Running head: SHAPER OF EXPERIMENTAL PSYCHOLOGY













G. E. Müller: The Shaper of Experimental Psychology



Edward J. Haupt

Montclair State University

haupt@email.njin.net






















































































Paper based on presentation, "G. E. Müller," invited address to Divisions 3 and 26, American Psychological Association, August .., 1995.

G. E. Müller: The Shaper of Experimental Psychology

Edward J. Haupt

Montclair State University

Table of Contents


INTRODUCTION 1

BRIEF BIOGRAPHY 1

MÜLLER'S AMERICAN STUDENTS 3

MÜLLER'S WORK 5

Müller's Work in the Core of American Experimental Psychology 6

Psychophysics 6

Memory 9

Müller's Work Rejected from the American Canon 12

Cognition and Herbartian mechanics 12

Color theory after Hering 13

The Physicalist/Behaviorist Rejection of Müller's Work from the American Canon 16

MÜLLER AS EXPERIMENTAL PSYCHOLOGY PERSONIFIED 17

SUMMARY 18

G. E. Müller: The Shaper of Experimental Psychology

Edward J. Haupt

Montclair State University

INTRODUCTION

After Georg Elias Müller died in 1934, his journal, the Zeitschrift für Psychologie, gave him the sort of treatment that we have only, in recent times, given to B. F. Skinner--an entire issue was dedicated to his work and his memory. Following German tradition, the speech at the grave by his student and Director of the Psychological Clinic in the city of Hannover, Wilhelm Hische (1935), came first. It was followed by 40 remarkably comprehensive pages written by his student, Oswald Kroh (1935), later to be the first professor of psychology at the Free University of Berlin. The last part was 60 pages by another student, Erich Jaensch (1935), which described the inspiration which Müller had provided for him. No obituary appeared in the Times of London or the New York Times.

It is clear, however, from E. G. Boring's (1950) descriptions below, that there was good reason to honor Müller so. Boring (p. 379) concluded his section on Müller with the quote,

E. G. Boring (1950, p. 379) concludes his section on G. E. (Georg Elias) Müller with the quote,

... As a power and an institution he [G.E. Müller] was second only to Wundt.

which, however, seems like an understatement when the range of laudatory quotes, which include

an attribution of the origin of Titchener's structuralism to Müller

G. E. Müller at one time placed the work Komplextheorie in opposition to Gestalttheorie, but the word complexes has a meaning pre-empted by psychoanalysis ... Titchener similarly onced used for this position [Komplextheorie] the term 'structural' psychology; ... Gestaltpsychologie was even once translated into English "structural psychology," thus adopting the name of tis chief enemy. (p. 431)

and, conversely, that Müller

G. E. Müller was closest to Titchener's position. (p. 419)

describes Müller's role in psychophysics as

... becoming on Fechner's death [1887] the leader in psychophysics ... (p. 374),

characterizes Müller's contributions to memory as

Ebbinghaus had opened up a field [memory] which the patience of G. E. Müller and his associates was soon to develop ... (p. 388),

contrasts it with Wundt's research on memory

[Wundt's studies] made little impression as compared with the effective research upon memory by Ebbinghaus and G. E. Müller that belongs to the same period. (p. 343),

describes Müller's importance for color theory

Müller adopted Hering's theory of the three reversible photochemical substances ... and added his concept of cortical gray as the zero-point from which all color sensations diverge. (p. 376),

calls Müller's color theory

a theory so well known as often to be called Hering's theory, which it includes. (p. 376),

summarizes Müller's research interests

In these three fields [psychophysics, memory, color], he took over problems from their originators, criticized them, corrected them, extended them and centered research about them. (p. 374),

includes Krohn's (1893) portrayal of Müller's Göttingen laboratory

'in many respects the best for research work in all Germany.' (p. 374),

and says, of Müller's students

The men who worked under him at Göttingen furnish the second most distinguished group of names of German psychologists, for the students of Wundt must surely rank first. (p. 374).

