Pixabay free pictures
Source: Pixabay free pictures

Some of the most fascinating experiments in education occurred in the 1920s and ‘30s, and almost nobody talks about them today. That was an era when progressive ideas about education were in the air. Even public schools were experimenting with the idea that less teaching and more opportunity for self-direction would pay big educational dividends.

Benezet’s experiment on the non-teaching of arithmetic

In a previous post (here) I described an experiment conducted by L. P. Benezet when he was superintendent of schools in Manchester, New Hampshire, in the late 1920s and early ‘30s. He altered the curriculum for half of the schoolchildren in the poorest schools in his district, so they would not be taught arithmetic until 6th grade. He found that those children, at the beginning of 6th grade, before they had received any arithmetic instruction at all, performed much better than the others on math story problems—the kinds of problems that require common sense applied to numbers. They were even better on those than were the kids in the rich schools, all of whom had been studying arithmetic all along. Of course, they were behind the others in doing calculations (adding, subtracting, multiplying and dividing) set up in the usual school way, but by the end of 6th grade they had fully caught up to the others on that and were still ahead on story problems. 

Benezet concluded that the early teaching of arithmetic was not only a waste of everyone’s time, but was counterproductive to the eventual learning of arithmetic. In his words, the early teaching of arithmetic was “chloroforming the children’s minds,” leading them to lose their common sense about anything to do with numbers. Nobody in education talks about Benezet’s experiment these days. Few people in education seem to have even heard of it. Benezet’s results fit well with research showing that school children make greater gains in mathematical reasoning during summer vacation than they do during the school year (see here), which is another finding that nobody in education talks about.

Williams’s experiment in which delinquent boys were freed from being taught

Now here’s yet another bit of education research that nobody today talks about. It was published in 1930 in the academic journal School and Society under the title “An Experiment in Self-Directed Education,” by Herbert Williams, the teacher who carried out the research.

The practical problem Williams was trying to address was what to do about delinquent boys, who were frequently absent from school and were causing trouble in the community. For the sake of this experiment, he went through the Juvenile Court records for the city of population 300,000 and identified the “worst” boys he could find. To that group the school principals added a few more, whom they considered to be their “most serious problems.” He ended up with a group that “ranged in age from eight to nearly sixteen, in IQ from 60 to 120, and included colored, Polish, Hungarians, and native white Americans.” 

The experiment was started in January, 1924, and lasted until the beginning of June that year. During that period the boys were excused from regular school classes and, instead, were assigned to a special room created for them in a technical school. The room was equipped with desks, blackboards, a large table, and a collection of books, including storybooks, nonfiction works, and textbooks for the various grades. The boys were given standard academic achievement tests in January and again, four months later, in May.

And now, I know no better way to convey what happened than to quote Williams directly:

“No formal instruction was given. In the beginning of the experiment the children were told to keep busy and refrain from annoying any of the others. This was the only rule that was enforced. Otherwise, they were permitted to occupy themselves as they saw fit.  The instructor [Williams] from time to time passed from one to another to see what was being done. One child might be busily occupied in copying a picture from one of the books; another might be reading a fairy story; another occupied with a problem in arithmetic; another reading a history; others might be looking up places on a geography map; and still others would be studying about some machinery.

“Whenever a child was found manifesting an interest in some particular thing, opportunity and encouragement were given him to develop that interest...The child with an interest and aptitude for mechanical work was given an opportunity to do this sort of work in the high-school machine shop. The same was true for those interested in automobile mechanics, woodworking, printing and the like. Arrangements were made for recreation at the neighborhood YMCA…” 

“Each child was told of his accomplishments on the achievement test and encouraged to make up for any deficiencies, but he was not forced to devote his time to these. It was a revelation to the writer how these children turned naturally from one subject to another. A boy might spend an entire day on some book that he was reading. The next day he might devote to arithmetic. One 10-year-old became interested in working square root problems and worked all of these he could find in the arithmetic book. A colored boy became interested in history and read all the histories we could supply. His accounts of interesting historical events kept the entire group keenly interested as he related them. Whenever one of the boys found something in his reading which he felt would prove interesting he was permitted to tell it to the group. However, they were not required to pay attention to the speaker if they wanted to continue what they were doing.

