E. L. Thorndike and the Origins of Animal Psychology

(1)

See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/232606340

E. L. Thorndike and the Origins of Animal Psychology: On the Nature of the Animal in Psychology

Article in American Psychologist · October 1998

DOI: 10.1037/0003-066X.53.10.1135

CITATIONS

13

READS

1,380 2 authors, including:

Some of the authors of this publication are also working on these related projects:

Stanley Milgram and obedience to authorityView project Henderikus J. Stam

The University of Calgary 196PUBLICATIONS 3,849CITATIONS

SEE PROFILE

All content following this page was uploaded by Henderikus J. Stam on 16 May 2014.

(2)

E. L. Thorndike and the Origins of Animal Psychology

On the Nature of the Animal in Psychology

Henderikus J. Stam and Tanya Kalmanovitch University of Calgary

E. L. Thorndike's deliberate use of the experimental method and rejection of anthropomorphism and anecdot- alism place him squarely at the beginning of a century of animal research in psychology. The authors examine Thorndike's critique of the extant research of his day and the development of his methods, including the important way in which animals and children were often taken as interchangeable sources of data for the new functional psychology. Thorndike's research clearly altered the rela- tionship between psychologist and animal, as is demon- strated by his own studies and those that were published in the decade that followed. The authors review this body of work in the early 20th century for (a) the manner in which it exchanges anthropomorphism for mechanico- theriomorphism and (b) its constitution of the animal as a conflation between organism and technology. The authors conclude by noting the importance of debates concerning anthropomorphism in psychology for the cen- tury-long question of consciousness.

J

ust as 1879 is celebrated as the founding o f experimental p s y c h o l o g y with the establishment o f W u n d t ' s laboratory in Leipzig, Germany, so is 1898 celebrated as the founding o f experimental animal psy- c h o l o g y with the publication o f Thorndike's " A n i m a l Intelligence" in the Psychological Review Monograph Supplement. Such dates are frequently combinations o f practical changes in the orientation o f a field as well as the source o f various origin myths created for particular purposes by subsequent generations.

Thorndike's doctoral thesis at C o l u m b i a University readily fulfills both functions. As practical reorientation it signals the beginning o f a century o f animal research in North American p s y c h o l o g y that would play an important role in the development o f the discipline' s first theoretical frameworks. In the a c c o m p a n y i n g article by Galef (1998, this issue) and in histories by Boakes (1984), D e w s b u r y (1984), and others (e.g., Bruce, 1997; Jonqich, 1968), the general c h r o n o l o g y o f Thorndike's animal research and his writings from 1896 to 1911 are readily available. I In this article we discuss the origin m y t h that accompanied this founding m o m e n t in animal psychology. In effect, Thorndike inaugurated a new set o f relationships between

the psychologist and the animal in the creation o f the field o f experimental animal psychology. His work was not just the application o f an experimental method or simply a precursor o f behaviorism. Rather, it led to the production o f a convention that treats animals as abstract devices for introducing concepts that were to b e c o m e c o m m o n in human psychology. That is, animals were to b e c o m e organisms o f convenience on which psychologists could script a variety o f processes that were made

" v i s i b l e " in ways that were not possible with human beings. The origin m y t h is that Thorndike's methods and explanations overcame anecdotalism, anthropomorphism, and introspectionism. We argue that, in the creation o f a new technology and in the concern with measurement and experimentation, his methods and explanations replaced anthropomorphism with mechanomorphism and theriomorphism (hereafter mechanicotheriomorphism). Mechanomorphism is the exclusive attribution o f mechanistic properties to psychological phenomena, whereas theriomorphism is the attribution o f the qualities o f n o n h u m a n animals to human beings (English & English, 1958). By mechanicotheriomorph- ism we mean the ascription o f mechanical properties to phenomena that are psychological in nature in n o n h u m a n animals that are, in turn, used to explain human psychological phenomena (e.g., instinct and habit).

Henderikus J. Stam and Tanya Kalmanovitch, Department of Psychol- ogy, University of Calgary, Calgary, Alberta, Canada.

Portions of this article were presented in our paper, Isolating the Mind: Animal Research in Early American Psychology, at the 104th Annual Convention of the American Psychological Association, To- ronto, Ontario, Canada, August 9-13, 1996.

We thank Barbara Beatty, Bennett G. Galef, Jr., and Cheryl Logan for their comments on earlier versions of this article.

Correspondence concerning this article should be addressed to Henderikus J. Stam, Department of Psychology, University of Calgary, Calgary, Alberta, Canada T2N 1N4. Electronic mail may be sent to stam @ acs.ucalgary.ca.

~ Washburn (1926), and many subsequent psychologists, placed the birth of experimental animal psychology in Thorndike's laboratory at Columbia University in 1898. Nevertheless, a comparable and contemporary experimental program was underway at this time under the direction of Small and Kline at the Clark University laboratory (see Dewsbury, 1984, p. 49). Such dating also requires that work that was all but psychological in name and conducted by physiologists such as Shepherd Franz and Ivan Pavlov be ignored.

October 1998 • American Psychologist

1135

(3)

Our use of the term animal psychology is meant to designate the entire set of studies that have typically used nonhuman animals as a source of data for psychological research. We deliberately do not use the term compara- tive psychology because a great deal of animal research carried on in the discipline was not truly comparative, even though it may have gone under that label. At the same time, serious attempts to compare species of animals were discussed, if not actually carried out in practice, and Linus Kline at Clark University wrote two papers in 1899 outlining the content of a laboratory course in comparative psychology (Kline, 1899a, 1899b). By retaining the older term, animal psychology, we mean to include all psychological research carried on primarily with nonhuman mammals (see Dewsbury, 1992).

In order to understand the manner in which Thorn- dike altered the nature of the relationship between the psychologist and the animal in North American psychology we discuss two aspects of this history. First, we focus on the predominant understanding of animal psychology prior to the publication of Thorndike's thesis. This requires a discussion of both the important role played by animals in the largely British, natural histories of the time as well as the criticisms of that role by the New Psychology of the late 19th century. Second, we examine the relationship between research on animals and children in early North American psychology and third, trace developments in animal psychology in North America that followed Thorndike's work in the first decade of this century.

