From Laws of Learning to a Science of Values Efficiency and Morality in Thorndike's Educational

Barbara Beatty Wellesley College

Edward L. Thorndike's educational psychology was the of human thought and action. Despite these intellectual beginning of an American behavioristic tradition that and cultural tensions, most turn-of-the-century psycholo- sought efficient, scientific solutions to educational, gists continued to do research on both intellect and char- moral and social problems. Thorndike used empirical acter and to see them as interrelated topics. methodology to explain behavior, intellect, and character. Psychologists in the Progressive Era were con- After rejecting developmentaIism, he combined laws of fronted with enormously difficult, contentious social is- learning derived from his experiments on animals with sues. Rapid industrialization, urbanization, immigration, quantitative measurement of individual differences in hu- poverty, violent labor disputes, changing relations both mans to construct a psychology of education. He applied between the sexes and within the family, and growing this educational psychology commercially and developed global unrest caused great dislocations and instability. many widely used tests and texts. Thorndike then pro- Psychology was pulled out of to help explain, posed a science of values that he hoped might be used contain, cure, and control these social ills, especially in as a guide for moral assessment and social policy. education (Danziger, 1990). Educators looked to psychol- ogists for assistance in categorizing, socializing, and in- structing the flood of immigrant and poor children enter- n the first sentence of the expanded 1911 edition of ing urban schools (Brown, 1992; Chapman, 1988). Edu- Animal Edward L. Thorndike listed "in- cational psychology arose in response to these practical ~ educational needs, and as a means of professionalizing tellect" and "character" (Thorndike, 1911, p. 1)as the two topics of behavioristic psychology. Thorndike education and expanding the profession of psychology. researched and reworked these themes throughout the Thorndike's project of creating a science of educa- successive phases of his long career. Following his path- tion was not new. Earlier nineteenth-century attempts had breaking animal experiments, he found employment in included deriving pedagogy from classroom practice; teacher education. He briefly explored G. Stanley Hall's from psychological philosophy, especially that of Herbert child study, but rejected developmentalism on intellec- Spencer and Alexander Bain; and from European peda- tual, methodological, and moral grounds and began mea- gogical theories such as those of Pestalozzi, Froebel, and suring individual differences. In the years before World Herbart (Roberts, 1968). G. Stanley Hall amassed survey War I, Thorndike combined learning theory, psychomet- data that he claimed constituted a scientific approach to rics, and applied research on school-related subjects to the study of children and education (Ross, 1972; White, form a psychology of education. In the 1920s, he helped 1990; Zenderland, 1990). Psychological methods were turn educational psychology into a mass-market industry also being used by educators outside of academia. But and produced numerous commercially successful tests some psychologists, such as , were skepti- and textbooks. In the final phase of his career in the cal as to whether education could become a science 1930s, Thorndike proposed a science of values and devel- (James, 1899; and see Beatty, 1996; Brown, 1992; Ca- oped quantitative indices of moral and social goodness. han & White, 1992; Cuban, 1993; Danziger, 1990; Klie- Thorndike's positing of intellect and character as bard, 1986; O'Donnell, 1985; White, 1991). the dual themes of behavioristic psychology reflected the unified view of truth and morality characteristic of nine- teenth-century educational philosophy. Character educa- Research for this article was supported in part by a grant from the tion was one of the main goals of public schooling (Ty- Spencer Foundation. I am especially grateful for the encouragement ack & Hansot, 1982). Religious and intellectual knowl- and suggestions of Geraldine Jonqich Clifford, whose definitiveb iogra- phy of Thorndike provided the groundwork for this article. The com- edge were linked in higher education as well (Reuben, ments of Emily Cahan, Bennett Galef, Ken Hawes, Richard yon Mayr- 1996). Darwin's theory of threatened this unity hauser, and Sheldon H. White, and the secretarial skills of Adele Rosen- of knowledge and morality, though Darwin was also in- thal were very helpful. Any errors of interpretation are, of course, my terested in explaining altruism (see Curti, 1980; Fancher, o w n . Correspondence concerning this article should be addressed to 1996; Richards, 1987; Ross, 1991; Sulloway, 1998). The Barbara Beatty, Departmento f Education, WellesleyC ollege, 106 Cen- construction of an empirical science of psychology posed tral Street, Wellesley, MA 02181. Electronic mail may be sent to a particular challenge to traditional religious explanations bbeatty@ wellesley.edu.

