Thorndike and cognitive learning theory
The strong, prima facie objection to Thorndike’s theory always was that it made learning transfer so limited that it was incapable of accounting for learning at all. It is common sense that problem solving abilities and abstract reasoning transfer to new and different tasks or problems.
Cognitive theories of learning have largely replaced earlier behavioral theories in this area. Moreover, while behavioral theories of learning like Thorndike's were the prevalent ones in the first part of the 20th century, there were prominent educational psychologists who rejected them even then.
One of the earliest critics was Alexander Meikeljohn, a famous educator and free speech advocate who served as dean of Brown University and later as president of Amherst College. In “Is Mental Training a Myth?” (Educational Review, February 1909, p. 130), Meikeljohn ridiculed Thorndike and Woodward’s theory:
What will you say of a theory that the training of the mind is so specific that each particular act gives facility only for the performing again of that same act just as it was before? Think of learning to drive a nail with a yellow hammer, and then realize your helplessness if, in time of need, you should borrow your neighbor's hammer and find it painted red. Nay, further think of learning to use a hammer at all if at each other stroke the nail has gone further into the wood, and the sun has gone lower in the sky, and the temperature of your body has risen for the exercise, and in fact, everything on each and under the earth has changed so far as to give each new stroke a new particularity of its own, and thus has cut off from all possibility of influence upon or influence from its fellows.
Harold Ordway Rugg (The Experimental Determination of Mental Discipline in School Studies, 1916, p. 116) was another prominent critic of Thorndike who defended transfer of training (another term at the time for what is now usually called transfer of learning). Although he later changed his views and became one of the leaders of the progressive education movement, in 1916 Rugg argued that students of descriptive geometry substantially improved their problem-solving abilities, and that from studies in school students developed the
ability to analyze the problems and to originate a method of procedure; to build up ideals, or to organize a method of attack. But it is undoubted that they also make habitual, or automatic, many specific constituents of the complex abilities that function in many complex situations.
In a work published in 1928, another educational theorist, Pedro T. Orata, saw that what Thorndike did was "profoundly misleading" because he had focused on narrow "habit formation and drill." Orata showed that psychologists who had trained students to understand "meanings, concepts, and principles or generalizations" had demonstrated considerable transfer of learning. Orata’s work was an early cognitivist view of learning, and quite remarkable when viewed against the backdrop of the dominant behavioral psychology of his time.
In 1950, when progressive educational theories were still very influential, Walter Kolesnik surveyed hundreds of studies and showed that in many cases transfer of learning does take place.
By 1966, Richard Hofstadter, in his classic work Anti-intellectualism in American Life, could describe Thorndike's theories and its dominance in American educational circles as "a major scandal in the history of educational thought."
Murray and Thorndike
Here is the connection I see between Thorndike’s behavioral theories of learning and Charles Murray’s views about education, and in particular his criticisms of the liberal arts and the B.A. degree.
Murray allows that a B.S. degree provides a genuine certification that the student has mastered measurable skills that mean something on the job. But there is always the strong suggestion in his writings that a liberal education and the B.A. degree, unlike training in the STEM fields of Science, Technology, Engineering, and Mathematics, do not involve the development of intellectual skills that are meaningful in the real world.
It is important to Murray that the graduate in a STEM field is qualified to build a bridge, do a chemical analysis, and solve equations that are used to build nuclear reactors or airplanes, etc. Similarly, the student who graduates with a concentration or major in accounting will have learned the rules of accounting and be able to keep the books in organizations and companies small and large. Real, measurable skills are also mastered in fields like medicine and law. But Murray seems to have no such conviction in the case of the liberal arts, where one apparently cannot point to a particular set of skills that can be applied in the real world. This comes very close to the Thorndikian view of learning.
One can imagine the identical elements involved in calculus problems that appear in college tests and real world tasks tackled by engineers and architects when they design and engineer buildings, bridges, and roads. And it is often emphasized that STEM fields involve a strict, steep, and rather well-defined "educational ladder," in which elementary courses are prerequisites for more advanced courses. One learns arithmetic, for example, which is a prerequisite for algebra, which is a prerequisite for trigonometry, which is a prerequisite for calculus, which is a pre-requisite for higher math, and so on. This sequencing of courses and requirements in important, of course, but it is not the same thing as learning transfer.
A little reflection will show that Thorndike's theory of identical elements could not be valid even as a theory of learning transfer for STEM fields. (Incidentally, learning theory in math and science is one of the most active fields of research in cognitive psychology.) It is true that students in these fields acquire a fund of knowledge and a repertoire of skills for dealing with particular sets of problems. But they are also developing along the way—though usually only implicitly—sophisticated models and ways of thinking and reasoning about mathematics and scientific problems and tasks. Real education in STEM fields is not just a matter of mastering a portfolio of tricks or skills, as Thorndike's theory of identical elements would have it. Instead, training in these fields is very much a matter of learning to think the way mathematicians and physicists and engineers think.