Thorndike and the Nature of Animal Psychology

Animal psychology, or more popularly, animal intellect, had been a source of considerable speculation since the

writings of both Charles Darwin and Herbert Spencer became well-known in the 1860s and beyond (e.g., Boakes, 1984; Galef, 1988; Richards, 1987). As Thorn- dike himself did, historians frequently make an important distinction between the anecdotal reports that character- ized the naturalist traditions of midcentury, particularly British, writings and the gradual development of an experimental science of animal intelligence. The progres- sion from Darwin to Romanes and from Romanes to Morgan "represented a continuous but evolving, tradition in the study of animals" (Boakes, 1984, p. 51).

George Romanes shared Darwin's interest in the evolution of mind and attempted to clarify the nature of animal intelligence by classifying the many reports available.

His Animal Intelligence (1882) and the succeeding Men- tal Evolution in Animals (1883) are indeed anecdotal and anthropomorphic in addition to being thoroughly Lamar- ckian with respect to the evolution of instinct and intelligence. The latter could be acquired, argued Romanes,

" b y the hereditary transmission of novel experiences"

(Romanes, 1883, p. 197).

Lloyd Morgan followed these developments with a critique of a number of key elements in the work of Romanes. In particular, Morgan made several crucial observations that proved to be useful to Thorndike in his experimental work. First, Morgan (1894) criticized the anecdotal tradition and distinguished between two types of psychology. One was inductive and introspec- tive and the other was observational and objective. Al- though both were required for a proper understanding of psychology, it was also the case that M o r g a n ' s mo- nism led him to e m p h a s i z e that these were dual aspects, subjective and objective, of the same mind. In addition, Morgan modified the L a m a r c k i a n views of Romanes, and Boakes (1984) went so far as to argue that Morgan returned to the principle of natural selection of the

" e a r l y D a r w i n " (p. 50). As G a l e f (1998) notes, Thorn- dike was strongly influenced by Spencerian doctrine.

Had he taken the early Darwin and Morgan more seri- ously he might have avoided the lapse out of what would later b e c o m e the mainstream of evolutionary biology.

Romanes and Morgan were influential, however, in suggesting to Thorndike a number of key elements of his method. 2 The puzzle boxes that Thorndike constructed were devices suggested to him, apparently, by some o f the descriptions o f animal abilities in the works of both of these earlier authors. Morgan had discussed two possible forms of trial-and-error learning as a basis for his argument that the acquisition o f skills could be attributed to the consequences of an action and not to c o m p l e x forms of association that 2 Morgan visited Harvard while Thorndike was in his first year of graduate school in 1896 but Thorndike makes no mention of attending Morgan's lectures. Thorndike would have been aware of Morgan's work, however, through William James, and he quotes Morgan exten- sively in his dissertation. In 1904, Morgan delivered the Lowell Lectures in Boston. and again there is no record of a meeting between the two men (see Boakes, 1984; Jonqich, 1968; Richards, 1987).

1136 October 1998 • American Psychologist

(4)

resided within the animal. Crude boxes with latches and doors were constructed, and Thorndike tested his animals in a state of " u t t e r hunger," a phrase he came to regret following repeated criticism (Thorndike, 1911 a). 3 These boxes were refinements of the method used in Thorndike's earlier work with chicks at Har- vard University as well as the systematic application of a technology to the observations contained in the works of the British zoologists.

Thorndike railed against the anthropomorphism o f the animal psychology o f his day in 1898, acknowledg- ing in 191 I that " w h a t one writes at the age of twenty- three is likely to irritate oneself a dozen years later"

and attributing it to the " e x u b e r a n c e , e v e n . . , arro- g a n c e " (p. vi) of his youth. He argued that previous work on animal psychology did " n o t give us a psychology, but rather a

eulogy,

of animals. They have all been about animal

intelligence,

never about animal

stupid- ity"

(1898a, p. 3). In addition to making animals ap- pear more rational than they might be, Thorndike chal- lenged the existing literature's anecdotal method. " T h e unconscious distortion o f the facts is almost harmless compared to the unconscious neglect o f an animal's mental life until it verges on the unusual and marvel- o u s " (1898a, p. 4). Anecdotes, argued Thorndike (1898a), provide only an " a b n o r m a l or supernormal psychology of a n i m a l s " (p. 5). Methodologically, Thorndike was highly critical of the single case and argued for repeated observations under regulated conditions.

Thorndike's criticisms have typically been inter- preted as a move away from the anthropomorphism o f the 19th-century naturalist. This is certainly true inso- far as Thorndike eschewed both the method and analysis of the naturalist. His work was written in the context of the Progressive Era. and he, like his colleagues

in the new profession o f psychology, was concerned with the scientific explanation of life and later would emulate the progressives' concern with popular school- ing and the professionalization of education in the early 20th century (Bledstein, 1976; Jonqich, 1968). This required a reconceptualization o f the dominant psychology o f his day, including the need to incorporate evolutionary theory, already widely popularized in its Lamarekian version in the United States by Herbert Spencer, in any understanding of psychology. It also included the application o f experimentation, which was so usefully developed in late 19th-century physiology, to the problems of animal psychology. More important, however, the new animal psychology served as much as the old animal psychology to provide psychologists with their first preoccupation, namely, represent- ing the human mind.

Thorndike was clear in articulating his intellectual heritage in t898 as well as in subsequent publications (Thorndike, 191 ta). However, the differences between him and his precursors have been emphasized, if not exaggerated, both by Thorndike himself and by most who have written about his work. This exaggeration of differences has two important implications: It suppresses the common understanding of evolution that was prevalent in the late 19th century, and it makes invisible the mechanicotheriomorphism inherent in Thorndike's work.