October 1998 • American Psychologist 1145 Copyright 1998 by the American Psychological Association, Inc. 0003-066X/98/$2.00 Vol. 53, No. 10, I145-1152 However different psychologists in the Progressive transmitting.., thought and feelings, which might exist Era conceived of the relationship of science, education, apart, but as light penetrates through transparent sub- and morality, most equated science and efficiency. They stances, so might they appear in connection with . . . believed their research would aid in the creation of a human brains" (Thorndike, 1901a, p. 210). The debate more rational, orderly, beneficent society (see Callahan, concludes with the passage from Plato's Apologia in 1963; Kloppenburg, 1986; Wiebe, 1967). G. Stanley Hall which Socrates goes to his death saying that whether or used developmentalism to guide educational, moral, and not there is an afterlife one must act on the basis of social decision making (Ross, 1972). was conscience (Thorndike, 1901a, pp. 212-213). also concerned with preventing waste and building intel- G. Stanley Hall's child study methodology was the lect and character and attempted to merge education, mo- topic of Thorndike's next major publication, and of some rality, and society through metaphors of organic growth of his courses at Teachers College, where he was hired and democratic community (Cahan, 1992; Ryan, 1995). in 1899 as an instructor in genetic psychology. Notes on Thorndike used connectionist learning theory to opera- Child Study, which appeared in 1901, contained criti- tionalize the relationship of education, morality, and soci- cisms of Hall but sounded some typically Hallian moral ety. How did Thorndike deal with intangible issues in and social themes. Thorndike recommended the morally morality and social policy? How did Thorndike attempt beneficial effects of fresh air, exercise, and healthy com- to maximize both efficiency and morality? How did panions. But he advocated athletic "games and social Thorndike evolve from thinking science was separate c l u b s . . , for girls as well as boys" (Thorndike, 1901b, from human ideals to believing that science should help p. 127), unlike Hall, who was notoriously anxious about set moral and social goals? preserving masculinity. Thorndike was particularly criti- Studying Children and Rejecting cal of Hall's untrammeled developmentalism. For Thorn- dike, the interesting aspect of studying children was not Developmentalism how they were developmentally the same, but how they Thorndike was initially drawn from animal psychology were individually different. General statements about into education for professional reasons. Like other young children "must be false," Thorndike wrote, "for no two psychologists who received doctorates around the turn of children are alike mentally" (Thorndike, 1901b, p. 14). the century, he encountered a dearth of academic posi- All statements about children were probabilities, Thorn- tions in psychology (Jon~ich, 1968). Psychology enroll- dike asserted, probabilities which could be stated with a ments grew more rapidly in the 1890s than did the de- level of statistical accuracy. mand for psychologists. As Hugo Mtinsterberg wrote in Thorndike knew that children were different from 1898 to James McKeen Cattell, "my elementary psychol- adults and included in his work a chart of children's ogy c o u r s e . . , has 360 students---what will this country developmental stages (Thorndike, 1901b, p. 13), but he do with all these psychologists?" (quoted in Brown, thought these differences were incremental. Even reason- 1992, p. 65). The answer was that many of them, like ing, which Charles Judd championed as a qualitatively Thorndike, would find jobs in child study and teacher different kind of higher order process, was essentially education programs (Dewsbury, 1992; O'Donnell, 1985). incremental for Thorndike. Thorndike asserted that all Thorndike' s early publications, after Animal Intelli- the rudiments of rational thought were present by the gence, showed the influence of William James and G. time a child entered school "and in fact long before Stanley Hall, two older psychologists who had written then" (Thorndike, 1901b, p. 86). The great importance about education and morality. James's well-known inter- of schooling was that it was one of the main ways chil- est in spirituality was evident in Thorndike's book The dren honed their reasoning. Step by step, "little by little," Human Nature Club, first published in 1900, which Thorndike wrote, "through constant correction and revi- Thorndike hoped would sell as well as James's very suc- sion," children gradually learned academic skills and cessful Talks to Teachers (1899). Intended for a popular knowledge, until they acquired "the usage of science or audience of high-school and teacher-education students literature of educated people" (Thorndike, 1901 b, p. 87). and adults interested in self-education, The Human Na- And step by step, Thorndike honed his own views on ture Club was written in the form of a fictional dialogue education, society, and morality. among members of a study group. After conversations Thorndike was vehemently opposed to Hall's notion on Jamesian themes of sensation, attention, , that the school curriculum should follow nature and reca- emotions, and so on, the club turns to "deeper questions" pitulate the stages of human cultural development. Not of free will and immortality. A minister joins the group only was Hall's developmentalism intellectually uninter- and expresses what Thorndike may have wished his own esting and methodologically inexact, Thorndike thought deeply religious, Methodist minister father might have it was immoral and potentially dangerous. Schooling said--that though the Bible taken "as a piece of history" should improve upon nature. Not everything in nature provided "evidence for continuance of mental life apart was good; there were base instincts as well as good ones. from the body," the psychological ideas the group was Educators should use the impulses of nature only if they discussing did not (Thorndike, 1901a, pp. 201-202). One "further the aims of education . . . . when they work of the characters paraphrases James's Human Immortal- toward moral ideals" (Thorndike, 1901b, p. 136). "What ity, saying that "nerve cells might be just the means for development is can never teach what it ought to be,"

