by Jim Wood
An old Swedish car or a Swedish old car? What is discovered when word patterns are literally mapped? Jim Wood looks at the Italian School of Linguists, Ginsberg’s adjectives and Guglielmo Cinque’s Functional Structure in DP and IP, surveying patterns that might predict the ways of the world’s languages’ words.
1.
Probably as long as people have been around, they have been drawing maps of things. The appeal is maybe obvious: it gives us the ability to see a much bigger picture than we would otherwise. We can see where things are, how to get places, and where we are. Even those among us who are the least interested in maps can become fascinated when we move to a new area, and spend many hours buried in them until we get used to the place. Some cartographers want maps to be as accurate as possible, others to be as useful as possible.

Guglielmo Cinque's Functional Structure in DP and IP
One recent movement in linguistics has been led by, but is by no means exclusive to, a group of Italian linguists. This is usually known as the “Cartographic Approach” to language, but is sometimes referred to as the “Italian School.” Their goal is to draw a map of the world’s languages; quite literally, they aim to discover a universal order the pieces of a sentence can come in. It is “cartographic” because they draw hierarchical maps of their results. To take an example, the “Cinquean Adverbial Hierarchy,” pioneered by Guglielmo Cinque, is a putative universal ordering of some 32 classes of adverbs, constructed after examining a great many languages.
What does a map of human language look like? This book, for example, has an excellent chapter by Gary-John Scott on adjective ordering. When two adjectives describe the same noun, without a “comma” (that is, without a pause) between them, what order do they come in? To illustrate, let’s consider some excerpts from Allen Ginsberg’s “Howl,” which is known for its strings of adjectives. For example:
angelheaded hipsters burning for the ancient heavenly connection to
the starry dynamo in the machinery of night,
Notice the relative order of ancient and heavenly. Why would one say “the ancient heavenly connection” instead of “the heavenly ancient connection”? Most people would respond that it sounds a bit funny. Others would say that it’s because you’re not talking about “ancient connections” which are heavenly, you’re talking about connections which are ancient and which are heavenly. But this just forces us to restate the question: why is it that when you want to talk about something which is both heavenly and ancient, the most natural order is “ancient heavenly” and not “heavenly ancient?” Taking some more Ginsberg examples:
who loned it through the streets of Idaho seeking visionary indian angels
who were visionary indian angels,
How would it have sounded different if he had been speaking about someone seeking “indian visionary angels” instead of “visionary indian angels?”
who faded out in vast sordid movies, were shifted in dreams, woke on
a sudden Manhattan,
How about “sordid vast movies?” “Who faded out in sordid vast movies?” There’s the urge to contend that Ginsberg fits no patterns, that he is spasmodic and entirely random. But is he? Even more to the point:
who created great suicidal dramas on the apartment cliff-banks of the
Hudson under the wartime blue floodlight of the moon
& their heads shall be crowned with laurel in oblivion,
Can we imagine “who created suicidal great dramas on the apartment?” No, that’s just ridiculous, right? No one would really say that. But again: why is “suicidal great” the ridiculous sounding one and “great suicidal” the natural sounding one?
If these observations were all there was to it, we could say that this is just the way English is. But there seems to be something deeper going on, and there are many languages, even those which supposedly have “free” word order, where a big red German ball would never be called a German red big ball. Other orders exist, for sure, but the cartographic project involves comparing orders which do exist with ones which do not – and many, many languages are taken into account.
2.
What is meant, then, by “universal?” Some take it to mean it is part of Universal Grammar – it is something we are born with and is just a fact of life specific to language. Just like there are many ways of walking, certain gaits probably never occur in any culture, probably because we are born with instincts which tell us (perhaps): don’t walk backwards all the time.
Others have argued that universal ordering comes from more general cognitive mechanisms, that it is not specific to language per se, but is instantiated in language via some other part of the mind. Either way would be interesting, really; either language is a way to learn about how some other part of the mind is organized, or it is a way to learn about how the language part of the mind is organized.
Cinque, in the introduction, justifies pursuing the hypothesis that potential universal ordering applies to all grammatical categories and all languages:
“What makes the enterprise all the more interesting is the mounting evidence of the last several years that the distinct hierarchies … may be universal … This is, at any rate, the strongest position to take, as it is compatible with only one state of affairs. It is the most exposed to refutation, and, hence, more likely to be correct, if unrefuted.”