First, the transition in British and North American aca- demic communities to a professional zoologist and biologist designation that came in tandem with the acceptance of evolutionary theory, and the gradual downgrading of the amateur naturalist, did not necessarily mean a change in the underlying metaphorical understanding of the relationship between animals and humans. As Ritvo (1987) has argued, Darwinian and Spencerian theory still ranked animals in relation to humans, and the continuity between animals and people that had emerged in evolutionary theory "'made it even easier to represent human competi- tion, and the social hierarchies created by those who prevailed, in terms of animals" (p. 40). Both the popular and scientific zoological literature promulgated a hierarchical vision of the animal kingdom and "provided its reade~ with a shared set of assumptions, values, and associations that simultaneously confirmed human ascen- dancy and supported the established social order" (Ritvo, 1987, p. 42). 4 In addition, world fairs, ethnographic ex- hibits, new museums of history and natural science, and a host of panoramas and other exhibitions put exotic animals on show along with natives from a wide range of colonized cultures. These were then measured, classified, pictured, and so on in the context of a discourse that

Thorndike akmg with many other animal psychologists (e.g., Pav- lov, Watson) had to contend with a sustained antivivisectionist move meat at the turn of tile century, a factor that aright account for a certain defensiveness on this issue in Thorndike's work (see Dewsbury, 1990, for a full account).

4 This hierarchical conception of the animal world has persisted in 20th century discussions of comparative psychology (see, e.g., Hodos &

Campbell, 1969).

October 1998 • American Psychologist 1137

(5)

relied on scientific, particularly evolutionary, theories to justify their educational value (Corbey, 1995). As back- ward ancestors, their exhibition was used to justify the colonialism of late 19th-century Europe and the inherent superiority of the White race. A gradual hierarchy of evolution became common knowledge. By sharply distin- guishing himself from his predecessors, the evolutionary view that Thorndike shared with his peers was eclipsed.

Second, the subtext that often informs discussions of the literature that precedes Thorndike was that this earlier work was largely unscientific due to its reliance on anecdote and mere description. The charge of anecdotalism is subtly related to the charge of anthropomorphism, namely that the anecdote is favored because it supports an anthropomorphic view of the animal as clever, rational, and possessed of human-like qualities.

Implied is that rigorous experimentation that does not rely on anecdote, but rather on observation and manipulation, is not anthropomorphic. Morgan's claim that attri- butions of consciousness in others (human or animal) are merely an image of one's own consciousness, however, was directly derived from his reading of idealist philoso- phy (Richards, 1987). But Morgan considered human and animal intelligences vastly different. The distinctions between animals and humans would gradually be turned on their heads, first by mechanizing the conception of animal intelligence and second, by attributing such forms of intelligence to all animals, human and nonhuman alike.

It was characteristic of Progressive Era intellectuals, however, to view mind as the outcome o f evolutionary processes whereby that which could be known about an animal could be understood as a graduated step away from " p r i m i t i v e " forms of mind, such as those found in children and in those who made up the lower strata of society (e.g., Noble, 1958). Thorndike's later turn to the problems of education and his interests in eugenics were therefore not unrelated to his animal work.

Thorndike, Children, and Other Animals

Like a number of his peers, Thorndike conducted research on both animals and children. In fact, Thorndike made his entry into animal research, by his own account, because of the failure to obtain approval from Harvard, in his first year of graduate studies, for his mind-reading experiments in young children. In these studies, young children from three to six years of age would attempt to guess which letter, object, or number was the one that Thorndike held in mind at that moment. The purpose of the research was to determine the "responsiveness of young c h i l d r e n . . , to facial expressions or movements made unconsciously" (Thorndike, 1936, p. 264). Ac- cording to his 1936 autobiographical statement, he said of this research in the 1896-1897 study year,

The children enjoyed the experiments, but the authorities in control of the institution would not permit me to continue them.

I then suggested experiments with the instinctive and intelligent behavior of chickens as a topic, and this was accepted. I kept these animals and conducted the experiments in my room until

the landlady's protests were imperative. (Thorndike, 1936, p.

264)

William James attempted to intervene and when he could not obtain s p a c e for Thomdike's animals elsewhere, he allowed Thorndike to keep the chickens in his basement for the duration of the year. As Thorndike recalled, " T h e nuisance to Mrs. James was, I hope, somewhat mitigated by the entertainment to the two youngest children" (p.

264).

Jonqich (1968) believed that Thorndike would not have so readily substituted chickens for children as he let on in 1936 and that this decision was not just a "fortu- itous second choice" (p. 88). Thorndike was already in- volved in other forms of animal research, an apparent rarity amongst his peers. Gertrude Stein, who was with Thorndike in the same seminar taught by James in the spring of 1896, noted the oddness of the chicken research (Jonqich, 1968). It appears that Thorndike had an aca- demic interest in animals and children from the onset of his graduate training.

Thorndike would devote the bulk of his professional career at Columbia's Teachers College not to animal research but to educational psychology. Indeed Thorndike would be at the center of the movement to professionalize education during the Progressive Era at the institution that became the leading voice of that movement. His first post at Western Reserve University in 1898 was to teach pedagogy, and Thorndike published an article in the No- vember 1898 issue of Psychological Review on the con- troversy initiated by Mtinsterberg that was then raging over the value of child-study in psychology. In this article Thomdike (1898b) compared the amateur with an interest in child-study to the naturalist in biology. And in the dissertation itself Thomdike argued,

Comparative psychology has, in the light of this research, two tasks of prime importance. One is to study the passage of the child mind from a life of immediately practical associations to the life of free ideas; the other is to find out how far the anthropoid primates advance to a similar passage, and to ascer- tain accurately what faint beginnings or preparations for such an advance the early mammalian stock may be supposed to have had. (1898a, p. 151)

Although Thorndike may have downplayed the importance of his animal research in 1936, it was not unusual to his career alone that he should move freely from research between children and animals. For example, Swiss psychologist Edouard Clapar~de said that in 1900 he resolved his conflicts between clinical neurology, laboratory research work, and animal psychology when he read Die Spiele der Tiere by Karl Groos, which "revealed to me, as in one illuminating flash, what should be the foundation of the art of education, i.e., the right use made of the child's natural tendencies" (Clapar~de, 1930, p.