1146 October 1998 • American Psychologist Thorndike stated, condemning Hall for committing the for more experimental studies of school learning, and he naturalistic fallacy. "No word perhaps is a poorer syn- did some research in schools during the early years of onym for 'the good' than 'the natural' (Thorndike, 1901b, his career (Jongich, 1968, p. 231). p. 136). Thorndike infused this school-related work with the- ories derived from his research on animals and statistical Measuring Individual Differences and analyses of individual differences in humans. Here again Constructing Educational Psychology his intellectual program, research agenda, and publica- Although Thorndike rejected Hall's developmentalism tions were influenced by the demands of his teaching and genetic psychology, he did not reject the genetic com- responsibilities and desire for professional advancement. ponent of individual differences. For Hall, genetic psy- By 1902 he was teaching Education 3, "Applications of chology meant the history of the development of the hu- Psychology in Teaching," a required elementary educa- man race; for Thorndike, genetic psychology meant tion course with large enrollments, which he had peti- biologically inherited characteristics in individuals. tioned dean James Russell that he be given to teach. He Thorndike spent the next decade and a half constructing changed its title to "Educational Psychology" and added an empirical psychology of education based on connec- a graduate level course, Education 3, "Application of tionist learning theory, statistical analyses of inherited Psychological and Statistical Methods to Education" and acquired individual differences, and other school- (Jonqich, 1968; Thorndike, 1901c). related research. The broad range of topics in his research As Thorndike focused more on statistical methodol- course, Psychology 13, reflected Thomdike's changing ogy, his work became increasingly filled with graphs, interests. The syllabus for the 1900-1901 academic year curves, and charts. This wealth of quantitative informa- included ' 'mental life of the primates," "verbal discrimi- tion was intended to help teachers and educational admin- nation in young children," the "correlations between the istrators deal with practical educational problems. For mental functions involved in school subjects," and the example, Thorndike studied the question of whether boys value of spelling and Latin as formal disciplines (Thorn- and girls should be educated differently, an issue about dike, 1901c, p. 10). which G. Stanley Hall, Dr. Edward Clarke, and others Some of this new research resulted in the publica- had made a great deal of noise, even though coeducation tion of a study on transfer of training which was as had proceeded quietly in most American public schools influential as Thorndike's earlier work on animal intelli- (Tyack & Hansot, 1990). After analyzing empirical data, gence. In association with his col- Thorndike stated that the "differences in ability" were league Robert S. Woodworth, Thorndike collected and "not of sufficient amount to be important in arguments correlated data from experiments on adults' ability to concerning differentiation of the curriculum or of meth- recognize misspelled words, ascertain accurate weights, ods of teaching in conformity with sex differences" add numbers, and associate word opposites. After only a (Thorndike, 1903, p. 118). few months, Thorndike and Woodworth reported that Although Thorndike did not think there were sig- there was relatively little transfer of training from one of nificant sex differences in intellectual ability, he thought these disparate mental skills to another. This finding, there was a strong genetic component to individual varia- which appeared in The in 1901, tion, including variation in morality. He published studies provided what was seen as one of the first empirical on heredity and on twins in 1903 and 1905, quoted Galton refutations of the concept of mental discipline, the pre- frequently, and thought intellect had such a high, fixed vailing theory in which the mental effort of learning sub- genetic component that, to be efficient, schools should jects such as Greek or Latin was thought to improve the group children into different classes and programs by powers of the mind generally and to aid in learning other ability. Thorndike's hard-nosed position on genetic limi- subjects (Thorndike & Woodworth, 1901). tations on human potential and the inefficiency of provid- Thorndike and Woodworth's (1901) findings ing equal education to all children was to become one spurred the growth of educational psychology by pointing of the main, longstanding criticisms of his work (see to the need for new studies on the learning of specific Jonqich, 1962, 21-22). "It is wasteful," he wrote in the academic skills and subjects. As Thorndike and Wood- 1903 version of his Educational Psychology, "to attempt worth concluded, the to create and folly to pretend to create capacities and interests which are assumed or denied to an individual next steps in the study of the interdependence of mental func- before he is born" (p. 44). Thorndike also thought there tions would seem to be the exact analysis of the influence of was a genetic component to morality. But in contrast to one on the other where such is present and the discovery of its his views on intelligence, at this early stage in his career amount and nature in cases of practical importance, for in- stance, in the case of the training given in school subjects. he felt that environmental influences had a greater impact (Thorndike & Woodworth, 1901, p. 563) on character than on intellect, a position he would later modify. "The important moral traits seem to be matters Although Thorndike remained personally ambivalent of the direction of capacities and the creation of desires about doing school-based research and relied heavily on and aversions by environment to a much greater extent data gathered by his graduate students and other research- than are the important qualities of intellect and effi- ers, his work with Woodworth convinced him of the need ciency" (Thorndike, 1903, p. 45).