BRIEF BIOGRAPHY

Georg Elias Nathanael Müller, the third surviving child of August Friedrich Müller and Rosalie Zehme, was the second son of the Pfarrer (both pastor and minister), religion teacher, and German teacher of the prestigious Saxon Gymnasium sometimes called the Illustre moldanum (illustrious school on the Mulde river) on July 16, 1850, the 300th year of the school. After his Abitur, Elias (as he was often known), completed a year of philosophy studies at each of the Leipzig and Berlin universities before he volunteered as a replacement soldier in the elite Prussian regiment usually nicknamed "Alexander" after the Russian Czars who were its nominal commanders. On his return to Leipzig, he spent one further semester with Moritz Wilhelm Drobisch, his first mentor, before continuing his studies with Rudolf Hermann Lotze in Göttingen. A year later, in 1873, he completed his dissertation, one of the sources of E. B. Titchener's discussion of attention (1908).

After three years, spent primarily as a resident teacher (Hauslehrer), he returned to present the Habilitation which enabled him to take the first step on the university career ladder and become a Privatdozent (in the German-speaking cultural areas, a doctorate was not sufficient to begin a university career; a "second dissertation" (Habilitationschrift) and a trial lecture, each reviewed by a committee of the Ordinarius professors, was required). This work, published in 1878, began his restructuring of psychophysics, although Müller did not complete this work until 25 years later. Through the firm support of Lotze, he first became Ordinarius in the recently founded Bukovinian university in Tschernowyzy for the Winter Semester of 1880 (Czernowitz in the orthography of the Austro-Hungarian Empire), and one semester later, in April of 1881, succeeded Lotze at Göttingen. He had begun to experiment in 1879, with weights provided by Fechner, and in 1887 got his first official support; those funds paid for the first memory drum. The subsequent experiments, aided by Friedrich Schumann, the first person to make a Habilitation with Müller as his sponsor, included the first retroactive transfer experiment.

Only after this publication did Müller attract many students who met his standards, including Victor Henri, Lillie Martin, Narziss Ach, Joseph Fröbes, Géza Révész, David Katz, Erich Jaensch, Walter Baade, Edgar Rubin, Oswald Kroh, and Georg Katona. While Müller was a close second on the short list for the Berlin chair in 1894, he was not chosen when Stumpf turned it down. His importance to experimental psychology, however, was established in 1904 by his position as chair of the German Society for Experimental Psychology, a position he held until 1927, five years after his retirement. By the time he made his summary statement on psychophysics in 1903, his research interests had broadened to include color theory and the psychology of thought, although his main publications in those areas were only to come later. The war and the subsequent inflation damaged the universities, but Müller had a cluster of students after the war and even allowed Wolfgang Köhler to hold the chair for a year while Müller lectured, just before Müller's legally enforced retirement in 1922. After retiring, he took up color theory in earnest, and after the publication of 1300 pages on color theory in 1930 (Müller 1930a; 1930b), still, with Käthe Müller's assistance, continued to publish small studies. He died just before Christmas 1934, renowned in the worldwide psychological community.

While Boring gave Müller generous treatment (certainly less than his due) more recent texts have tended to place Müller among the secondary figures in German psychology. Since almost nothing of Müller's work was translated and he had few American students, we now know very little of Müller's work. What follows is an all-too-brief overview of Müller's American students, his work, some thoughts on why particular parts of this work are less well known, and some of the characterizations of him as a sort of experimenter's experimenter.

MÜLLER'S AMERICAN STUDENTS

While Müller's German students, such as Oswald Külpe, Friedrich Schumann, Narziss Ach, Géza Révész, David Katz, Joseph Fröbes, and Erich Jaensch, form a dominant stratum of the next generation of German psychology, Müller had several American students who deserve to be better known. The first of these was Christine Ladd-Franklin, who, in 1891, got the part of her German education in color theory devoted to Hering from Müller (Franklin, 1973, p. 158).