75). Two decades later John Watson would switch to the study of infants from animals in a seamless move, and Jean Piaget switched from research on mollusks to research on children at about the same time, seeking in this move to base morality on science (Vidal, 1987).

(6)

Animal psychology in early 20th-century psychology, based on hierarchies of intelligence and adaptivity, was readily moved to the problems of the hierarchical environment of the classroom. For as animals were re- strained and tested so was mass education a setting in which to regulate and adapt the child to the new industrial age. And just as labor could be conceived of in terms of its constituents and timed accordingly (as Frederick Tay- lor so aptly demonstrated) so could the classroom be regulated and work divided according to subjects and tasks. This analogy is supported by the development of the concept of intelligence. Spencer had already argued for the special role of intelligence from the very first edition, published in 1855, of the Principles of Psychol- ogy. His associationism led him to argue that

since all modes of consciousness can be nothing else than incidents of the correspondence between the organism and its environment; they must all be different sides of, or different phases of, the co-ordinated groups of changes whereby internal relations are adjusted to external relations. (Spencer, 1897, pp.

495 -496)

Here nature was one, and " n o impassable chasm" ex- isted between physical and psychical life as far as Spencer was concerned. Danziger (1997) has argued that Spen- cer's terminology eventually won the day and, as noted above, intelligence was a word readily adopted by both Romanes and Morgan. By the time Thorndike moved to Teacher' s College in 1899, it would only be a matter of a few years before Pearson, Spearman, and Binet published their articles on intelligence, bringing the terminology fully into educational circles and practice. However, its biological roots ensured that it would eventually replace the older and more established, strictly human concept of "intellect." Instead, intelligence became the preferred term during the first decade of this century because it referred to individual adaptability and a graded set of abilities. It also sped the process whereby learning gradually became one of the dominant topics in psychology (Danziger, 1997). Again, Thorndike stands at the head of this tradition, being the first to use "learning" as a psychological category in 1901 (Thorndike, 1901/1911b, p. 283; earlier even than Danziger's dating of 1902).

Mechanicotheriomorphism had arrived.

The need to develop objects of scientific interest in the animal laboratory also had the unfortunate effect of diminishing the importance of evolutionary theory, the one position that might have prevented the theriomorphic stance. By enlisting animals as analogs for human psychology, Thorndike, like many other animal psychologists in the period from 1890 to 1920, simply ignored the dictates of evolution (Mills, 1997, in press). Thorndike (191 l a) noted only that connections between sensations and reactions formed the basis of adaptive habits if these connections were accompanied by beneficial consequences. As Mills (1997, in press) has argued, Thorndike displayed a remarkable lack of interest in questions of inheritance and instinct, relegating both of these to a minor role in his thesis. The emerging conception of

individual adaptation was deeply at odds with the Dar- winian notion of species adaptation (Sohn, 1976). For animal psychologists of the period, "Darwin's name would be intoned but his voice, if heard at all, would be muffled and distorted" (Mills, 1997, p. 15). Indeed, for Thorndike, Spencer played a more critical role than Darwin.

Although Thorndike's interests shifted after his ap- pointment to Teacher's College, research in animal psychology began to emerge as a distinct subfield and spe- cialty. Next we examine the first decade of this emer- g e n c e - l a t e r years have already been covered in detail elsewhere (cf. Beach, 1950; Boakes, 1984; Dewsbury, 1997). Our purpose is to reconstruct the mechanicotheriomorphism of the "animal subject" that emerged out of Thorndike' s work.

Animal Research After Thorndike

In the years immediately following the publication of Thomdike's dissertation, animal psychologists attempted to establish their field as a scientific project, their goal being the production of positive knowledge about the animal mind. The cornerstone of animal psychology as natural science was the use of experimental methodologies, where the anecdote gave way to the experiment, and the naturalist to the experimentalist. The original ontological question of the animal mind, as considered by Romanes and Morgan, was transformed into a methodological one, a product of the budding functionalism and pragmatism of early 20th-century psychology. The question of interest rapidly became not "was there an animal mind," nor "what are the kinds of qualities of the animal's mental experience," but rather " h o w might psychologists collect facts about the psychological features and processes of the animal mind such as intelligence and instinct." The uniquely American answer, brought to wide attention by Thorndike in 1898, was the laboratory experiment, in which the animal could be systematically isolated, controlled, and observed. From 1898 onward, the number of experimental reports citing the use of animals as subjects rose steadily as the experiment rapidly eclipsed any rival method of animal study in psychology (see Figure 1). The figure shows our calculation of the number of experimental papers in animal psychology that appeared in the five journals that published such research in the United States. The trend was influenced by a number of factors. In particular, the establishment of a dedi- cated journal for animal psychology (the Journal of Ani- mal Behavior, 1911, an offshoot of the Journal of Com- parative Neurology and Psychology) was associated with a marked increase in the number of published experimental reports using animals as subjects. Moreover, in 1909 the Wistar Institute in Philadelphia began its breeding program with the albino rat creating a colony that would expand to 11,000 rats by 1913 to be widely distributed to other laboratories (Clause, 1993).

For animal psychology to establish itself within the existing institutional and intellectual structures of Ameri- can psychology, its advocates needed to demonstrate that

(7)

20

15-

10-

5 -

0

1895 1900 1905 1910 1915

NUMBER OF PUBLISHED

R E P O R T S

Figure 1

Number of Experimental Reports on Animals in Five Psychology Journals Published in the United States

1898- 1914

YEAR

knowledge of the animal mind would be both scientific and useful. Animal laboratories were established because they provided an inexpensive means of science education even if the methods were time consuming. Until the ar- rival of psychological tests, animal psychology was a sure means of demonstrating the relative roles of instinct and habit in psychology (Mills, 1997). Utility would ulti- mately be based, however, on the light such research would shed on the mental processes of humans. Thus research in the early years, mainly in response to Thorn- dike's publication, dealt largely with the learning and sensory capacities of animals. The vibrancy of early animal research was due in part to the fact that once the problems of animal psychology had been reconceptual- ized as problems of learning, then it was a short step to the obvious claim that very little precise knowledge about learning and the perceptual capacities of various species had previously been established. Despite this reconceptualization, the change from anecdotalism to experimental laboratory methodologies proceeded slowly over the first decade.