He then points out a danger in taking a weaker position, namely that these universal hierarchies are not really universal:
“Even if this position should eventually turn out to be right, methodologically it would be wrong to start with it … That would only make us less demanding with respect to the facts and could lead us to miss subtle evidence supporting the stronger position (a risk not present under the other option).”
So let the map drawing begin. The book is divided into two parts. The first part is concerned with mapping noun phrases (which most linguists now call ‘determiner phrases’), and the second is concerned with mapping full sentences. Each part has several authors who are mapping very different areas and/or using very different methods. After all, one would hope that radar, satellites, ships, and airplanes, when combined, would give a more complete map… and hopefully not contradict each other.
3.
This book’s biggest strength, I think, is in its breadth, rather than its uniformity. It is not the most topically focused book out there. But it achieves breadth in a way that is actually quite rare. On the one hand, there is a wide variety of detailed theoretical work. One chapter, written by Anna Cardinaletti and Ian Roberts, focuses on languages such as German, where the verb consistently appears as the second position of the sentence. Thus, instead of “Yesterday I left” you get “Yesterday left I.” Giuliana Guisti’s chapter is a detailed approach to noun phrase structure drawing on recent theoretical proposals by Noam Chomsky. Laura Brugè studies the position of words like “that” in Spanish, discussing varieties of Spanish where it is common in conversation to say things like el libro ese ‘that book’ (literally translated: ‘the book that’).
In addition to this theoretical cross-linguistic work, there is a chapter on first-language acquisition by Maria Teresa Guasti and one on American Sign Language by Carol Neidle and Dawn MacLaughlin. The chapter on language acquisition focuses on where the -s in sentences like John go-es is located. It doesn’t always appear on the verb; in questions and negative sentences, for example, it appears on a dummy auxiliary do:
(1)
a. Gertrude go-es.
b. Gertrude do-es not go.
c. Do-es Gertrude go?
Children often go through a stage where they do not pronounce “-es,” and instead say things like:
(2) Gertrude don’t go.
However, when they form questions, they always use “-es.” They never say “Don’t Gertrude go?”, but always say “Do-esn’t Gertrude go?” They argue that there is a connection between moving do to form a question and pronouncing the “-es,” and it is one of cartography: “do” literally has to pass by “-es” and pick it up as it moves to the beginning of the sentence. This is like saying you have to stop by the post office to get a passport before you take a plane out of the country.
The chapter on American Sign Language (ASL) is an impressive and insightful close to the book. The authors even provide a web address (http://www.bu.edu/asllrp/) where videos of the ASL constructions they discuss are publicly available. Contrary to popular misconception, ASL is a fully-fledged language which is unrelated to English. Those who use ASL and can read English are in a very real sense bilingual. The gestures of ASL are multi-modal (using hand signs, facial gestures, body position, etc.) and occur in three-dimensional space. This allows us to see several dimensions of language occurring at the same time. In spoken language, we have to linearize everything. Negation must appear before or after the verb, or sometimes attached to it. But in ASL, certain (facial) gestures representing negation can occur simultaneously with the verb. Interestingly, the way that this is done ends up making ASL look much more like spoken languages than one might expect.
I’ll give a brief example involving negation. Certain words, such as “anyone,” generally have to occur after negation. (This is an over-simplification, but it will suffice for now.) It is strange to say, “Anyone wasn’t in the room,” but perfectly natural to say, “There wasn’t anyone in the room.” The part of the sentence starting with “n’t” and continuing to the end of the sentence is called the “scope” of negation. Interestingly, ASL negative marking can continue across the entire scope of negation. In the ASL equivalent of “John does not buy a house”, this kind of negative marking starts on “buy” and extends for the rest of the sentence. That is, it extends over the same part of the sentence that allows words like “anyone”. This chapter shows how convincingly the study of American Sign Language (and, of course, other signed languages in the world, of which there are many) can contribute to the cartography of human languages.
4.
The appeal of work like this stems from our curiosity about language and our fascination with maps; it makes apparent bizarre chaos start to make sense. The cartographic approach doesn’t always answer the “why” question, just as geographic cartography doesn’t always explain why the Swiss Alps are here and not there. But it tells you a lot about what pieces of language go where. Its focus is on being useful, in that we can predict where things will appear in the languages of the world and in human language in general, and its focus is on being accurate, even if such accuracy ends up requiring upwards of 17 positions for adjectives. This book is, of course, not the complete story, and does not attempt to be. But as the first volume in the “Cartography of Syntactic Structures” series, it is an excellent start.
*
Posted in features