Lillie Martin, when she arrived in Göttingen in 1894, was only one year younger than her mentor and had left a prestigious position as science teacher and vice-principal of Girls High School in San Francisco. She is the only American student who spent five years learning the new psychology in Germany; she apparently chose not to seek a doctorate; and in the end, she published much more than a dissertation with Müller as second author (Martin & Müller, 1900)--the only time Müller took that role. This book was an important precursor of Titchener's psychophysics (Titchener, 1905; introduction), and the document in which the method of right and wrong cases was renamed the method of constant stimuli. She maintained her role as an American Suffragette, since she took the place of Mrs. August Belmont as head of the American delegation in a London march. Through the accident of her acquaintance with David Starr Jordan (the then President of Stanford University), she became Frank Angell's sabbatical replacement at Stanford in 1899-1900, rose to full professor there, returned frequently to Germany to work, received an honorary doctorate from the University of Bonn for work which Oswald Külpe found important to thought research, was the only woman teacher of this generation who could supervise doctoral dissertations by men, and after her retirement, developed clinical psychology (she had apparently spent a year with Forel learning about hypnotism) and invented geriatric clinical work. It is easy to consider her the most distinguished of the first generation of American women psychologists.

Margaret Keiver Smith had taught psychology at the New York State Normal School at Oswego. She then translated Herbart's Lehrbuch zur Psychologie (1895), thus providing the first direct treatment of Herbart's mathematizations in an American publication. Several years later, she spent a year in Göttingen, went on to work with Ernst Meumann in Zürich for several years in experimental pedagogy and receive a doctorate there in 1900, contributed to Wundt's Festschrift, and returned to New Paltz Normal School where she directed the study of psychology and languages from 1901 to 1909.

Lottie Steffens, as did her sister Laura, got a bachelor's degree from Stanford. These sisters of Lincoln Steffens followed their brother to Leipzig to study psychology, but after one semester left for Göttingen. Lottie completed a doctorate on part vs. whole learning, the first psychology doctorate that Göttingen gave to a woman. She seems not have used this degree in any professional way. Laura published a paper on motorical attitude (motorische Einstellung) that probably could have served as a dissertation, but did not take the examinations for the doctorate.

Eleanor Acheson McCulloch Gamble completed her Ph.D. with E. B. Titchener on the psychophysics of odor, which made her, for many years, the leading American expert on the psychophysics of smell, and she regularly published reviews of new research on this topic. In 1906 she was given unpaid leave by Wellesley for a post-doctoral year with Müller (only full professors were allowed sabbaticals). She learned much of her craft there, and, one year after her return, became the director of the Wellesley Psychological Laboratory; a post she held until her death in 1933. She was, thus, for the next quarter century the exclusive provider to Wellesley undergraduate and graduate students of basic training in laboratory skills. Almost all of the theses and most of the published work from this laboratory were in Müller's tradition.

Thorleif Grüner-Hegge (usually T. G. Hegge), who had completed his dissertation in Oslo, went to Göttingen and continued Müller's cognitive work with the mathematician/-"memory-artist" Rückle. Hegge became the Scientific Director of the Wayne State Training School in Michigan in 1929 and remained in that post into the 1950s. As director of this research/educational component, he attempted to make the study of teaching retarded persons more scientific, basing it on experimental studies of memory. One of the significant residues of his work remains in the Hegge-Kirk-Kirk remedial reading drills and Samuel A. Kirk's book, Teaching reading to slow-learning children (1940). Under his auspices, Cruikshank, Boyd McCandless, and Sidney Bijou also started research programs that were important to the early development of American Special Education. Hegge thus served as a godparent to this important component of applied psychology.

That George Katona did his dissertation with Müller is scarcely known. However much his book, Organizing and Memorizing (1940), is seen as Gestalt-oriented, reading the book shows that he considered Müller's work important--and included it. His later work in economics and psychology, aside from its mathematical sophistication, shows little imprint of his teacher. And William McDougall, who spent a semester at Göttingen in 1900, shows even less effect.

Thus, without counting those who did not study with Müller, but followed his work (e.g., Titchener on psychophysics; Karl M. Dallenbach on memory; Deane Judd on color theory), Müller had a significant presence in American psychology. The number of women students is also remarkable, since Wundt had only one woman doctorate and Carl Stumpf had no women doctorates, although Rosa [Heine] Katz did work with him during World War I.

MÜLLER'S WORK

In spite of an almost complete lack of translation of Müller's work and (in my judgment) distortion of much of his work in Woodworth's presentations, the importance of Müller's work was still appreciated in the 1950s. In each of the three major works on experimental psychology, Stevens (1951), Osgood (1953), and Woodworth and Schlosberg (1954), there are more citations of Müller than of Wundt.