October 1998 • American Psychologist 1147 Using empirical data to make specific educational institutions and methods made and reported with quanti- decisions was the raison d'etre of Thorndike's educa- tative precision" (Thorndike, 1906, p. 163). Without such tional psychology. He did not think there was, or ever meticulous testing, theories of education were mere could be, a grand, overarching theory of education• Nor speculation• was he interested in creating one. His goal was to mea- Thorndike knew the science of education was in its sure individual and group differences and the myriad infancy and worried that its complexity and labori- particularities of human learning. As he wrote in his 1903 ousness might put many psychologists off. But he was Educational Psychology, "there is no chance for any firmly convinced that education could be made into a simple general theory"; the "true general theory must science and proceeded to do the organizational and pro- be the helpless one that there can be no general theory" fessional work necessary to do so. In the lead article of (p. 163). What educational psychology could do was the first issue of The Journal of Educational Psychology, more modest• It could provide educators with lots of which Thorndike founded in 1910, he presented an detailed information• "Multiply the number of different agenda for how psychology could contribute to education changes desired by the number of different original na- and vice versa• Psychology could make the goals of edu- tures to be changed and the resulting number of concrete cation clearer and more exact, could measure the proba- problems will measure the number of separate concrete bility that these goals were attainable, and could enlarge precepts which the art of education must include," he and refine the aims of education• It could also contribute concluded, with a specificity that would carry over to his to understanding the content and means of education and views on moral education and later attempts to measure to improving of methods of teaching (Thorndike, 1910, morality (Thorndike, 1903, p. 163). pp. 5-7). Thorndike also thought that education could In the years before World War I Thorndike wrote in turn help psychology• "The science of education can about instructional methods and assessment and estab- and will itself contribute abundantly to psychology" lished educational psychology as an academic discipline• (Thorndike, 1910, p. 12). Psychology laboratories were In his first pedagogical work, The Principles of Teaching, not the only places where valid scientific research could published in 1906, he applied his learning theory to class- be done. "School-room life itself is a vast laboratory in room teaching• Designed as a text for Education A, the which are made thousands of experiments of the utmost largest course at Teachers College, this widely used book interest to 'pure' psychology" (Thorndike, 1910, p. 12). combined summaries of basic psychology with practical In Education, his book on educational philosophy, exercises for student teachers• Thorndike emphasized which was published in 1912, Thorndike set forth his throughout that there were no shortcuts to learning. utilitarian views on education, morality, and the behavior- "Each special task adds its mite to the general store•" istic approach to moral education• In "the broad sense" "Intellect and character" were strengthened "not by any morality was "simply such thought and action as pro- subtle and easy metamorphosis, but the establishment of mote the improvement and satisfaction of human wants•" particular ideas and acts under the law of habit" (Thorn- The " 'right' thing to do," Thorndike said, was that dike, 1906, p. 247). In the chapter on moral training he "which a man who could foresee all the consequences included a long, convoluted quotation on the "good and of all acts, and who considered fairly the welfare of all efficient character" from his Elements of Psychology, men, would in that case choose." The "aims of education which stated that character was in part dependent on "the as a whole" were thus "identical with those of morality" presence of worthy ideals" (Thorndike, 1906, p. 179). (Thorndike, 1912, p. 29). Thorndike admonished teachers He laboriously explained each step in the process of to model positive moral behavior, rather than talk about character formation and stated that although moral train- it. Using bullying as an example, he gave practical advice ing in the school was more difficult than in the home, on what today would be called behavior modification because of limitations of time, class size, and other cur- techniques, in which undesirable behavior could be in- ricular subjects, school education had "high moral hibited by being ignored, followed by "discomfort," or value•" Schools helped teach the everyday virtues, the substituted with positive behavior (Thorndike, 1912, p. small "ordinary moral acts" that Thorndike valued as 200). With characteristic concern for efficiency, Thorn- much or more than dramatic moral choices (Thorndike, dike advised that positive reinforcement was more effec- 1906, pp. 106, 192). tive than punishment• Associating "good responses" In the final chapter of Principles of Teaching, Thorn- with "satisfaction is in general preferable to the elimina- dike stated the scientific creed of his educational psychol- tion of bad responses by pain or deprivation•" In situa- ogy. The "scientific study of teaching," he said, rested tions where there were many possible responses punish- on testing for results. "Testing the results of teaching ment could be "very wasteful" (Thorndike, 1912, p. and study is for the teacher what verification of theories 201). Finding the right mixture of "incentives and deter- is to the scientist,--the sine qua non of sure progress" rents" depended on the particular case and was "an intri- (Thorndike, 1906, p. 264). " A true educational science cate problem" (Thorndike, 1912, p. 202). • . . must be made up from the study of the particular In his 1913-1914 magnum opus, Educational Psy- facts in answer to thousands of different questions," chology, Thorndike moved closer to a biological view of Thorndike stated, and "must rest upon direct observa- morality in which ideals were a product of evolution. tions of and experiments on the influence of educational Dedicated to the memory of William James, the three