Though it is certainly not limited to graduate studies, the point is perhaps most apparent there. In graduate studies, it is not enough to simply learn a repertoire of specific skills. It is even more important to impart a general way of thinking and reasoning that will enable the graduate to advance the field itself. This can't be done only by teaching a particular skill set. (The attempt to find a heuristic or set of procedures for advancing science was once one of the cherished goals of positivist philosophers of science, but the project was abandoned as quixotic long ago.) Nor is it a matter of simply imparting important scientific facts. It is a matter of developing the ability to do research and discover presently unknown facts. The graduate in STEM fields must develop skills in solving new problems, and the developing of problem solving abilities is the very essence of learning transfer. Here the importance—and the reality—of learning transfer are particularly striking and obvious.
The Collegiate Learning Assessment test (CLA) and its findings
Since I had recently read Richard Hofstadter’s Anti-intellectualism in American Life and its strong condemnation of Thorndike’s learning theories when I read Murray and learned about the Collegiate Learning Assessment, I was able to appreciate the relevance of the CLA to Murray’s claim that the B.A. degree is meaningless. The belief that formal education has value relies largely on the notion that learning in school will prepare students for new and unexpected situations. The early findings from the first longitudinal study of the CLA indicate that the social sciences, humanities, and STEM fields do develop reasoning, critical thinking, and communication skills that are engaged in tasks faced by business managers and others in the "real" world. Thorndikian theories of learning cannot accommodate these findings.
The CLA is also relevant as I pointed out in “Is the B.A. degree meaningless?” to the issue of educational stratification—and hence at least indirectly to the issue of social stratification. Like Thorndike and the progressives, Murray believes that an academic curriculum can benefit only a relatively small minority of the population. He believes, in fact, that it can benefit only a minority of those currently going to college. He believes that most of the students who are presently attending college should drop out and do something else. But Murray’s position is directly challenged by the finding that there is no SAT score below which college students fail to show gains on the CLA’s performance tasks.
In his later years, Thorndike appeared more willing to concede that education can indeed train the intellect, but (like Murray), he believed that significant gains in academic skills would be limited to the most gifted in the population. Thorndike, who was one of the pioneers in IQ theory and testing, did not offer a break off point, but he did believe that those best endowed with intellect were the ones likely to make the most gains. This claim, too, has been undermined by the early results of the CLA, which found that that gains made by students on the CLA who score in the lowest decile on the SAT are as large as those made by students scoring in the highest decile.
The development of problem solving skills is an essential component of learning itself, and learning transfer of this sort is clearly engaged by the CLA's performance tasks. The development of such skills is also an essential component of what is valuable about education and the entire academic enterprise. In a real sense, it makes people more intelligent. That is why it is significant that the early results show an improvement on a quintessential test of learning transfer: taking broad reasoning and communication skills learned in an academic curriculum and applying them to a real world performance task on the CLA.
On politics and the historical contexts
There is a chapter in Left Back (“IQ Testing: ‘This Brutal Pessimism’”) that is devoted to the history of intelligence testing and its close association with progressive education. It covers the ascendancy of the intelligence testing movement in education by educational psychologists like Edward L. Thorndike, Louis M. Terman, Robert Yerkes, and Carl C. Brigham. It also includes a short section, “Reaction Against the IQ Concept,” in which Ravitch cites the growing doubts of men like Brigham and Terman about their own views on IQ, long after those views had become dominant in education. More careful studies of the connection between years of education and the quality of education and the IQ scores of individuals and racial and ethnic groups had undermined their theories significantly. It had also led to widespread questioning of their view that only a relatively small minority of the population could benefit from a high quality, academically oriented curriculum.
Significantly, the American public never bought the progressives’ “brutal pessimism” about intelligence. (The phrase, which contains a criticism of the misapplication of IQ testing, is from Alfred Binet, the father of psychological and IQ testing.) Public resistance was, understandably, particularly strong among blacks, Hispanics, and immigrants from southern and Eastern Europe (including Jews), who were regarded by educational psychologists as genetically inferior and incapable of benefiting from an academic curriculum. These groups, in particular, saw the progressives’ “brutal pessimism” as a classist and racist effort to keep them from bettering themselves and their children. The leading educational theorists of the day would have consigned most of the students in these groups to vocational training, beginning at a quite early age.