The scope of Müller's work is rarely appreciated. While few authors expect to match the more than 50,000 pages of Wundt or Piaget, Müller's work comprises more than 6000 published pages; almost all these works are primary contributions to the scholarly literature; most of his non-research works were critical reviews in high-prestige journals which were often cited; only 124 pages could be called a textbook.

Müller's work encompasses four principal areas, psychophysics, memory, thought psychology, and color theory. The psychophysics anticipates the color theory, which is founded on Hering, and the memory work leads into the thought psychology, which is founded on Herbart and Külpe.

What has surprised me most is the selective use of the work in the U.S.; two parts of Müller's work, psychophysics and memory became elements of the core of American experimental psychology. Two other parts, in spite of clear American representation (Ladd-Franklin and Deane Judd for color; Lillie Martin, Eleanor Gamble, and Grüner-Hegge for Denkpsychologie) were almost completely ignored in the construction of the American psychology that took place in the 1920s.

Müller's Work in the Core of American Experimental Psychology.

Psychology includes many examples of a discontinuity between the originator of an area of study and another scholar who later developed the techniques which make the work a substantial part of the psychological canon. While we recognize E. L. Thorndike as the originator of the study of behavioral consequences (i.e., the law of effect), it is clearly Skinner who elaborated the formidable array of techniques that became operant conditioning. It was Müller's role or choice to have such a second position in both psychophysics and memory.

Psychophysics. That Müller was important to psychophysics is frequently remarked; in fact, it was Müller who was responsible for the American version that J. P. Guilford (1954, p. 25) called "classical" psychophysics (Haupt, in press).

Müller started his study of psychology in the middle of the origin of psychophysics. Drobisch, Müller's first philosophy teacher, had sponsored Fechner's physics work at Leipzig; Müller took courses from Fechner in each of his first two semesters at Leipzig in 1868-1869, and Lotze, Müller's sponsor at Göttingen, Fechner's former student, and a close personal friend of Fechner, had announced Fechner's results to the world in 1852. Fechner worked closely with E. H. Weber (Lotze's anatomy and physiology professor), whose brother, Wilhelm, was Müller's Göttingen physics teacher and a member of both Müller's doctoral examining and Habilitation committees. And, by the time that Müller chose thresholds as the foundation (thus, Grundlegung) of psychophysics for his topic in 1873, scholars such as Théodule Ribot had written (La psychologie Allemande) that psychophysics would form the basis of the new scientific psychology. Thus, this work, which would first enable Müller to become a member of the faculty at Göttingen, was an ambitious topic and one which would be closely watched by the intellectual community which Müller thereby entered. Fechner rewarded Müller both with a set of weights like the ones he had used and by pointing out, in the foreword to his Revision der Hauptpuncte der Psychophysik, that Müller's work made this new book, four years before Fechner's death, necessary.

Müller even told a story about himself (Misch, 1934-1935, p. 48) in which it seemed that others knew he would do psychophysics before he did. The story concerns the young Müller's meeting at a railroad station a very special couple, Hermann Grimm (son of Wilhelm) and his wife who was often known by her unmarried name, Gisela von Arnim. Both Gisela von Arnim and her mother, Bettina, were part of Fechner's circle of acquaintances; a group centered on the Leipzig music publisher, Herman Haertel. Frau Grimm was known for her outgoing manner, and she engaged the young Müller in conversation, at the end of which she said, "Oh. You are so like Fechner!"

Psychophysics comprises two major parts--thresholds and scaling. Scaling, in the controversy between the logarithmic law and the power law, has received far more attention than the development of threshold methods, which, after Müller, were not a matter of controversy. This problem of scaling has three sorts of solutions; a simple Fechner solution, in which it is assumed that the ratio of the difference threshold to the stimulus value at which it is found is constant; a power law, with more parameters, but a vastly simplified methodology, and an agnostic middle ground which uses measured thresholds as the metric. Thresholds are important for the first and third of these methods and have many other reasons for being studied as well.

