As Easy as ‘One’, ‘Two’, ‘Three’?—The Morphology and Syntax of “Numerals” in Old Church Slavonic, Old Russian, and modern Russian
“Numerals”, that is words designating cardinal numbers, in Old Church Slavonic (OCS) and Old Russian (OR) were nouns or pronouns rather than a separate morphosyntactic category, as they are in modern Russian. Hence, for OCS and OR, I use the term “numerals” in quotes.
Let’s first consider OCS “numerals”. They can be divided into two major sets: the first set includes ‘one’, ‘two’, ‘three’ and ‘four’, and the second set includes ‘five’, ‘six’, ‘seven’, etc. “Numerals” in the first set (‘1’-‘4’) behave syntactically as adjectives in that they agree with their nouns in gender and case. There is no number agreement, because unlike other adjectives, ‘one’ always takes singular nouns, ‘two’ (and ‘both’) always takes dual nouns, and ‘three’ and ‘four’ always take plural nouns. Morphologically, some “numerals” are pronouns and others are nouns. Specifically, the words for ‘one’ (masculine edinъ, neuter edino, and feminine edina), ‘two’ (masculine and neuter dъva and feminine dъve), and ‘both’ (masculine and neuter oba and feminine obě) are pronouns, i.e. they follow the pronominal declension. In contrast, trьe ‘three’ and četyre ‘four’ are nouns following the anomalous subtype of the simple nominal declension (for details, see Lunt 2001: 71-74).
Higher “numerals” (‘5’-‘10’) behave syntactically and morphologically as nouns: they have inherent gender (feminine), follow the simple nominal declension (pętь ‘five’, šestь ‘six’, sedmь ‘seven’, osmь ‘eight’, and devętь ‘nine’ follow the regular subtype of the simple nominal declension, while desętь ‘ten’ follows the anomalous subtype of the simple nominal declension), and take nouns in the genitive form. (Since they denote multiple entities, they always appear with plural nouns.) For “numerals” higher than ten, see Lunt (2001: 153).
In modern Russian, the so-called “higher numerals” also appear with the genitive plural form of the noun: pjat’ knig ‘five books.GEN’. However, in modern Russian these numerals are considered to be a separate part-of-speech, whereas in OCS and OR they were syntactically nouns. Babby (1987) describes three contrasts between the “numerals proper” of modern Russian and what I shall call “cardinal nouns” of OR. (OCS here patterns with OR; OR examples below are from Babby’s article, boldface indicating the relevant agreement) First, in OR demonstratives agreed with the “cardinal nouns” in gender, number, and case, whereas in modern Russian demonstratives agree in number (though not in case) with nouns proper:
Old Russian:
ta | pjat’ | butylok |
that.NOM.SG.FEM | five.NOM.SG.FEM | bottle.GEN.PL |
‘those five bottles’ |
modern Russian:
te | pjat’ | butylok |
that.NOM.PL | five.NOM | bottle.GEN.PL |
‘those five bottles’ |
Second, in OR the “cardinal nouns” triggered agreement on the verb, whereas in modern Russian the verb is either plural or appears in the default, non-agreeing form (3rd person singular neuter):
Old Russian:
ta | pjat’ | butylok | prišla |
that.NOM.SG.FEM | five.NOM.SG.FEM | bottle.GEN.PL | arrived.SG.FEM |
‘those five bottles arrived’ |
modern Russian:
te | pjat’ | butylok | prišli / prišlo |
that.NOM.PL | five.NOM | bottle.GEN.PL | arrived.PL / arrived.SG.NEUT |
‘those five bottles arrived’ |
Third, in OR “cardinal nouns” always appeared with nouns in the genitive, regardless of the position of the entire phrase in the sentence. Thus, both nominative/accusative and oblique noun phrases had exactly the same structure in OR. In modern Russian, oblique noun phrases with numerals show the oblique case on all the elements, including the numeral itself and its complement noun.
Old Russian:
ta | pjat’ | butylok |
that.NOM.SG.FEM | five.NOM.SG.FEM | bottle.GEN.PL |
‘those five bottles’ |
toju | pjat’ju | butylok |
that.INSTR.SG.FEM | five.INSTR.SG.FEM | bottle.GEN.PL |
‘with those five bottles’ |
modern Russian:
te | pjat’ | butylok |
that.NOM.PL | five.NOM | bottle.GEN.PL |
‘those five bottles’ |
temi | pjat’ju | butylkami |
that.INSTR.PL | five.INSTR | bottle.INSTR.PL |
‘with those five bottles’ |
A detailed investigation of the morphosyntax of modern Russian numerals goes beyond the scope of this post, and I have dealt with these issues elsewhere (see Pereltsvaig 2006 a,b, 2007, 2010). Interestingly, while modern Russian developed a distinct morphosyntactic category (“part-of-speech”) for numerals, it also has some “cardinal nouns”, such as pjatok ‘five’, desjatok ‘ten’ etc., which behave exactly like “cardinal nouns” in OR (see Pereltsvaig 2007 for details). Also, unlike their numeral counterparts, “cardinal nouns” do not participate in the so-called Approximative Inversion, where the noun inverts around the numeral to denote an approximate number (see Pereltsvaig 2006b for details):
modern Russian, numeral:
butylok | desjat’ | vodki |
bottle.GEN.PL | ten.NOM | vodka.GEN |
‘approximately ten bottles of vodka’ |
modern Russian, “cardinal noun”:
*butylok | desjatok | vodki |
bottle.GEN.PL | ten.NOM | vodka.GEN |
intended: ‘approximately ten bottles of vodka’ |
Thus, modern Russian developed a distinct morphosyntactic category of numerals (for a detailed analysis of the parallel development in Polish, see Rutkowsky 2002, 2006, 2007). Yet, some of the morphosyntactic complexity of Russian numerals reflects the patterns inherited from Common Slavic (as also reflected in OCS) and Old Russian. Recall that in OCS, numerals for ‘one’ through ‘four’ behaved syntactically as adjectives. In modern Russian, ‘one’ is still an adjective in that it agrees with the noun in gender and case. (Normally, ‘one’ takes nouns in the singular, but pluralia tantum nouns, which are morphologically always plural, appear with odni ‘one.PL’, as in odni brjuki ‘one pair of trousers’.) Numerals ‘two’, ‘three’, and ‘four’ in modern Russian still pattern differently from the higher numerals, which as discussed above, take nouns in the genitive plural. ‘Two’ through ‘four’—called “paucal numerals”—appear with what looks like genitive singular form, as in tri knigi ‘three books’ (compare with pjat’ knig ‘five books’). (Note on the image on the left: četyre goda ‘four years’, but dvadcat’ pjat’ rublej ’25 rubles’.) However, as has been pointed by Mel’čuk (1985: 430-437) and discussed in Pereltsvaig (2010), at least four nouns have a separate form that appears only with the paucal numerals: šagá ‘step’, šará ‘ball’, rjadá ‘row’, časá ‘hour’ (with the stress on the ending rather than on the stem, as in the genitive singular forms of these nouns). Where does this special morphosyntax of ‘two’ through ‘four’ in modern Russian come from?
