Language Log

« previous post |

Tuesday's Frazz:

George Miller was eloquently skeptical about the value of traditional vocabulary lessons, compared to everyday experience. From George Miller and Patricia Gildea, "How Children Learn Words", Scientific American 1987:

Our findings and those of other workers suggest that formal efforts to build vocabulary by sending children to the dictionary are less effective than most parents and teachers believe. […]

In the early grades schoolchildren are expected to learn to read and write. At first they read and write familiar words they have already learned by means of conversation. In about the fourth grade they begin to see written words they have not heard in conversation. At this point it is generally assumed that something special must be done to teach children these unfamiliar words. This educational assumption runs into serious problems. Although children can recognize that they have not seen a word before, learning it well enough to use it correctly and to recognize it automatically is a slow process. Indeed, learning a new word entails so much conceptual clarification and phonological drill that there simply is not enough classroom time to teach more than 100 or 200 words a year in this way.

Miller notes that informal learning through speech and reading is where the real action takes place, with about 13 words a day learned over the course of 16 years:

For the purpose of counting, a word can be defined as the kind of lexical unit a person has to learn; all the derivative and compound forms that are merely morphological variations on the conceptual theme would not be counted as separate words. For example, write is a word and its morphological variants (writes, writ, wrote, written, writing, writer and so on) are relatives in the same family. If such a family is counted as a single word and knowing a word is defined as being able to recognize which of four definitions is closest to the meaning, the reading vocabulary of the average high school graduate should consist of about 40,000 words. If all the proper names of people and places and all the idiomatic expressions are also counted as words, that estimate would have to be doubled. This figure says something about the ability of children to learn words. If the average high school graduate is 17 years old, the 80,000 words must have been learned over a period of 16 years. Hence the average child learns at the rate of 5,000 words per year, or about 13 per day. Children with large vocabularies probably pick up new words at twice that rate. Clearly a learning process of great complexity goes on at a rapid rate in every normal child.

If 2/3 of the Mrs. Olsen's test items was 16, then the whole list was 16*3/2 = 24 items long. So by Miller's calculation, Mrs. Olsen's 24-item vocabulary test was about two days worth of natural word learning.

Of course, natural word learning is a more gradual process, with several encounters in context generally involved for each item. Miller notes that an experiment from 1978 showed somewhat-effective learning from a single encounter (Susan Carey and Elsa Bartlett, "Acquiring a single new word", Papers and Reports on Child Language Development 1978). But he describes two stages of the word-learning process:

[A] child's appreciation of the meaning of a word seems to grow in two stages, one rapid and the other much slower. Children are quick to notice new words and to assign them to broad semantic categories. […]

The slow stage entails working out the distinctions among words within a semantic category. […]

This stage ordinarily takes much longer than the first and may never be completely finished; some adults, for example, correctly assign delphinium and calceolaria to the semantic field of flowering-plant names but have not learned what plants the words denote and cannot identify the flowers on sight. At any given time many words will be in this intermediate state in which they are known and categorized but still not distinguished from one another.

Miller's educational conclusion is that "a computer program providing lexical information about new words encountered in the context of a story might be more effective." And modern educational practice has largely learned this lesson, I think, though some Mrs. Olsens are no doubt still Out There.

March 6, 2025 @ 7:29 am · Filed by Mark Liberman under Linguistics in the comics

Permalink