The development of experimental methodologies was to have significant consequences for the animal. The animal as the subject of naturalistic observation was to undergo significant changes when it was cast as an object of psychological experimentation. The technology of experimental research transformed not only the knowledge that could be obtained from an animal but the insertion of the isolated animal into a piece of machinery became an entirely new way of " s e e i n g " the animal. Gradually, this form of knowledge production became a privileged and definitive source of knowledge about animals in psychology. Confined, food-deprived, restricted to limited

actions, and interbred, the animal in experiments was sufficiently unlike its natural counterpart to provide abstract knowledge of a hierarchical notion of intelligence that paid no heed to the concrete bodies of specific species of animal. Although comparative and descriptive work continued to be done, it gradually became less important to North American psychology (see Dewsbury, 1984; Kalikow & Mills, 1989). A brief description of the major transformations that allowed this transition to occur can be found by examining the research conducted by Thorndike (1899, 1901), Yerkes (1903), Allen (1904), and Watson (1907).

Thorndike's (1899, 1901) two experimental reports published after Animal Intelligence were both presented as extensions of the 1898 work, and the first of these (in 1899) consisted of data collected contemporaneously with the data from his doctoral thesis. The studies reported are observations on young chicks ranging in age from a few hours to several months under various conditions (e.g., jumping from a height, the presence of a cat). The methods and analyses, however, owe more to Morgan's anecdotalism than to the experimentalism that would become established by 1911. Indeed, Thorndike compared many of his observations of the reactions of young chicks with those reported by Morgan in An Intro- duction to Comparative Psychology (1894). Notable in the 1899 article is an absence of formal laboratory apparatus. Instead, Thorndike reported using living cages, col- ored cardboard squares, and a box off of which the chicks were induced to jump. Auditory stimuli were produced by the industrious experimenter himself by "clapping the hands, slamming a door, whistling sharply, banging a tin- pan on the floor, mewing like a cat, playing a violin, thumping a coal-scuttle with a shovel, etc., etc." (Thorn- dike, 1899, p. 290). In contrast to the 1899 paper, the 1901 monograph displays many of the elements of later experimental animal study. Thorndike studied three Cebus monkeys, referred to only by number, and conducted numerous experiments that required the monkeys to carry out activities such as opening complex boxes, obtaining food, or discriminating between two signals.

The apparatus consisted of various puzzle boxes similar to those from the 1898 monograph, and the central motive used to induce association was hunger. Thorndike switched early in the experiments from the egress-type configuration of his 1898 work (in which the hungry animal was required to escape confinement to reach the food) tO an ingress configuration when Monkey 1 proved to be " n o t easily handled at [the] time" (1901, p. 9).

Results were once again presented in the form of learning curves and tables, with additional anecdotal information provided only to further illustrate and support the infer- ence drawn from the quantitative data.

Along with vestiges of anecdotalism, the key elements of what would later become the standard of experimental method in animal psychology were present in these reports; animal study was carried out in a laboratory setting, with concern and attention given to standard- izing the animals' environment. The use of apparatus,

(8)

though not prominent in the 1899 paper, further served to isolate and control the animal and assisted in the manipulation of motives such as fear, hunger, and constraint, which provided the experimenter with a means to elicit the requisite behaviors.

In 1903, Robert Yerkes reported his development of apparatus to further control the animal subject, and in particular, he reported the first published use of electric shock as a stimulus in animal psychology. Yerkes devised a reaction box in which the subjects of these experiments, green frogs, were required to escape confinement. When confinement alone was not a sufficient motive to induce the green frog to learn the task, Yerkes attempted to stimulate the subject with a stick. This only seemed to frighten the animal (1903, p. 582). By means of a solu- tion, Yerkes introduced a piece of apparatus which he termed a "labyrinth." The apparatus consisted of an en- trance chamber and a two-choice path over a series of electrical circuits which led to a shallow tank of water.

Either path could be obstructed with a glass plate. Upon selecting the obstructed path, the animal was stimulated with an electric shock. As Yerkes explained, the electric shock "facilitated the experiment by forcing the animal to seek some other way of escape, and it also furnished material for an association" (1903, p. 584).

Yerkes explicitly acknowledged selecting subjects of exceptional suitability for this particular experimental task, a practice that would soon lead to the breeding of laboratory animals. Nevertheless, when Yerkes came across a problem in audition that he could not seem to address in the laboratory, he conducted a post facto naturalistic study. Thus research in a natural habitat re- mained an option in 1903. By 1913, however, Yerkes would have adopted what became a more standard view of experimentation. In his lecture notes of that year, he wrote " I d e a l - - t h e animal should [be] in a perfectly controlled and describable situation [to] take a 'picture' (ac- curate record) of its behavior" (Yerkes, 1913).

In 1904, Jessie Allen at the University of Chicago published what, to our knowledge, is the first experimental psychological study using guinea pigs. She acknowledged the influence of Thorndike:

My work on the guinea pig has been undertaken from a point of view somewhat similar to that assumed by Thorndike; viz., the point of view that the law of parsimony must govern inter- pretation, and a sufficient number of control experiments must condition every statement made. (Allen, 1904, p. 297) This paper was still transitional, however, largely describing guinea pigs' development and habits and comparing these with the white rat. Allen gave the guinea pigs various maze tasks to solve but in doing so still argued that

" n o problem should be given to an animal which involves the inhibition of a deep-seated instinct" (1904, pp. 3 0 9 - 310), a (soon to vanish) recognition that experimental arrangements could restrict or alter natural activities.