Fechner was not always the champion of Fechner (logarithmic) scaling and he revised his ideas on this topic; this was an active area of controversy in the 1870s and 1880s. While Fechner's Elemente used three methods (limits, average error, and constant stimuli) for studying thresholds, it is Müller, in his Habilitation of 1876 (which, after one and one-half years of further refinement became his first book; Müller, 1878), who makes the decisive turn to the method of constant stimuli as the preferred choice (Haupt, in press). Müller's mathematical sophistication in this area is clear and nowhere represented in American texts. While the decision to split doubtful judgments differently from Fechner is known, the 10 pages of calculus in the Grundlegung that supported it is nowhere discussed. In his successor paper (1879), Müller presented what became the Müller weights. (These weights are used in connection with the raw data--physical stimulus values and percentages correct--from a method of constant stimuli experiment; they are used to determine the regression constants from which the Upper Limen and Lower Limen are estimated, and make the linear regression equation produce values which are closer to those found with a proper fit to the non-linear cumulative normal curve function). The rationale for these weights reveals knowledge that I have not seen represented in American texts. Müller was apparently able to fit the normal ogive to psychophysical data using non-linear methods. The simpler method, of a linear fit of the z (normal curve) functions to the stimulus values, did not give equal predicted values. The Müller weights were a method to match the non-linear fit more closely. And he demonstrated that the method produced the desired result.

Müller was also responsible for explanation of psychophysical judgment of weight lifting in terms of muscle response, thus emphasizing his deep commitment to physiological explanation of psychophysical results; he had a theory of absolute impression, one of the major attempts to provide a psychological theory for the nature of the sensation which is responded to. Müller's work should be better known in English, but Titchener made giving the list of Müller's contributions to psychophysics largely unnecessary with his Experimental Psychology, which he dedicated to Delboeuf, very nearly Fechner's first critic.

Müller's second psychophysical publication was his extensive criticism of Fechner's second book, In Sachen der Psychophysik (In matters of psychophysics), in which Fechner replied to critics such as Plateau, Delboeuf, and Hering. The third such work, in Pflügers Archiv (Müller, 1879), was the justification of and use of the Müller weights; this weighting function remained a standard procedure until Urban's work created a major debate about the choice of weights. After a small work with Schumann, Müller's next papers on psychophysics are the important works on visual sensations which include the "psychophysical axioms." This was followed by a last major empirical work with Lillie Martin (1900), followed by the summary (Müller, 1903), which Titchener (1905) used as his lodestone. These publications are not only important, they are the works which defined the threshold methods of psychophysics until it was expanded by ideas of signal detection theory.

Memory. We have overemphasized Ebbinghaus in the history of the study of memory. Ebbinghaus did memory experiments in 1879 and in 1883-1884, and published a combined result (1964). He carried out no further experiments in this area and did not create any of the later standard methods (see Haupt, in press, for a concise summary).

Briefly, in a publication of 190 pages (in issues which appeared in 1893), Müller originated the use of the memory drum and established the standard methods for serial anticipation (Müller and Schumann, 1894). Müller presented very clearly the advantages of the memory drum, which include anything that a modern experimental psychologist might consider. In addition, he purposely maximized, using the variable speed control, the rate of the memory drum individually for each participant to eliminate rehearsals between presentations of a list and between syllables. That Müller made some improvements in the syllable set is known; his extensive and meticulous efforts to provide lists of equal difficulty are less frequently remarked. He was the first to study retroactive inhibition and, while he was not the first to study proactive inhibition, he was the first to use nonsense syllables to study proactive inhibition. He also performed the first transfer experiment in which the two conditions (trochaic and iambic accenting of the series of 12 syllables) were factorially organized; thus Müller planned for all combinations of methods in training and in test, as is done in state-dependent memory experiments. Another experiment in this paper compared the importance of associations and serial position in the relearning of elements from a serial anticipation list. The paper includes as well an articulate discussion of single-blind (in which only the subject is unaware of the goals of the experiment) and double-blind controls (in which both subject and experimenter are unaware). I have described these studies more fully elsewhere (Haupt, in press). Müller also established the standards for methods of paired associates, elaborated the concepts of consolidation and perseveration, and further developed the memory drum in a later book (Müller and Pilzecker, 1900).