1148 October 1998 • American Psychologist volumes dealt exhaustively with "man's original mental because they met educators' practical needs. His arithme- equipment--the inherited foundations of intellect, mor- tic books were easy for teachers to use and fit well with als, and skill," "the laws of learning," and individual existing curricula and methods. His tests provided admin- differences (Thorndike, 1913-1914, p. vii). Thorndike istrators with information on student achievement in a formalized his research on learning into his three famous format that was useful and convincing for reporting pur- laws, "the laws of Readiness, Exercise and Effect" poses. And the standardized results of his assessments (Thorndike, 1913-1914, p. 1). He now stated that al- provided arguably fairer rationales for the selection func- though there was "warfare of man's ideals with his origi- tions society demanded of schools than did teachers' sub- nal tendencies," "ideals themselves came at some time jective ratings and grades. from original yearnings in man" (1913-1914, V.I, p. True to his early rejection of developmentalism, 311). In his 1903 version of Educational Psychology Thorndike did not try to alter subject matter but accepted Thorndike had still thought there were some intangible the traditional curriculum. His goals of making school questions that could not be usefully addressed by facts. curricula more rational and efficient and eliminating Answers to the question of the aim of education, for waste and redundancy were ideas that American teachers instance, of what "people ought to be," would come and parents could understand and accept. This public from "conceptions of ultimate values" and would be acceptance and commercial success, combined with the "judged not by facts but by ideals," (Thorndike, 1903, scientific "look" and usefulness of educational psychol- p. 163). A decade later, Thorndike's views on ideals were ogy, explains much about why Thorndike's methods won more genetic. And as his career progressed, there would out over those of John Dewey (see Lagemann, 1989) and be fewer reservations about questions which facts could other more socially radical and child-centered educa- not helpfully address or answer. tional philosophers and psychologists. Publishing Tests and Texts and Measuring Morality and Proposing a Commercializing Educational Psychology Science of Values World War I provided confirmation of the usefulness In the last phase of his career, Thomdike applied some of Thorndike' s educational psychology and served as a cata- the methods of educational psychology to broad societal lyst for him to apply his research on a larger scale. The issues. Supported by a large grant from the Carnegie disappointing results of recruits' performance on the Corporation and by the staff and facilities of Columbia' s Army intelligence tests, which Thorndike had helped de- Institute for Educational Research, he now had the time, sign and analyze, gave further impetus to the movement resources, and inclination to focus on the social and to develop widely available psychological products. Edu- moral questions that had been underlying themes cational psychology boomed as a mass-market industry throughout his earlier work. Although the Depression had in the 1920s. In the years after the war, Thorndike pro- shaken faith in the power of science to solve social prob- duced a number of commercially successful school mate- lems, Thomdike remained a firm believer. He thought rials. In addition to the rating scales for handwriting, psychology could serve as the basic science for the rest English composition, and drawing he had developed ear- of the social sciences (Jonqich, 1968) and began collect- lier, he created a word frequency book for teachers, dic- ing data on industry, crime, labor, management, consum- tionaries, and tests to measure oral and silent reading, erism, government, law, economics, and all manner of geography, spelling, and other academic skills. other topics. Thomdike became involved in commercial ventures In Your City (1939), and other of the many studies that extended the influence of his work. Along with Cat- that resulted from this enormous project, Thorndike went tell and Woodworth, he founded the Psychological Corpo- beyond measurement of specific facts to making norma- ration in 1921 and standardized and copyrighted numer- tive and comparative evaluations of generic goodness. ous educational and psychological tests. His most influ- After statistical analyses of "over three hundred items ential piece of curriculum "psychologizing" was his of fact" on cities with populations of over 30,000 in Thorndike Arithmetics series, published by Rand- 1930 (Thorndike, 1939, p. 21), he developed a "good- McNally, (Thorndike, 1917, 1922) which sold extremely ness" index based on 37 supposedly significant variables well and augmented his academic salary handsomely ranging from mortality rates, crime, home ownership, (Jonqich, 1968). These tests and texts were used through- and teachers' salaries, to "per capita park acreage" and out the and probably had more direct, last- circulation of Better Homes, National Geographic, Good ing impact on children and schools than anything else Housekeeping, and Literary Digest (Thomdike, 1939, pp. Thorndike did. 29-31). Aside from the almost comical subjectivity and Thomdike's educational products were successful class bias of some of Thorndike's variables, Your City in part because they were based on empirical research shows how hard it was to do social science before the and appeared to be scientific. He claimed that the messi- advent of computers and multivariate regression analysis. ness of classroom teaching and learning could be trans- Thorndike painstakingly calculated correlations formed into a science; the mystique of science and his among his variables and other statistics and reported his wealth of charts, curves, and statistics made his claims findings in detail. A "high percentage of Negro fami- seem credible. Thorndike's products were also successful lies," for instance, was "a bad sign" for a city's good-