One final, major innovation that would come to de- fine experimental animal research in psychology was provided by Watson in 1907. He considered this monograph

to constitute the introduction of a new research paradigm in comparative psychology in which vivisection would be performed on animals that had been trained in an association task. Although he acknowledged in a footnote that a 1902 paper by Shepherd Ivory Franz had already presented a virtually identical method, Watson announced with typical grandiosity, " H e r e is, undoubtedly, a new field for the student of animal psychology; a field which must be worked over from the psychophysical standpoint, in contradistinction to the purely physiological o n e "

(1907, p. 1).

The method reported in this paper consisted of training white rats on a modified Hampton Court Maze (in a virtual replication of one reported by Small in 1901) and successively removing or lesioning sensory functions in order to address questions about sensory organ processes in animals. 5 Watson admitted to not going far beyond the interpretive scope of Small's earlier work, but he did claim to make certain minor methodological improve- ments. This consisted in large part of the use of tamer animals, which, accustomed to laboratory life, were thought to settle into their tasks straight away without needing time to be familiarized to the laboratory environment or to adjust to the rigors of maze-learning. Watson expected the animal as psychological subject to be a sophisticated denizen of the laboratory. Fittingly, the rats in this report were referred to by number or as " n o r m a l "

or "defective." In an appendix, Watson considered that the successive, but temporally separated, elimination of separate sensory functions might allow a source of error into his conclusions. To address this, he performed an operation in which the eyes, the olfactory bulbs, and the vibrissae of a young male rat were simultaneously removed. According to Watson's optimistic report, "Nat- urally recovery was slow with this a n i m a l . . . The animal finally completely recovered and is still alive . . . and in absolutely perfect condition. He shows the same eager curiosity, which is so characteristic of the normal animal" (p. 98). The destruction of sensory functions, whether successive or simultaneous, nevertheless carried the same consequence for the subjects, rendering the rats essentially and enduringly unsuitable for any other life than that of a laboratory subject.

Primary in the accomplishment of the transition of the animal from natural object to object of psychological science was the implementation of experimental methods that were intended to force the formation of associations through the manipulation of the animal's physical and psychological experience, thereby eliciting the required or desired behavior of the animal. Although most of the researchers that are cited above expressed a certain eager- ness to demonstrate that no harm came to their animal subjects (the irony of this stance perhaps most evident in the appendix to Watson's 1907 report), the successive elaboration of experimental apparatus and methodology

5 See Dewsbury (1990) for a description of the public outcry that followed the presentation of this study at the meetings of the American Association for the Advancement of Science in 1906.

(9)

that began with Thorndike in 1898 only carried increas- ingly severe consequences for the animal subject. The construction of more sophisticated laboratory appara- tuses may have had as its impetus the need to refine, standardize, and optimize inherently fallible observational techniques, but its larger, implicit, goal was the erosion of the experimental animal as a sentient being with ties to the natural world. The further incorporation of surgical techniques of control of the animal's sensory functions also had serious consequences for animals and, explicit in the mandate of experimental animal psychology, the conclusions drawn for human psychology. In 1907, Watson stated that " f r o m the standpoint of method, the study of the minds of animals (normal or defective) does not essentially differ from the study of human defective minds and the minds of children" (p. 2), and it is of no little significance that he later referred to surgical manipulation of the animal's body as no more than a

" v e r y serviceable" method of environmental control (Watson, 1914, p. 69).

Technologies, Mechanicotheriomorphism, and Animal Consciousness

In his analysis of the social organization of Pavlov's

"physiology factory," Todes (1997) has argued that the modifications surgically produced on Pavlov's dogs con- stituted these animals not only as objects of study but also as technologies that were created to produce particular kinds of knowledge. Hence the animal served as both object and intermediary to the facts. Donna Haraway (1989) made this case about Yerkes' s work with primates:

" T h e laboratory animal in general possessed the highest value for human beings precisely because it was designed and standardized, in short, engineered, to answer human queries. But the animal's epistemological status was also as natural object yielding objective understandings" (p.

62). There is a gradual narrowing from understanding the animal as the outcome of, and as understandable in, a unique environment to the animal as a subject of convenience, a process that extends over approximately 20 years (see Beach, 1950; Dewsbury, 1997). It is also con- sistent with methodological developments in human psychology at the turn of the century (Stare, 1997).

Animal research played an important role in the development of a graded theory of intelligence, and it continued to support the development of theories of behavior that could be applied to children and adults alike.

In order for animals to fit this role they had to be shorn of their unique characteristics as animals, that is, they needed to be the equivalent of a laboratory tabula rasa.

In 1913 when Watson presented his behaviorist mani- festo, that is exactly what the animal had become! By first making animals fit for laboratory consumption through restraints, isolation, and food deprivation and then taking the resulting animals as representatives of "nature," the refined knowledge that emerged from the animals could be treated as entirely objective. Properties of a mechanical nature are ascribed to the animal, and having been

thus inscribed they are fit objects for explaining the properties of humans and nonhumans alike.

Thorndike's work heralded a period of productivity and engagement with animals that would eventually alter the substance of North American psychology. It was a new orientation with new aims, methods, and explanatory goals. However, we challenge the view that the orientation was exactly what Thorndike and those who followed took it to be. By introducing physiological forms of experimentation and control into the psychological laboratory, Thorndike replaced anthropomorphism with mechanicotheriomorphism.

In the late 20th century, psychologists are still not decided about the place and function of animals in psychological research (e.g., Davis & Balfour, 1992; Pious, 1993; Wertz, 1994). Anthropomorphism was a term whose origin in the 18th century was reserved for the ascription of human forms and attributes to deities, and only in the 19th century did it become a general term for the attribution of human form or character to anything impersonal or irrational such as nations or animals. As Thorndike made so clear, anthropomorphism became a problem for the scientific community after the introduction of evolutionary theory. Adherents of the "new psychology" at the turn of the century attempted to resolve this issue by recourse to the so-called argument from analogy: If others (persons, species) behave like I do then I can reasonably infer similar experiences. Boring (1929) described this position in retrospect as, "the experimenter penetrates the subject's mind by way of analogy with his own. If one dislikes, as Titchener did, this em- phasis on behavior in discussing introspection, one can substitute the conception of 'empathy' for analogy" (p.