The theoretical questions Müller employed provide a sense of what was important for theories of memory, particularly those of Herbart, in the years before the turn of the century. In these experiments, the methods of investigation originated designs which only became widely used many years later. Such designs included factorial designs, transfer-test list paradigms, initial list learning to equate syllable (response) learning and which define the associations which can later be used through the imposition of a trochaic rhythm, and what seems to be the first example of a retroactive transfer study. More than the "methodology," these studies introduced important theoretical issues, such as the control of rehearsal, interference, and the use of procedural controls for distinguishing among theories; these techniques became part of the standard repertoire of American memory research. In particular, these thirteen experiments seem to have established the standards which were assiduously followed, even if their source was not so well understood, for serial anticipation procedures.

Thus, this Müller and Schumann paper (as well as the later Müller and Pilzecker work) established conceptual problems and methodological standards for many aspects of memory experimentation. While the studies seem to be relatively well known for equating the items which were to be remembered and the meters, many other components which provided models for later research are present; these components seem to be scarcely known.

In addition, well-known studies, frequently cited in American work, were done by Müller's students. Adolf Jost (1897) carried out an important investigation which showed the relation of memory strength and memory decay. Lottie Steffens (1900) did a well-known study on the efficiency of part vs. whole learning. The title, however, (Experimentelle Beiträge zur Lehre von ökonomischen Lernens), shows that the topic was indebted to Mach's notions that the true nature of thought could be explained by least-effort principles. Later, Rosa Heine (1914; who was later to become David Katz' wife) showed that there was no evidence to be found for retroactive inhibition in recognition memory. While Ebbinghaus founded the experimental study of memory, with this range of achievement, Müller established its standards.

It is not easy to find a full appreciation of Müller's work in American sources; when you look for Müller's work in standard American memory texts such as McGeoch (1942) or Hovland's (1951) chapter in the Stevens Handbook, the absence of page numbers in the citations tells more about their authors' secondhand knowledge of Müller than the high praise he is given. The absence of attribution for Müller's ideas, and thus for his direct knowledge of these ideas, by Woodworth often amazes and distresses me. Thus, in one of American psychology's classic texts on memory, McGeogh (1942, p. 402) identified Müller and Schumann's (1894) priority in bringing "The problem [of negative transfer] to a clear experimental focus ..." as the "Müller-Schumann law of associative inhibition, "when any two items, as A and B, have been associated, it is more difficult to form an association between either and a third item, K." This statement of what we now call proactive inhibition clearly ignores the origin of such ideas in Herbart and Müller and Schumann's (1894) use of "associative inhibition" as an umbrella term for both proactive and retroactive inhibition.

Müller's Work Rejected from the American Canon

However, Müller, continuing his psychological work, developed two further areas which are little known to American historians of psychology. By themselves, Müller's contributions to Denkpsychologie (the experimental study of thought) and color theory would qualify him as a major contributor to the development of experimental psychology, even if he had not done anything else. They can scarcely be found in American psychology textbooks.

Cognition and Herbartian mechanics. That Müller developed his memory work on the basis of Herbartian ideas might be disputed, but I think that would not get far with readers who are well acquainted with Herbart. Thus, Müller's study of memory was characterized as a Vorstellungsmechanik, or a mechanics of the dynamics of mental flow, an idea that would surely have upset James (who often ridiculed Herbartian concepts, e.g., 1890, p. 603). While much similar cognitive work seems to belong to Külpe (the leader of the Würzburg school, who was much more Müller's student than Wundt's) and his "Würzburg" school, I believe that Müller was early implicated in the study of Denkpsychologie. One of the early summary works on thought processes (Über die Willenstätigkeit) was written by Narziss Ach (1905). It is well to remember that, by this time, Ach had made the first step to a university teaching career (habilitated) under Müller's sponsorship and had been Müller's Assistent for four years. Under those conditions, and given Müller's reputation for criticalness, it is unlikely that Ach would have published anything on the topic which Müller thought was ill-founded.

To develop ideas about cognition, Müller went on to write three major volumes on thought psychology (they appeared as supplementary volumes to the Zeitschrift für Psychologie 1911 [vol. I], 1913 [vol. III], and 1917 [vol. II]). These volumes contain the empirical development of Müller's theory of complexes, a word that specifically recalls Herbart (e.g, Murray, 1995, pp. 81-83). Complexes were used to describe everything from the linking of colors and shapes to the linking of elements into thoughts, in ways which were later elaborated by Otto Selz as Külpe's student. This Komplextheorie was later deployed as an alternative to the newly emerging Gestalt work and forms the empirical heart of Müller's criticism of Köhler (Müller, 1923). These densely packed volumes were concerned with, among other things, an investigation of a mnemonist (more precisely, "artist" of calculations and Göttingen Ph. D. in mathematics) Rückle. Some of these results were further elaborated by Grüner-Hegge in his later role as an administrator for retarded children in Detroit.

