A linguistic mystery yields clues in Russian

Feb 07, 2014 by Peter Dizikes

When it comes to numbers, Russian grammar has a bewildering thicket of rules. A singular noun such as "table" ("stol" in Russian), used as the subject of a sentence, takes a special "case form" called the nominative singular. When used with numbers five and above, table takes a different form called the genitive plural ("pjat' stolov"). And with numbers from two to four, it takes still a different form, the genitive singular ("dva stola").

But with any number whose last digit is 1, the proper case is again the nominative singular: In Russian, "5,281 children" would be translated literally as "5,281 child." Many of these forms change again if the noun is the object of a preposition.

"Russian has this astonishing mess with the numbers," says David Pesetsky, the Ferrari P. Ward Professor of Modern Languages and Linguistics and head of MIT's Department of Linguistics and Philosophy.

What interests Pesetsky most about this mess are the rules of grammatical case: the prefixes and suffixes that, in many languages, tell us what part of speech a word holds in a given sentence. To read any Latin sentence, for instance, you must recognize the suffix that denotes the case ending of each noun, in order to see what part of speech that noun has, and to glean the proper meaning of the sentence. Today, case features prominently in German, Japanese, and Korean, among other languages. (In English, case exists only in small traces, such as the distinction between "she" and "her.")

But the oddity of Russia's case rules for numbers, Pesetsky thinks, may tell us something important about case more generally, by giving us a rare glimpse into aspects of the modern language that are normally hidden from view by other complexities. Case, he argues, does not comprise a stand-alone set of grammatical rules; instead, certain case endings naturally belong to, and derive from, certain parts of speech. Nouns, as Pesetsky puts it, are "born genitive," and the existence of case may simply reflect how the different parts of speech interact with each other.

Now Pesetsky is making his case for this theory in a book, "Russian Case Morphology and the Syntactic Categories," just published by MIT Press.

"The idea is to reduce the notion of case to the notion of parts of speech," Pesetsky says. "You can get rid of an extra complication in the entire theory of grammar." In so doing, however, Pesetsky says he is addressing a "deeper, more meaningful [question]: to ask why languages should have case in the first place."

Born this way: Are nouns really genitive?

To anyone who has even struggled through, say, elementary Latin, the idea that have a natural case seems quite peculiar. Consider the forms of the word for dog: "Canis," the nominative case ending, means it is the subject of a sentence ("The dog ate Virgil's food"), while "canem," the accusative, means it is the direct object ("Virgil fed the dog"), and "cani," the dative, is the indirect object ("Virgil took the food to the dog"). The existence of case provides such linguistic flexibility.

The cover of "Russian Case Morphology and the Syntactic Categories" by David Pesetsky.

"The idea that nouns are born genitive is completely counterintuitive to everybody," Pesetsky acknowledges. "It's counterintuitive to me. It's not an obvious proposal."

To see why nouns might be inherently genitive, though, consider a Russian phrase Pesetsky discusses in his book: "eti poslednie pjat' krasivyx stolov," or "these last five beautiful tables," which is an example of "case mismatch." The words "these," "last," and "five" are all in the nominative case, but "beautiful" and "tables" remain genitive.

Pesetsky—drawing on a particular comparison first developed by his MIT colleague Norvin Richards, between Russian and the vanishing Australian language Lardil—asserts that this mismatch occurs because Russian is actually a "case-stacking" language, that is, one in which multiple cases can legitimately apply to a given word at a given time. In Russian, that case-stacking tendency is usually invisible, because of the interaction of other processes.

"My idea is that what nouns and verbs are doing when they do case marking is they're really just marking some other phrase as having been dependent on them," Pesetsky says. "A noun assigns genitive because genitive is just another name for noun. Verbs assign accusative because they are accusative. Prepositions assign instrumental or dative [cases] because they are [originally] instrumental or dative."

In this view, the persistence of the genitive case in the numeral-based Russian nouns is not just a curiosity, but a window into the underlying mechanics of language.

"These Russian numerals are a place where you can see nouns in their primeval natural state," Pesetsky asserts.

Pesetsky's proposals about Russian are not what is taught in Russian textbooks, notes Pesetsky, who first worked on the language in the 1970s.