Progressive educators opposed academically oriented public secondary schools, while the American public supported them strongly and were intensely proud of them even in periods of American history when jobs were plentiful and one didn't need a secondary education credential to get a job. This last point is important. Today, Murray and others allege that most Americans believe in the college degree only because of the credentialing factor, and that it serves this function only because business uses it as a screening device for good jobs. But the historical situation during the 19th and early part of the 20th century was very different. At that time there was a plenitude of vocational and industrial jobs. Furthermore, business in this early period was not interested in academically oriented high schools and post-secondary education. It was interested instead in lower taxes and in having a pool of well-trained cheap labor for industrial jobs. Even then, however, the new educational theorists—the Sneddens, the Thorndikes, the Rosses—couldn't sell most Americans on non-academic education for their children.
For all the similarities between them, there are two important historical differences between the progressive educators of the past and Murray’s conservative/libertarian views. It seems to me, though, that the differences make the progressives’ views more understandable and defensible, and Murray’s less so.
First, since the heyday of progressive education, we have moved towards a much more complex society and economy. In today’s information-based and technologically sophisticated society, intellectual and cognitive skills have become far more important than they were in the 19th and early 20th centuries. Globalization has forced the American work force to compete increasingly with labor forces abroad, and many blue collar jobs have already been lost to foreign competition. It is true that many tech jobs have been outsourced abroad as well, but America’s competitiveness is still greater in jobs requiring technological and intellectual skills than in blue collar jobs. In fact, our competitive disadvantage in the global market place for jobs that do not require cognitive skills is likely to deteriorate steadily. Parents and students understand this. This is an important part of the reason for the continuing demand for higher education despite its steeply increasing costs.
Second, the progressives’ views are more defensible and understandable than Murray’s often similar views because they could believe in social engineering in a way we do not and cannot now. Some of the statements of early educators, like those of sociologist Edward A. Ross in 1901 that free public schooling was an “engine of social control,” sounds to the contemporary ear more like propaganda emanating from the Politburo or a Petrograd Soviet in the 1920s than to the views of educators in the American tradition.
In order to understand the progressives, it is also necessary to place their views in the context of socialist and Marxist thinking. (Ravitch covers some of the issues in her chapter “On the Social Frontier,” which explores the progressive’s attraction to Soviet educational theory and practice.) To be sure, progressive educators were not socialists or Marxists, but there can be little doubt that progressive theories of education developed under the long shadow of socialism and Marxism.
Marxists, after all, had clearly articulated views about “worker culture” and proletarian culture, the whole purpose of which was to denigrate the Western intellectual and academic tradition as anti-people, feudal, or bourgeois. Most socialists and Marxists of the day valued homo faber over homo cogitans, and rejected the academic, liberal arts tradition in favor of education (or more precisely vocational training) that "fits the child for work in the real world and produces or conduces to his or her higher development physically, intellectually, and morally."
It is an interesting question how many of the leading theoreticians in the Marxist and socialist movements believed this ideology, since almost without exception they were well-educated individuals themselves. Many of them were in fact academics. Some were undoubtedly hypocrites, and many were probably self-hating intellectuals. (Trotsky was one Marxist ideologue who believed that there was not and could not be a proletarian culture, but he was an exception.) In any case, Marxist ideology, combined with an unfettered belief in social engineering, were important influences that allowed the Sneddens of the progressive movement to believe that educators could adequately train students for lifetime careers in tailoring, jewelry salesmanship, poultry farming, and coal cutting. We can no longer believe in this project, since we no longer share either the ideology or the belief in social engineering that buttressed it.
Like the socialist and Marxist philosophers of the not-so-distant past, Murray often appears to present intellectual and academic values as elitist, even aristocratic, class values, like a taste for fine wines or fine art—of no real educational or intellectual significance or import. He sometimes describes the traditional liberal arts curriculum as a kind of class luxury, reminiscent of Thorstein Veblen’s theory of conspicuous consumption. Like the gentleman scholars of the Oxbridge universities in the 19th century, who went there to become cultured and educated, but without any of the passionate—even deep moral commitment—to either the scholastic or classical learning that was an essential component of those two universities in their formative years when they grew to greatness, America’s elite can be equally indulgent, if they are so inclined, but that is all.
Murray gives this impression despite the fact that it runs counter to his convictions about the importance of educating his elite, because he shows little or no interest or concern for the purely intellectual values of the liberal arts curriculum—the aspect of it that educators in previous days called its “mental discipline.” When Murray speaks of the liberal arts, it is usually only about the part of it that he believes is still capable of sustaining and transmitting conservative and traditional cultural values (albeit with a nod at one point to Aristotelian and Confucian virtues as well). To that extent, Murray does stand within the conservative tradition. He departs from it, however, when he asserts that liberal education “in a real sense” is only of potential benefit to an intellectual class. Ravitch’s book shows that such a view of liberal education has more affinity with the progressive educational theories of the past than it has with the conservative educational tradition.