The roots of the special “paucal morphosyntax” in modern Russian are to be found in Old Russian. Recall that in OCS (and more generally in Common Slavic), ‘two’ appeared with nouns in the dual number (e.g. dъva slěpьca ‘two blind men’). But already in the Old Russian period (starting in the 1200s; cf. Garbuzova 1975) the dual as a productive morphological category begins to erode. Independent morphophonological changes decrease the number of distinct dual forms: while there were six distinct case forms in the singular, there were only three distinct case forms in the dual. The different declensions (still distinguished in the singular and the plural) now share the same dual endings. Moreover, when no word for ‘two’ was present, the plural form was used instead of the dual, as in pomogi rabomъ svoimъ Ioannu i Oleksiju ‘help your slaves John and Alexey’ (from a text dating from 1219 CE). Here, rabomъ svoimъ clearly denotes two slaves, yet its morphological form is dative plural rather than the dative dual, raboma svoima. Even with the word ‘two’ is present, the distinct dual form is gradually replaced the genitive singular. Why? Masculine nouns in the so-called C-stem declension (corresponding essentially to the masculine twofold declension in OCS and masculine first declension in modern Russian) have the homophonous (i.e. same-sounding) form for genitive singular and for nominative/accusative dual, e.g, druga ‘friend’. It is this pattern of homophony that is being gradually extended to other types of nouns in Old Russian, first to nouns of the o‑declension (e.g. nominative/accusative dual čislѣ is replaced by čisla ‘number’), then to nouns other declensions: a‑declension (e.g. nominative/accusative dual ženѣ is replaced by ženy ‘wife’), soft-C-declension, ŭ‑declension, i-declension, and so on. In another development, the use of a special form appearing with ‘two’—historically, the dual, but by now largely replaced by genitive singular—is extended to ‘three’ and ‘four’, which in Common Slavic, OCS, and OR patterned with ‘two’ in their distinctively adjectival syntax.
Sources:
Babby, Leonard H. (1987) Case, Prequantifiers, and Discontinuous Agreement in Russian. Natural Language and Linguistic Theory 5(1): 91-138.
Garbuzova, Ekaterina Petrovna (1975) Imja suščestvitel’noe v drevnerusskom jazyke. [Noun in Old Russian]. Smolensk.
Lunt, Horace G. (2001) Old Church Slavonic Grammar. Mouton de Gruyter.
Mel’chuk, Igor A. (1985) Poverxnostnyj sintaksis russkix čislovyx vyraženij. Wiener Slawischer Almanach Sonderband 16. Institut für Slawistik der Universität Wien, Vienna.
Pereltsvaig, Asya (2006) Small nominals. Natural Language and Linguistic Theory 24(2): 433-500.
Pereltsvaig, Asya (2006) Passing by Cardinals: In support of Head Movement in Nominals. In James E. Lavine, Steven Franks, and Hana Filip (eds.) Formal Approaches to Slavic Linguistics 14: The Princeton Meeting. Ann Arbor, MI: Michigan Slavic Publications. Pp. 277-292.
Pereltsvaig, Asya (2007) On the Universality of DP: A View from Russian. Studia Linguistica 61(1): 59-94.
Pereltsvaig, Asya (2010) As easy as two, three, four? In: Browne, Wayles; Adam Cooper; Alison Fisher; Esra Kesici; Nikola Predolac and Draga Zec (eds.) Formal Approaches to Slavic Linguistics 18: The Cornell Meeting. Ann Arbor, MI: Michigan Slavic Publications. Pp. 417-434.
Rutkowski, Paweł (2002) Numerals as grammaticalised nouns: a generative approach. Interlingüística 13(3): 317-328.
Rutkowski, Paweł (2006) Why Polish numerals should not be analyzed as nouns. In: Changguk Yim (ed.) Minimalist Views on Language Design: Proceedings of the 8th Seoul International Conference on Generative Grammar. Seoul: Hankook/Korean Generative Grammar Circle. Pp. 249-263.
Rutkowski, Paweł (2007) Grammaticalization in the nominal domain: the case of Polish cardinals. In: Blake H. Rodgers (ed.) LSO Working Papers in Linguistics 6: Proceedings of WIGL 2006. Madison: Department of Linguistics, University of Wisconsin-Madison. Pp. 89-102.