"If you open up a Russian book, they'll tell you the numeral assigns the genitive case," Pesetsky says. "My suggestion is we turn that on its head. What the numeral is actually doing is preventing the genitive that the nouns were born with from being erased due to assignment of another case by the verb."

Spur for further research

Pesetsky says that his ideas have surprised some linguists, especially specialists in Slavic languages, but the book has received praise. John Frederick Bailyn, a linguistics professor at Stony Brook University who has written about Russian syntax, has called the book a "brilliant and astonishingly original account of one of the most notorious problems in Russian morphosyntax," and adds that in so doing it "provides a radical simplification of the overall architecture of syntactic theory."

As Pesetsky readily acknowledges, the discovery of certain examples showing that case does not operate as we have long assumed is a departure point for further inquiry. 

"It's unlikely that I've completely gotten rid of case as a separate system in the world's languages," he says. "What I suspect is that what we call case is really several interacting systems." As a specialist in syntax, however, he does think that what might emerge from rethinking case is a more robust understanding of how case operations mesh with syntax, the fundamental rules of languages.

"Either I'm wrong, in which case it's a spur for someone to come up with a better theory, or we have to overcome our intuitions," Pesetsky says.

Explore further: World population likely to peak by 2070

Related Stories

Study shows humans and apes learn language differently

Apr 02, 2013

(Medical Xpress)—How do children learn language? Many linguists believe that the stages that a child goes through when learning language mirror the stages of language development in primate evolution. ...

Historical context guides language development

Apr 14, 2011

Not only do we humans enjoy talking -- and talking a lot -- we also do so in very different ways: about 6,000 languages are spoken today worldwide. How this wealth of expression developed, however, largely remains a mystery. ...

Recommended for you

World population likely to peak by 2070

5 hours ago

World population will likely peak at around 9.4 billion around 2070 and then decline to around 9 billion by 2100, according to new population projections from IIASA researchers, published in a new book, World Population and ...

Bullying in schools is still prevalent, national report says

6 hours ago

Despite a dramatic increase in public awareness and anti-bullying legislation nationwide, the prevalence of bullying is still one of the most pressing issues facing our nation's youth, according to a report by researchers ...

Study examines effects of credentialing, personalization

9 hours ago

Chris Gamrat, a doctoral student in learning, design and technology, recently had his study—completed alongside Heather Zimmerman, associate professor of education; Jaclyn Dudek, a doctoral student studying learning, design ...

Data indicate there is no immigration crisis

Oct 22, 2014

Is there an "immigration crisis" on the U.S.-Mexico border? Not according to an examination of historical immigration data, according to a new paper from Rice University's Baker Institute for Public Policy.

User comments : 3

Adjust slider to filter visible comments by rank

Display comments: newest first

nanotech_republika_pl
not rated yet Feb 07, 2014
This example about the table form is good for Polish language too and I'm guessing that it is true for some other Slavic languages. Russian is here an example of a member coming from a family of Slavic languages that come in turn from some common Slavic language or in general maybe from old language like Sanskrit.
antialias_physorg
not rated yet Feb 07, 2014
It's interesting to note that having a lot of cases lets you construct longer sentences and express more complex ideas without ambiguity. This in turn shapes the mind (simple thinkers vs. complex thinkers). It is no coincidence that many of the 'great' philosophers wrote in germanic languages - or that many of the great mathematicians were/are russian.
dnatwork
not rated yet Feb 07, 2014
Or, as Pesetsky notes at the end, Pesetsky could be wrong. It is simpler to say that the numeral is the head of the nominal phrase in "these last five of tables."

The article and adjective (these last) agree in case with the numeral (five)--they are all nominative.

The so-called noun (tables) is in the genitive to answer the question "five of what?"

The numbers 2 through 4 are small sets, so Russian uses the singular genitive of the following noun. The numbers 5 through 20 are larger sets, so the plural genitive is used.

The number 1 is really an adjective (unlike any other numeral, it can used as a stand-alone adjective, as in "Ya odin" [I am alone]), so the following noun is really the head of the nominal phrase. As such, it agrees in case with the preceding adjectives.

Thus, the rules are evident on the surface, marked by case agreement. There's no deeper structure, and no speaker of the language has to deduce it from knowledge they don't have of Latin.