“Many of the boys went to the blackboard to work arithmetic problems, primarily for the activity involved. They made up certain games involving arithmetic processes… For example, two or more boys would start at a given signal to add by seventeens to a thousand. The rivalry was often intense, and for some of the boys the increase in speed and accuracy in the fundamentals was striking. The reports of the various boys on interesting material read would stimulate other boys to read the same thing or something of like nature. It is quite possible, too, that the desire to obtain recognition from their fellows motivated them to do tasks that would not have been otherwise attempted.” 

“Although a total of twenty-six boys were in attendance in this special experimental group for shorter or longer periods, only thirteen were present for both the January, Form A, and May, Form B, Stanford Achievement Tests. This was due to out-of-school adjustments, transfers and other causes. Social adjustment was given first importance, and completeness of the experimental records was not allowed to prevent placing a boy on a farm, for example, if this met a pressing need.”

Here are the results from the achievement tests:

Over the 4 months period of this experiment, the thirteen children gained an average of slightly over 15 months in language age, 14 months in arithmetic; 11 months in reading; 11 months in science; and 6 months in both history and literature. By the end of the experiment all of these children were above grade level overall. The three boys who showed the least gains were also the three who, for reasons of health or family problems, were most often absent from the group. The average gains for the ten students who were regularly present were 17.4 months for language and arithmetic; 15.8 months for science; and 15.5 months for reading. 

In concluding the article, Williams wrote:

“The most striking fact is that such marked improvement could and did result from such informal, self-directed activity. The writer was not greatly interested in the educational development of these boys. The problem of social adjustment entirely outweighed it in his estimation. He used the special room merely to get better acquainted with the individual boys and to keep them from violating the compulsory attendance law. Whether they learned reading, arithmetic, geography, history and the other subjects was considered relatively unimportant…It should be remembered, too, that these boys spent less time in the classroom and more in shops and the gymnasium and on the playground than is usually the case...In accounting for this increase in educational achievement the writer can only surmise that...a personal interest on the part of the supervisor in each child’s home conditions, neighborhood, recreation, health and the like as well as an interest in the child individually may have stimulated the child.”

My own suspicion, not mentioned by Williams, is that age mixing also played a role. The boys ranged in age from 8 to almost 16. Self-directed education always works best in age-mixed environments (see here and here). Also, of course, these boys were free to spend as much time as they liked on whatever they were studying, which allowed them to dig much more deeply than is ever possible in a standard classroom; and because they were always free to talk with one another they learned from one another. While regular classrooms are perfectly designed to prevent the development and pursuit of genuine interests, this “classroom” did not prevent these.

Wouldn’t it be great if education authorities would take a look back at some of these old research studies and try repeating them today? Today education authorities seem to think the only solution to educational deficiency is more teaching—more of the same of what already isn’t working. But research such as Benezet’s and Williams’s suggests that the solution might lie in less teaching and more trust.

Basic Books, with permission
Source: Basic Books, with permission

As regular readers of this blog know, I’m not a fan of standardized academic testing, nor of any sort of school system that sees high scores on such tests as a primary educational goal. In my view (and I suspect Williams’s as well), the years that we think of as school years should be devoted to discovering who you are and what you like to do, to developing skills in what you like to do, to acquiring social and emotional competence, and to gaining the confidence that you can learn whatever you want, on your own initiative, at the time you need to know it. That all comes from real Self-Directed Education, where young people are free to explore the world in ways that are not dependent at all on a special room with textbooks, nor on encouragement to improve scores on someone else’s concept of “achievement.” Williams’s experiment is, to me, just one more example showing that the kinds of “achievements” that we fret and sweat about in our schools are actually quite easily and painlessly attained by young people who for one reason or another decide to attain them and are free to do so in the ways that work best for them.


And now, what do you think about all this? This blog is, among other things, a forum for discussion, and your views and knowledge are valued and taken seriously by me and other readers. Make your thoughts known in the comments section below. As always, I prefer if you post your comments and questions here rather than send them to me by private email. By putting them here, you share with other readers, not just with me. I read all comments and try to respond to all serious questions if I feel I have something useful to add to what others have said.


See also: Free to Learn; the website of the Alliance for Self-Directed Education; and follow me on Facebook.


Herbert D. Williams (1930).  Experiment in Self-Directed Education.  School and Society, 31, 715-718.

You are reading

Freedom to Learn

Another Example of Less Teaching Leading to More Learning

Delinquent boys made huge academic gains when freed from classroom lessons.

Mend the Gap Between Rich and Poor in School Achievement

We know how to reduce the school achievement gap, but we do the opposite.

Facts and Fiction About the So-Called “Summer Slide”

Do children really learn less during summer vacation? Less of what?