October 1998 • American Psychologist 1149 ness index (Thorndike, 1939, p. 77). Not surprisingly, the of imbeciles and ignoramuses, who do not know what is cities with the lowest " G Scores" were in the South, for their own good, much less what is for the good of although Woonsocket, Rhode Island, fared poorly as well. others" (Thorndike, 1940, p. 351). Unabashedly heredi- The highest scoring city was Pasadena, California tarian and meritocratic, Thorndike thought more tradi- (Thorndike, 1939, pp. 33-34). Thorndike then showed tional, moral, political, and humanitarian views did not readers how they could measure their own cities, using give enough weight to individual differences and was a simple "ten-item city yardstick" that involved calcula- determined to find a scientific way to do so. "Ethics, tions such as obtaining the annual infant mortality rate, politics and philanthropy have been guilty of neglecting subtracting this number from 120, and multiplying the individual differences, partly because doing so simplifies result by 2 (Thorndike, 1939, p. 153). Simplistic counting all problems, and partly because of the retention of theo- and numerology such as this was characteristic of the logical and sentimental prejudices in favor of the similar- 1920s and 1930s, when statistical surveys were the rage ity and equality of man" (Thorndike, 1940, p. 369). in American education (see Lagemann, 1998). Thorn- Using methods similar to those he had used to assess dike's work provided the methodology for this kind of handwriting and cities, Thorndike proposed a detailed, quasi-scientific research. quantitative moral rating scale. This "system of weights" That Thorndike was trying to quantify a measure of would combine both intellect and character and assign goodness beyond a simple urban quality-of-life index numerical scores for each. As on an intelligence test, an was apparent in his last major book, Human Nature and ordinary man would get a score of 100, while "Newton, the Social Order, which was published in 1940, the same Pasteur, Darwin, Dante, Milton, Bach, Beethoven, Leo- year he resigned from the Teachers College faculty. nardo da Vinci, and Rembrandt will count as 2000, and Thorndike returned to some of the religious and moral a vegetative idiot as about 1" (Thorndike, 1940, p. 372). questions he had dealt with in a lighter vein 40 years Points would be added for "unselfishness, benevolence, earlier in The Human Nature Club. On the question of and cooperativeness," exemplified by "Jane Addams, the existence of life after death Thorndike now stated Madam Curie, Sidney and Beatrice Webb," and sub- unequivocally that science was a more powerful explana- tracted for "meaness and cruelty" (Thorndike, 1940, pp. tory system than religion. Although science could not 372-373). Young children would get extra points de- provide information about the existence or nature of a pending on their age, because they were "innocent," but "supernatural world," Thorndike argued that the value no points were to be added for sex, family, wealth, or of believing in life after death should be viewed empiri- religious creed (Thorndike, 1940, p. 373). An individual cally and tested by its consequences. "If men had, during child's score, however, was to be determined in part by the past hundred years or thousand years, lived in the those of her parents. Thus, "twenty-five percent of the belief that the death of the body was the end of the person, plus or minus difference of his parents' average weight who can be sure that they would have been less moral?" from 100 is combined with each child's intrinsic credit he asked. And in the future, would people be better off until the age of twenty-five" (Thorndike, 1940, p. 373). retaining traditional religious beliefs about immortality If a child's background was unknown, credits or penalties or by believing "that the fate of all men rested entirely would be added "for being a member of a certain racial with nature"? (Thorndike, 1940, p. 147). Thorndike stock" (Thorndike, 1940, p. 373). thought answers to these and other "hard" questions One of the most troubling aspects of this supposedly should be determined by measuring the effects of such scientific moral weighting scale, other than its obvious beliefs. racial bias and subjectivity, was Thorndike's belief that In Human Nature and the Social Order, Thorndike people who came out with high numbers were, and should proposed the creation of a science of values that could be, more powerful. "Effective valuation," was "the total inform moral and social decision making. He wanted to net result of the valuations of all the persons concerned, explore the "possibility and desirability of the existence each weighted by the person's power" (Thorndike, 1940, of a natural science of values" that would "progress p. 388). Thorndike was aware of the unfair, arbitrary, from and improve upon the best present opinions about racist, and politically biased implications of disparities in what is good and what is bad by studying consequences power. He said, for instance, that a "magnate in business, of various conditions and events for the satisfaction of government, the church, or literature" might see himself wants present and future" (Thorndike, 1940, p. 347). as "God, or God's special representative," and "grossly He rejected two other approaches to social valuation: overrate the value" of his own wants, "or his family, or religious and democratic methods. Thorndike said these his church, or his dogs, or white men, or artistic people, "inferior procedures," trying "to discover what God's or thrifty people, or members of the Communist party" will is," and putting "it to a vote of all citizens, each (Thorndike, 1940, p. 388). However because of his posi- being given equal weight" (Thorndike, 1940, p. 351), tivistic ideology, and possibly because of his own success were problematic because they were not scientific. Sci- in the business world, Thorndike accepted such power ence was Thorndike's religion. He did not approve of differentials as inevitable. political or moral systems that did not maximize effi- Nor was Thorndike able to resist the tendency to ciency and rationality. "The God of science is revealed play God himself. His faith in the power of science and in reality, and science rebels against counting the votes valuing of efficiency led him to espouse some highly

