo-Saharan, such as Russian gorox 'peas', gorošina 'pea'. In the Native American language Kiowa, nouns which are inherently singular or dual take a number suffix to mark plural, while other nouns which are inherently dual or plural ('bone', 'rib', 'wing') take a singulative marker.

So, basically, mass nouns are 'marked', or indicated, by virtue of being morphologically unmarked but having a clear singulative counterpart, and unique noun morphology.

The above points do not take into account the complex interactions between noun classifier systems and number marking in Nilo-Saharan languages, nor external ways of indicating the 'mass-ness' of a noun (the classificatory verbs discussed by @Daniel Briggs) - there are also many other ways to indicate this beyond the noun morphology.

Answer 3

I'm not sure if there are languages that highlight the mass distinction in particular--the boundaries are of course less clear-cut than for number, but that doesn't mean a language wouldn't have two or three "masses," with its speakers allowed to use either/multiple ones for masses at boundaries; however, there are languages that make finer, qualitative distinctions that correlate with mass/scale. For instance, in The Semantics of Classification in Koyukon Athabaskan by Melissa Axelrod, we read:

Koyukon Classificatory Verbs

G+ø+'o      'compact object is in position, is sitting there'
G+ø+ton     'flat, rigid, or sticklike object is there'
G+ø+lo      'plural objects are there'
ø+kko       'object in an open, shallow container is there'
G+ɬ+ton     'bag or enclosed object is there'
G+ɬ+kooɬ    'flat, flexible, or clothlike object is there'
G+ɬ+koot    'food, edible object is there'
de+ɬ+t'on   'burning object is there'
G+de+dzok   'disorderly, scattered plural objects are there'
G+de+tlaakk 'mushy, wet, messy object is there'
de+de+nokk  'granular or powdery substance is there'

The capital G refers to a gender prefix, which is chosen from:

Koyukon Gender Prefixes

ne-    faces, berries, beads, thimbles, string, rope
de-    wood, plants, furniture, mittens, boots, rigid containers
dene-  round, heavy objects: animal heads, cabbages, apples, rocks;
       long, cylindrical objects: pipes, bridges, pencils, guns
hu-    the areal prefix, used also to classify areas, events, abstractions
hede-  weather
(zero) people, animals

She then works through an example changing classification and gender with flour, and its implications. It seems as though, if you were referring to rice, for instance, you might choose G+ø+lo, G+de+dzok, or de+de+nokk depending on whether it was just a few grains, several, or many, which would correlate with the mass. So this is somewhat related to the question.

Answer 4

In Asturian, they are marked by gender. Uncountables (or mass nouns) are neuter. Words that be both countable and not will change gender with the countable form being masculine or feminine.

• L'arena ye blanco_NEUT. (the sand is white, as mass)

• L'arena ye blanca_FEM. (the [grain of] sand is white, as countable)

Likewise, the pronouns change.

• L'agua, ¿veslo_NEUT? (the water, you see it?, as mass)
• L'agua, ¿quiéresla_FEM? (the [bottle of] water, you want it?, as countable.

What's interesting is that the split to the neuter from masculine/feminine didn't get fully realized with nouns (the neuter aspect of verbs and clauses — conserved even in Spanish — is total). It got stuck on prepositioned articles and adjectives, which still agree with the inherent gender of the nouns. Everywhere else — postpositioned adjectives, predicative adjectives, pronouns, article+adjective sans noun, etc — display the neuter form. So if you say "La blanca agua" (the white sand) blanca has a feminine form, but you could, for instance, say "La blanca arena ye guapo" (the white sand is pretty) where guapo is in neuter even though blanca shows a feminine form.

Per comment's request, here's some information on the Asturian neuter (thankfully Asturian is well documented, you can get the Gramática de la llingua asturiana on-line from the Academy of the Asturian Language, and it's not too terribly hard to read if you already know one of the Iberian Romance languages). Their examples are probably better than mine, but mainly show off the material-as-uncountable-as-neuter aspect that a lot more nouns can pick up.

VII.2.1 El xéneru y el númberu - El neutru - Carauterístiques xenerales
El neutru ye un xéneru qu'apaez nel artículu, nos axetivos, nos referentes y nos pronomes personales qu'entren en concordancia con sustantivos non cuntables y con otros elementos d'asemeyaes carauterístiques. […] la presencia del neutru nel axetivu o nel referente ye pura repercusión del calter non cuntable del términu principal.

The neuter is a gender that appears in the article, in adjectives, in referents and in personal pronouns that enter into agreement with uncountable nouns and with other elements _{[verbs, clauses, and situations]} of similar characteristics. […] the presence of the neuter in the adjective or in the referent is pure consequence of the uncountable character of the principal term.

XXII.2.3.2.B El grupu nominal Subordinación del grupu nominal - La concordancia nel grupu nominal - La concordancia de xéneru - La concordancia del neutru
Ha recordase qu'esisten, per un llau, sustantivos siempre non cuntables y, per otru llau, sustantivos que puen ser cuntables o non cuntables acordies col contestu en que s'empleguen. Los primeros piden concordancia neutra; los segundos, según los casos, piden concordancia de masculín o femenín, o de neutru.
Exemplos de concordancies con sustantivos siempre non cuntables: echa lleña seco al fueu, fartucóse de carne asao, ye dineru bien ganao, lleche preso, lleche cuayao, ¿d'ú sal esi fumu blanco?, llegó xente mozo.
Exemplos con sustantivos que puen ser cuntables o non cuntables: atopó nel llibru una fueya seca, pero ta tol suelu enllenu de fueya seco; mancóse con una piedra picuda en güeyu, pero a la vera'l ríu hai muncha piedra menudo; apuntó'l teléfonu nun papel blancu, pero baxó a la llibrería a mercar papel blanco; nun comí más qu'una pera madura, pero llegó un camionáu de pera maduro; tien un güesu rotu, pero ye una figura de güesu trabayao.

It ought to be remembered that on the one hand there exists nouns that are always uncountable, and the other, nouns that can be either countable or uncountable according to the context in which they are employed. The former needs neuter agreement; the latter, according to the case, need agreement in masculine or feminine, o in neuter.
Examples of agreements with nouns that are always uncountable: put some dry_NEUT wood on the fire, he/she/it engorged him/her/itself with roasted_NEUT meat, spoiled_NEUT milk, spoiled_NEUT milk, where did this white_NEUT smoke come from?, some young_NEUT people arrived
Examples of nouns that may be countable or uncountable: he/she found a dry_FEM leaf in the book, but the ground is covered with dry_NEUT leaves/foliage; he/she got by a chipped_FEM rock in his/her eye, but at the river's edge there's lots of smooth_NEUT rock(s); he/she wrote down the number on a white_MASC sheet of paper, but he/she went to the bookstore to buy white_NEUT paper; he/she/it has a broken_MASC bone, but it's a figure made of sculpted_NEUT bone.