Very little of this work can be found in English. Müller's theory is rarely presented comprehensively, and when someone like Humphrey (1951) discusses thinking, Müller is only indirectly mentioned through citations of Ach. Brown and Deffenbacher (1975), in a plea to restudy the work do cite Müller, but it is the only citation in PsychLit. However, a treatment of some of Müller's work on grouping can be found in George Katona's (1940) book, Organizing and Memorizing. It is only through citations of Müller's former Assistent Ach and Külpe's student Selz, that an inkling of the Komplextheorie fitfully appears. This Komplextheorie was the prominent alternative to Gestalt cognitive theories in the 1920s, as the controversy between Müller and Köhler shows.

Color theory after Hering. The development of color theory is often portrayed as an unrelenting combat between the theories of Helmholtz and Hering, and the incompatibility of the two theories seems to have been a common idea in the 1920s. This ignores the proposal by Donders in 1881 (Judd, 1949, p. 1) and the development, as early as 1905 by von Kries (Judd, 1951, p. 836), of two-stage color theories. These two-stage color theories accepted the further development of the Helmholtz position by König and Dieterici of the three receptor pigments and proposed a transformation of these receptor functions, through some neural or chemical mechanism, into Hering's opponent colors; the physiological responses of these colors were first demonstrated by De Valois in the late 1950s and further developed by Hubel & Wiesel through the 1960s to 1980s. Müller's summary contribution, published in nearly 1300 pages of supplementary volumes to the Zeitschrift für Psychologie (Müller, 1930a; Müller, 1930b), was characterized by Judd (1951, p. 836) as the best articulated of these two-stage color theories.

Müller and Hering knew each other well; and color theory was long an interest of Müller's. Müller was responsible for the awarding of an honorary degree in philosophy to Hering in 1887 and Hering returned the favor, from Leipzig's medical faculty, 10 years later (in between, as noted before, Christine Ladd-Franklin learned her Hering from Müller). In the same year, Müller presented his first major paper on visual function, the "Psychologie der Gesichtsempfindung (The psychology of visual sensations)." Jaensch (1935) clearly identifies the strength of their relationship.

Müller's contributions to this field were many. Before the turn of the century, Müller had originated the concept sometimes called "cortical" or "intrinsic/subjective" gray (Augengrau), which provided an elaboration of Hering's theories from phenomenological observations (Katz, 1935a). This intrinsic neutral gray of these subtle observations was often theorized to be in the cortex, thus fulfilling the first of his psychophysical axioms. To improve the physical basis of Hering's equations, Müller attended his colleague Walter Nernst's classes on physical chemistry (one of the side benefits of Göttingen) and apparently was the first to use reversible chemical equations to characterize the breakdown and recombination of photosensitive pigments as early as 1904. While Hering had put the locus of the opponent processes somewhat implausibly in adjacent cells, Müller managed to use the cortical neutral gray as a device for placing the opponent processes in a single cortical, but not receptor, cell.

The problems of including the clear finding of three pigments in the cones and the four psychological primaries further on required a solution that did not reject one while keeping the other, the only sort of solution that Boring seems to have understood. Müller was among the first to propose such a multi-stage theory that would put the receptor pigments first and transform their response to the four primaries of the mind (see Massof, 1985). Müller's theory is thus a clear predecessor of the important work of Hurvich and Jameson (1955).

From my compilation of Müller's publications, it is clear that Müller articulated many concepts about color until his death in December 1934. In the course of these publications (Katz, 1935a, pp. 76-80), he also emphasized concepts which we would identify as deriving from very subtle descriptions of color; as if Husserl's call for phenomenology as the fundamental psychology had been especially well heeded by Müller (Müller' students did attend Husserl's classes [Schuhmann, 1977]). Such concepts included the "pronouncedness" (Ausgeprägtheit) of a particular gray in which the same color chip's appearance could vary depending on its distance from the observer; and the "penetratingness-/insistence" (Eindringlichkeit) of a color, the quality which describes the way a color disappears as it becomes more peripheral (p. 196). Thus, he also experimentally supported the concept that lightness level is one-dimensional.