1150 October 1998 ° American Psychologist undemocratic ideas. Like many other progressives, and rationalization of educational and in- Thorndike supported and now proposed giving structional methods made educational psychology com- political power to the genetically more able (see Curti, patible with the "grammar of schooling" (Tyack & Cu- 1980; Hailer, 1984; Kevles, 1986). Thorndike had lec- ban, 1995) and useful to schools and teachers. Thorn- tured on eugenics as early as 1913, but had stopped short dike's awareness and exploitation of the commercial of wholeheartedly recommending it (Thorndike, 1913). possibilities of educational psychology established it in After more than 900 pages, the conclusion of Human the economic sector and provided a profit motive which Nature and the Social Order was an alarmingly simplistic propelled further expansion of the field. Encouraged by list of 20 suggestions. Along with "increasing capital this success and supported by a wealthy private founda- goods" and "the elimination of wars between and within tion, Thorndike then applied his empirical methodology nations," Thorndike recommended "better genes," to a wide range of other social issues and attempted to "guidance by science," and a "national council of the measure moral worth, just as he had objectified, quanti- able, good, and impartial endowed so as to be utterly fied, and commodified human learning. their own masters" (Thorndike, 1940, pp. 957-961). It is not surprising that Thorndike's proposal for a Thorndike said he supported absolute meritocracy be- science of values was not as successful as his educational cause quality "was better than equality" (Thorndike, psychology. It was subjective, unwieldy, and went against 1940, p. 962). Despite the unreasonableness of some of the grain of American religious and political traditions. these conclusions, his final recommendation was to have But Thorndike should not be blamed if the thorniness of "Reasonable expectations" (Thorndike, 1940, p. 963). morality and politics proved unsuited to scientific analy- sis, or if traditional moral reasoning and democratic eth- Conclusion ics did not diminish his satisfaction with what others When the state of the job market in psychology led Thorn- might see as the troubling nature of many of his conclu- dike to shift his attention from animals to education, he sions about moral and social values (Clifford, personal maintained the nineteenth-century linkage between intel- communication, January 14, 1998, p. 3). Most progres- lect and character. In The Human Nature Club, Thorndike sives believed unquestioningly in the ability of science questioned traditional religious views but allowed for the to solve societal problems, and many supported eugenics. possibility of Jamesian spirituality. In Notes on Child Few, if any, saw, or could be expected to foresee, the Study he rejected Hall's developmentalism as uninterest- problematic consequences of some of their research and ing, inexact, and immoral and advocated differential recommendations (see McCormick, 1990). If Thorndike rather than genetic psychology. In the first version of was blinded by the precision of his statistics, so were Educational Psychology Thorndike maintained that the many other psychologists and social scientists. moral aims of education should be determined by ideals, Character and intellect were pervasive topics in not facts, and that morality was more influenced by envi- Thorndike's work and continued to be themes in behav- ronment than intellect was. In The Principles of Teaching ioristic psychology, as the ideas of John Watson and he described morality as a product of learning acquired books such as B. E Skinner's Walden Two (1948) and through numerous specific connections. In the second Beyond Freedom and Dignity (1971), and Richard version of Educational Psychology he stated a genetic Herrnstein and Charles Murray's The Bell Curve (1994) view of the origins of ideals. At the end of his career, in attest. Above all, however, Thorndike was a methodolo- Human Nature and the Social Order, he proposed a moral gist. He was an educational, moral, and social technician. measurement system and advocated selective breeding to Thorndike believed in lots of little measurements, not big enhance individual and social goodness. Incrementally, theories. He excelled at doing exact, exacting analysis of like the connections in his theory of learning, Thorndike' s data and reveled in work and findings other psychologists explanations of morality became more efficiency-ori- might have thought drudgery or minutia. Eagerness for ented, biologically-deterministic, and positivistic. "facts no matter how uninspiring," Thorndike wrote as Most of Thorndike's early positions on education the last sentence of Human Nature and the Social Order, and morality were relatively conservative. Unlike some would contribute more to "beneficent reforms" than progressives, he did not challenge traditional assumptions "governments, churches, and social reformers in search about the content of curricula or the moral and political of wholesale salvation" (Thorndike, 1940, p. 963). purposes of education. Thorndike transposed the culture of the laboratory onto the school. He created a seemingly REFERENCES empirical, marketable knowledge base for education that met the needs of psychologists and educators and re- Beatty, B. (1996). Rethinking the historical role of psychologyi n educa- sponded to societal demands for order and efficiency. tional reform. In D. Olson & N. Torrance (Eds.), The handbook of Thorndike's systematization and transformation of older, education and human development (pp. 100-116). Oxford,E ngland: quasi-scientific theories of education to fit the emerging Blackwell. intellectual and methodological frameworks of experi- Brown, J. (1992). The definition of a profession: The authority of metaphor in the history of intelligence testing, 1890-1930. mentalism made educational psychology understandable Princeton, NJ: Princeton University Press. and acceptable to academics (White, personal communi- Cahan, E. D. (1992). Deweya nd development.D evelopmentalP sychol- cation, November 11, 1997). Thorndike's regularization ogy, 28, 205-214.

