Research Review

Professor Eve Clark

"Puller-Wagons, Plate Eggs, and Pants with Pockets:
Children's Strategies for Acquiring Language"

by Sarah Felstiner, Teacher - Center P.M.
The Bing Times - November 1991

On September 23rd, the Bing staff attended a presentation by Eve Clark, a professor of Linguistics at Stanford. She discussed her research in the field of language acquisition, most of which was done in the game rooms of Bing Nursery School. In particular, she has studied how children learn the meanings of words, and build their vocabulary. Professor Clark described the design and outcome of several of her important pieces of research.

In one study, Clark measured children's understanding of the temporal prepositions "before" and "after." In this study, she presented children with phrases that used these conjunctions in four different ways, and asked the children to act out these phrases with small toys provided. For example, a child might be asked to demonstrate "the horse jumps over the fence before the dog goes inside," or "after the dog goes inside, the horse jumps over the fence." Clark found that 3-year-olds were very successful at the [event 1 before event 2] and [after event 1, event 2] configurations, but had more trouble with sentences phrased in the form [before event 2, event 1] and [event 2 after event 1].

This finding led her to believe that the children in her study were not attending to the conjunctions "before" and "after" at all, but rather performing the events in the order they were stated, whatever the preposition used. By age 5, she noted, children can perform all four forms correctly. They come to learn that these words carry a temporal meaning, and later, that they are opposites. Most importantly, Clark mentioned that parents and teachers should take into account a child's understanding of words like "before" and "after" when giving that child a series of instructions. For instance, with a very young child, the phrase "before you eat, please wash your hands" will meet with less sanitary results than "please wash your hands before you eat." Until they understand these temporal prepositions, children rely on specific strategies (e.g. do the first thing you hear, and then the second thing) to interpret the unfamiliar words.

Eve Clark found the same to be true of spatial prepositions such as "in, on, and under." When she asked young children to put an object "in" an open box , they did so easily. However, when she turned the box on its side, and asked the child to place an object "on" the box, he would put the object in, and place the box face up again. If she demonstrated "on" with one box, and asked a child to do the same thing with another box, he would almost always turn both boxes, and put both objects inside. In other words, young children have an ordered set of strategies for interpreting spatial prepositions. These strategies are based on conceptual preferences for organizing objects in their world (e.g. objects go in boxes, and those boxes must sit face-up).

Another area of the English language that takes some time for children to master is the group of deictic or pointing words such as come and go, this and that, bring and take, and here and there. Children may not differentiate between these various words until age 8 or 9, and can produce phrases like "come me to school" instead of "bring me to school." One possible cause of confusion is that children's early egoistic view of the world makes it hard for them to adjust to speaker-centered phrases. For example, does "here" represent the location of the child, or that of the speaker? By about age 3, children use "here" and "there" in a non-deictic manner, such as "here you go" or "there, I did it." By 4 or 5 they have mastered this/that and come/go, and can use "bring" and "take" appropriately by age 8 or 9.

Professor Clark also studied children's understanding of kinship relations. She asked students from Bing and the Escondido school, aged 3 to 9, to define kinship terms like mother, uncle, etc. She found that their answers were completely incidental, having little to do with actual familial relations. For instance, "a brother is someone who wears pants with pockets." By age 4 or 5, children may gain some relational understanding, but they seldom grasp reciprocity (if he is my brother, then I am his sister) until age 6. Even some 9-year-olds have trouble with the aunt/niece and uncle/nephew configuration, perhaps because many American families address close family friends as "Uncle Harry" or "Aunt Jessie." Whether or not a young child learns kinship terms has little to do with the relatives a child has frequent contact with, but rather is related to the age of the child, and the difficulty of the kinship term to be mastered.

Much of Clark's work has explored children's understanding and production of compound words, particularly those with a prefix or suffix. Children spontaneously coin such words from about age two (e.g. someone who shovels is a "shoveler") and will also offer spontaneous analyses of compound words. For instance, "it's called a runway because you run on it." Clark asked the children in this study a series of 30 questions in the form "What is a ___er?" Even with nonsense words, children at age 3 always responded "he ___s." Yet these same children could not reverse this process grammatically. When she asked them, "here's a man who builds walls. What shall we call him?" they usually responded "a build-man" or "a wall-man," rather than "a wall-builder."

Similarly, when she asked what one might call a person who pulls wagons, children often responded "a pull-wagon." However, if she later asked what a "pull-wagon" might be, children suggested it was "a wagon that you pull." So again, this process was not reversible. Four-year-olds might call someone who pulls wagons a "puller-wagon," correctly adding the "-er" suffix, but reversing the word order. By conducting this same study in other languages, Clark determined that the order reversal was due to an order conflict inherent in English, German, Dutch, and related languages, but not existent in some other languages. She concluded that the structure of language has a clear effect on the language learning of young children.

In the Toddler rooms at Bing, Clark carried out a study of compound-noun use by very young children, and found that they invented some very creative compound nouns. Toddlers use compounds to contrast sub-categories of words, for instance, a "plate egg" is fried, while a "cup egg" is hard boiled. In some cases, a child will use compounds to organize larger categories. If you tell a child, "that's a poodle," for example, she may respond "poodle-dog." Toddlers will also use compound nouns to describe the inherent properties of an object, calling a car made of apple an "apple-car." In fact, Clark estimates that 30 to 40 percent of the words we all use each day aren't found in dictionaries, yet their meaning is transparent.

In conclusion, Professor Clark pointed out several general principles that seem to be true across different languages. Words with "simplicity of form" (no prefix or suffix) are easier for young children to understand, as are those with "semantic transparency," in which all forms are similar (e.g. children can't hear the relationship between "magic" and "magician," and may use a created word like "magic-man" instead). Similarly, two different forms of a word may have two different meanings for a child, especially if the forms are as disparate as "go" and "went." The forms of words that children use and understand will be determined largely by which forms the adults around them favor.

The staff was thrilled to have this visit from Eve Clark, particularly because she has been such an active researcher at Bing. We also shared with her some of the creative words Bing children have come up with recently. A favorite was the phrase coined by a Center Room child, who felt that if the area above the slide was called "upper sand," then the area down below should of course be known as "downer sand."