551). By 1929 Boring dismissed this position, saying that

"it is not possible to penetrate the animal's consciousness by way of analogy" (p. 551), and under the influence of Jacques Loeb, a committed mechanist, he was forced to ask whether an animal had any consciousness at all. It is interesting to note that this exact argument from analogy was repeated recently by the ethologist Marian Daw- kins (1993), when she argued for the existence and significance of animal consciousness.

For the later behaviorists, anthropomorphism was a continuing embarrassment fueled no doubt by the public avowals of its necessity in the writings of psychologists such as Tolman (1938/1978) and Waters (1939). Waters argued that animal behavior can be made intelligible only by describing it in anthropomorphic terms, and Tolman justifies his being "openly and consciously just as anthropomorphic" as he pleases and continues " I in my future work intend to go ahead imagining how, i f I were a rat, I would behave as a result of such and such a demand combined with such and such an appetite and such and such a degree of differentiation; and so o n " (1938/1978, p. 356). Yet, as anthropomorphic as he was, Tolman was equally theriomorphic. At the end of his 1937 APA Presi- dential Address, he proclaimed his belief that

everything important in psychology (except perhaps such matters as the building up of a super-ego, that is everything save

(10)

such matters as involve society and words) can be investigated in essence through the continued experimental and theoretical analysis of the determiners of rat behavior at a choice-point in a maze. (1938/1978, p. 364)

A n t h r o p o m o r p h i s m b e c a m e e v e r m o r e p r o b l e m a t i c w h e n the w o r k o f ethologists such as L o r e n z and T i n b e r g e n b e c a m e m o r e w i d e l y k n o w n in the 1950s. A n d along w i t h t h e r i o m o r p h i s m , it r e m a i n s an o p e n question, for as M a r i a n D a w k i n s asks, " i f b e i n g sufficiently like us d o e s n o t m e a n l o o k i n g like us or h a v i n g h u m a n c u s t o m s and l a n g u a g e , w h a t d o e s it m e a n ? I f c o n s c i o u s n e s s d o e s n o t h a v e to h a v e a h u m a n face, s p e a k w i t h a h u m a n v o i c e or l o o k out f r o m h u m a n eyes, inside w h a t sort o f b o d i e s should w e l o o k for it? A n d h o w will w e r e c o g n i z e w h e n w e find i t ? " ( D a w - kins, 1993, p. 16; see also B u r g h a r d t , 1991). C o n t e m p o r a r y e t h o l o g y grants the rat far m o r e abilities and s o c i a l i t y than w a s e v e r c o n s i d e r e d p o s s i b l e b y the early a n i m a l e x p e r i - mentalists (e.g., D a w k i n s , 1993). W h a t a n i m a l s did p r o v i d e w a s a r e a d y s o u r c e o f data points f o r the e x p e r i m e n t a l project. H i g h l y restricted in their ability to eat, sleep, r e p r o - duce, interact, and so on, they p r o v i d e d a b o d y o n w h i c h m e c h a n i s t i c m o d e l s c o u l d be i n s c r i b e d and f r o m w h i c h i n f e r e n c e s c o u l d be d r a w n about h u m a n beings. In this r e s p e c t T h o r n d i k e ' s 1898 dissertation is surely to b e c o u n t e d the first study a m o n g many.

REFERENCES

Allen, J. (1904). The associative processes of the guinea pig. A study of the psychical development of an animal well medullated at birth.

Journal of Comparative Neurology and Psychology, 14, 293-359.

Beach, E A. (1950). The snark was a boojum. American Psychologist, 5, 115-124.

Bledstein, B.J. (1976). The culture of professionalism: The middle class and the development of higher education in America. New York: Norton.

Boakes, R. (1984). From Darwin to behaviorism: Psychology and the minds Of animals. Cambridge, England: Cambridge University Press.

Boring, E. G. (1929). History of experimental psychology. New York:

Appleton.

Bruce, D. (1997). Puzzling over animal intelligence [Review of animal intelligence: Experimental studies]. Contemporary Psychology, 42, 879-882.

Burghardt, G. (1991). Cognitive ethology and critical anthropomorphism: A snake with two heads and hognosed snakes that play dead.

In C. A. Ristau (Ed.), Cognitive ethology: The minds of other animals (pp. 53-90). Hillsdale, NJ: Erlbaum.

Claparbde, E. (1930). Edouard Clapar~de. In C. Murchison (Ed.), A history of psychology in autobiography (Vol. 1, pp. 63-97). Worces- ter, MA: Clark University Press.

Clause, B. (1993). The Wistar rat as a right choice: Establishing mammalian standards and the ideal of a standardized mammal. Journal of the History of Biology, 26, 329-349.

Corbey, R. (1995). Ethnographic showcases, 1870-1930. In J. N. Piet- erse & B. Prekh (Eds.), The decolonization o f imagination: Culture, knowledge and power (pp. 57-80). London: Zed Books.

Danziger, K. (1997). Naming the mind: How psychology found its lan- guage. London: Sage.

Davis, H., & Balfour, D. (Eds.). (1992). The inevitable bond. Cam- bridge, England: Cambridge University Press.

Dawkins, M. S. (1993). Through our eyes only? The search for animal consciousness. Oxford, England: W. H. Freeman.

Dewsbury, D. A. (1984). Comparative psychology in the twentieth cen- tury. Stroudsburg, PA: Hutchinson Ross.

Dewsbury, D. A. (1990). Early interactions between animal psychologists and animal activists and the founding of the APA Committee

on Precautions in Animal Experimentation. American Psychologist, 45, 315-327.

Dewsbury, D. A. (1992). Comparative psychology and ethology: A reas- sessment. American Psychologist, 47, 208-215.