October 1998 • American Psychologist 1151 Cahan, E. D., & White, S. H. (1992). Proposals for a second psychology. Ross, D. (1991). The origins of American social science. New York: American Psychologist, 47, 224-235. Cambridge University Press. Callahan, R. E. (1963). Education and the cult of efficiency. Chicago: Ryan, A. (1995). John Dewey and the high tide of American liberalism. University of Chicago Press. New York: Norton. Chapman, E D. (1988). Schools as sorters: Lewis M. Terman, applied Skinner, B. E (1948). Walden two. New York: Macmillan. psychology, and the intelligence testing movement. New York: New Skinner, B. E (1971). Beyond freedom and dignity. New York: Knopf. York University Press. Sulloway, E J. (1998). Darwinian virtues. New York Review of Books, Cuban, L. (1993). How teachers taught: Constancy and change in XLV, 6, 34-40. American classrooms, 1880-1990. New York: Teachers College Thorndike, E. L. (1901a). The human nature club: An introduction to Press. the study of mental life. New York: Longmans, Green. Curti, M. (1980). Human nature in American thought. Madison: Univer- Thorndike, E.L. (1901b, June). Notes on child study [Monograph]. sity of Wisconsin Press. Columbia University Contributions to Philosophy, Psychology, and Danziger, K. (1990). Constructing the subject: The historical origins of Education, 8, (Nos. 3-4). New York: Macmillan. psychological research. Cambridge, England: Cambridge University Thorndike, E.L. (1901c). The study of children. Teachers College Press. Record, 2, 1 - 11. Dewsbury, D.A. (1992). Triumph and tribulation in the history of Thorndike, E. L. (1903). Educationalpsychology. New York: Lemcke & American comparative psychology. Journal of Comparative Psychol- Buechner. Thorndike, E. L. (1905). The elements of psychology. New York: A. G. ogy, 106, 3-19. Seiler. Fancher, R.E. (1996). Pioneers of psychology (3rd ed.). New York: Thorndike, E. L. (1906). The principles of teaching, based on psychol- Norton. ogy. New York: A. G. Seiler. Haller, M. H. (1984). Eugenics: Hereditarian attitudes in American life. Thorndike, E. L. (1910). The contribution of psychology to education. New Brunswick, NJ: Rutgers University Press. The Journal of Educational Psychology, 1, 5-12. Hernstein, R. J., & Murray, C. (1994). The bell curve: Intelligence and Yhorndike. E. L. (1911). Animal intelligence. New York: Macmillan. class structure in American life. New York: The Free Press. Thorndike, E. L. (1912). Education, afirst book. New York: Macmillan. James, W. J. (1899). Talks to teachers. New York: H. Holt. Thorndike, E. L. (1913, August). Eugenics. Monthly Jon9ich, G. M. (1962). Science: Touchstone for a new age in education. 70, 125-138. In G. M. Jongich (Ed.), Psychology and the science of education. Thorndike. E.L. (1913-1914). Educational psychology. New York: Selected writings of Edward L. Thorndike (pp. 1-26). New York: Teachers College Press. Bureau of Publications, Teachers College, Columbia University. Thorndike. E. L. (1917). The Thorndike arithmetics, Books 1-3. Chi- Jon~ich, G. (1968). The sane positivist: A biography of Edward L. cago: Rand-McNally. Thorndike. Middletown, CT: Wesleyan University Press. Thorndike. E.L. (1922). The psychology of arithmetic. New York: Kevles, D. J. (1986). In the name of eugenics. New York: Knopf. Macmillan. Kliebard, H.M. (1986). The struggle for the American curriculum, Thorndike. E. L. (1939). Your city. New York: Harcourt, Brace. 1893-1958. Boston: Routledge & Kegan Paul. Thorndike, E. L. (1940). Human nature and the social order. New York: Kloppenburg, J. T. (1986). Uncertain victory: Social democracy and Macmillan. progressivism in European and American thought, 1870-1920. New Thorndike, E. L., & Woodworth, R. S. (1901, May, July, November). York: Oxford University Press. The influence of improvement in one mental function upon the effi- Lagemann, E.C. (1989). The plural worlds of educational research. ciency of the other. Psychological Review, 8, 247-261, 381-395, History of Education Quarterly, 29, 185-214. 556-564. Lagemann, E. C. (1998). Contested terrain: A history of education re- Tyack, D. B., & Cuban, L. (1995). Tinkering toward utopia: A century search in the United States, 1890-1990. Educational Researcher, 26, of American public school reform. Cambridge, MA: Harvard Univer- sity Press. 9, 5-17. Tyack, D. B., & Hansot, E. (1982). Managers ofvirtae: Public school McCormick, R. L. (1990). Public life in industrial America. In E. Foner teachership in America, 1820-1980. New York: Basic Books. (Ed.), The new American history (pp. 93-118). Philadelphia: Temple Tyack, D.B., & Hansot, E. (1990). Learning together: A history of University Press. coeducation in American schools. New Haven, CT: Yale University O'Donnell, J. M. (1985). The origins of behaviorism: American psy- Press. chology, 1870-1920. New York: New York University Press. White, S. H. (1990). Child study at Clark University: 1894-1904. Jour- Reuben, J. (1996). The making of the modern university: Intellectual nal of the History of the Behavioral Sciences, 24, 131-150. transformation and the marginalization of morality. Chicago: Univer- White, S. H. (1991). Three visions of a psychology of education. In sity of Chicago Press. L. T. Landesman (Ed.), Culture, schooling, and psychological devel- Richards, R. J. (1987). Darwin and the emergence of evolutionary theo- opment (pp. 1-38). Norwood, NJ: Ablex. ries of mind and behavior. Chicago: University of Chicago Press. Wiebe, R.H. (1967). The search for order, 1877-1920. New York: Roberts, J. R. (1968, Winter). The quest for a science of education in Hill & Wang. the nineteenth century. History of Education Quarterly, 8, 431-446. Zenderland, L. (1990). Education, evangelism, and the origins of clinical Ross, D. (1972). G. Stanley Hall: The psychologist as prophet. Chicago: psychology: The child study legacy. Journal of the History of the University of Chicago Press. Behavioral Sciences, 24, 152-165.

1152 October 1998 • A m e r i c a n Psychologist