Dewsbury, D. A. (1997). Comparative psychology in journals, 1911- 1927: Another look at the snark. Unpublished manuscript.

English, H. B., & English, A. V. (1958). A comprehensive dictionary of psychological and psychoanalytical terms. London: Longmans, Green.

Galef, B. G., Jr. (1988). Evolution and learning before Thorndike: A forgotten epoch in the history of behavioral research. In R. C. Bol- les & M. D. Beecher (Eds.), Evolution and learning (pp. 39-58).

Hillsdale, NJ: Erlbaum.

Galef, B. G., Jr. (1998). Edward Thorndike: Revolutionary psychologist, ambiguous biologist. American Psychologist, 53, 1128-1134.

Haraway, D. (1989). Primate visions: Gender, race, and nature in the world of modern science. New York: Routledge.

Hodos, W., & Campbell, C. B. G. (1969). Scala naturae: Why there is no theory in comparative psychology. Psychological Review, 76, 337-350.

Jonqich, G. (1968). The sane positivist: A biography of Edward L.

Thorndike. Middletown, CT: Wesleyan University Press.

Kalikow, T J., & Mills, J.A. (1989). Wallace Craig (1876-1954), ethologist and animal psychologist. Journal of Comparative Psychol- ogy, 103, 281-288.

Kline, L. W. (1899a). Methods in animal psychology. American Journal of Psychology, 9, 256-279.

Kline, L. W. (1899b). Suggestions towards a laboratory course in comparative psychology. American Journal of Psychology, 10, 399-430.

Mills, J. A. (1997). Early American comparative psychology. Unpub- lished manuscript, University of Saskatchewan, Saskatoon, Saskatch- ewan, Canada.

Mills, J. A. (in press). Control: A history of behavioral psychology.

New York: New York University Press.

Morgan, C. L. (1894). An introduction to comparative psychology. In D. N. Robinson (Series Ed.), Significant contributions to the history of psychology: Series D, Vol. 2. Comparative psychology. Washing- ton, DC: University Publications of America.

Noble, D. W. (1958). Theparadox of progressive thought. Minneapolis:

University of Minnesota Press.

Pious, S. (Ed.). (1993). The role of animals in human society [Special issue]. Journal o f Social Issues, 49, 1-167.

Richards, R. J. (1987). Darwin and the emergence o f evolutionary theo- ries of mind and behavior. Chicago: University of Chicago Press.

Ritvo, H. (1987). The animal estate: The English and other creatures in the Victorian age. Cambridge, MA: Harvard University Press.

Romanes, G. J. (1882). Animal intelligence. New York: Appleton.

Romanes, G. J. (1883). Mental evolution in animals. London: Kegan, Paul, Trench.

Sohn, D. (1976). Two concepts of adaptation: Darwin's and psychology's.

Journal of the History of the Behavioral Sciences, 12, 367-375.

Spencer, H. (1897). The principles of psychology (Vol. 1, 3rd ed.). New York: Appleton.

Stam, H.J. (1997, May). Technologies "R" us: Psychology and the production of new bodies. Paper presented at the seventh biennial meeting of the International Society for Theoretical Psychology, Ber- lin, Germany.

Thorndike, E. L. (1898a). Animal intelligence: An experimental study of the associative processes in animals. Psychological Monographs, 2(4, Whole No. 8).

Thorndike, E. L. (1898b). What is a psychical fact? Psychological Re- view, 5, 645-650.

Thorndike, E. L. (1899). The instinctive reaction of young chicks. Psy- chological Review, 6, 282-291.

Thorndike, E. L. (1901). The mental life of monkeys. Psychological Monographs, 3(5, Whole No. 15).

Thorndike, E.L. (191 la). Animal intelligence: Experimental studies.

New York: MacMillan.

Thorndike, E. L. (191 lb). The evolution of the human intellect. In E.

Thorndike (Ed.), Animal intelligence: Experimental studies (pp. 282- 294). New York: MacMillan. (Original work published 1901)

O c t o b e r 1998 • A m e r i c a n P s y c h o l o g i s t 1143

(11)

Thorndike, E. L. (1936). Edward Lee Thorndike. In C. Murchison (Ed.), A history of psychology in autobiography (Vol. 3, pp. 263-270).

Worcester, MA: Clark University Press.

Todes, D. P. (1997). Pavlov's physiology factory, 1891-1904. Isis, 88, 1 - 4 2 .

Tolman, E. C. (1978). The determiners of behavior at a choice point.

In E. R. Hilgard (Ed.), American psychology in historicalperspective (pp. 337-370). Washington, DC: American Psychological Associa- tion. (Original work published 1938)

Vidal, E (1987). Jean Piaget and the liberal Protestant tradition. In M. G. Ash & W. R. Woodward (Eds.), Psychology in twentieth-cen- tury thought and society (pp. 271-294). Cambridge, England: Cam- bridge University Press.

Washburn, M. E (1926). The animal mind." A text-book of comparative psychology (3rd ed.). New York: MacMillan.

Waters, R. H. (1939). Morgan's canon and anthropomorphism. Psycho- logical Review, 46, 539-540.

Watson, J. B. (1907). Kinaesthetic and organic sensations: Their role in the reactions of the white rat to the maze. Psychological Mono- graphs, 7(2, Whole No. 33).

Watson, J. B. (1914). Behavior: An introduction to comparative psy- chology. New York: Henry Holt.

Wertz, E J. (1994). Of rats and psychologists: A study of the history and meaning of science. Theory & Psycholog}; 4, 165-197.

Yerkes, R. M. (1903). The instincts, habits and reactions of the frog.

Psychological Monographs, 4( 1, Whole No. 17).

Yerkes, R.M. (1913). Methods in comparative psychology [Lecture notes]. Akron, OH: Archives of the History of American Psychology, Beach Papers (Box M272).

1 1 4 4 O c t o b e r 1 9 9 8 • A m e r i c a n P s y c h o l o g i s t