The Physicalist/Behaviorist Rejection of Müller's Work from the American Canon.

There is very little of Müller's extensive psychological work that we would not be better for, were we to pay more attention to it today. However, American psychology accepted part of it and rejected other parts. Why?

In order to understand the absence of Müller's cognitive theories (Komplextheorie), I would like to point to the 1920s, when American psychology was not simply behaviorist, but produced a general sense that more abhorred mentalism/spiritualism than desired to make behavior a standard. Dewsbury, writing about Norman R. F. Maier (1993), has mentioned that Clifford Morgan considered Maier's concept of conflict to be outside the permissible range of "real science" (p. 875). Thus, the anti-cognitive sentiments of American psychology in the 1930s and 1940s were part of Müller's problem. We might get a less distorted view of Müller's importance, however, if we had queried Gestalt psychologists about him when they first arrived in the 1930s, as they were still creating formulations which excluded the Komplextheorie.

More strongly even than for psychophysics or memory, Müller is a major conceptual as well as methodological contributor to the 20th century development of vision theory. Unfortunately, Müller's major work did not appear until 1930, and apparently was ignored by most Americans. Since Müller produced such important work in color theory, how can we account for his neglect? In the 1920s Helmholtz was canonized by the Optical Society's translation of his Physiologische Optik--a choice which pointedly ignores Hering and theories closely related to his. This choice seems to be the result of an American interpretation of color theory in the 1920s, led by the Columbia University Professor of Biophysics, Selig Hecht. This scientific direction seems to me best called physicalism, and much to derive from Jacques Loeb. While he was almost completely ignored in the U. S., in Germany, Müller, as a color theorist, never really went out of style, since his pupils David Katz and Edgar Rubin continued to acknowledge him into the 1950s. We probably need to look to a few sources, such as David Katz, Wilhelm Ostwald, and Deane Judd, for a fuller appreciation of Müller's role.

As American physicists, physiologists, and psychologists adopted a color theory that refused to accommodate Helmholtz to Hering, it is no surprise that Müller felt himself somewhat isolated. Germany, too, was still suffering the effects of the war and the hyperinflation of 1923, and the high cost of American journals to Germans also contributed to difficulties in getting academic sources. Thus, in response to letters by Boring which requested a biography, Müller, who was somewhat reticent, and did not find it easy to write autobiographically, did write something. In some letters of 1929 (selections of which appear in Boring's 1935 Obituary for Müller), show that the quid pro quo for details about Müller's early academic life was a collection of papers on color theory which were not then available in Germany. These papers then formed the background for Müller's two volume work on color theory (Müller 1930a, 1930b) that anticipated Hurvich and Jameson.

Thus Müller's theoretical contributions to the study of thought and color, while scarcely ever mentioned in American texts, were a major influence for European psychologists and physiologists in the early 20th century. Here, I have argued that two components, psychophysics and memory, were simply adopted from Germany. The acceptance of parts of this German legacy, while ignoring other parts equally well established, argues against any simple interpretation on purely nationalistic grounds.

MÜLLER AS EXPERIMENTAL PSYCHOLOGY PERSONIFIED

It was widely acknowledged that work from Müller's laboratory represented a gold standard for psychological research in the areas (psychophysics, memory, cognition, color vision) in which he worked. David Katz (1935b, p. 235), in his European eulogy for his teacher described Müller as the methodologist of experimental psychology. Joseph Fröbes (1931, p. 127) described his teacher at the time he started to study at Göttingen in a similar way, "George Elias Mueller, ... , who was never equalled in exactness and method, was at that time [1902] at the height of his influence." In the Spring of 1910, Raymond Dodge, then teaching at Wesleyan, spent the second half of his sabbatical working under Max Verworn, during Verworn's last semester at Göttingen before he moved to Bonn. Dodge was (most probably) a Versuchsperson (experimental participant) in experiments in Müller's lab. He described this time (Dodge, 1961, p. 116) as follows: "It was during this sabbatical that I served as observer for G. E. Mueller and came to know him as a friend. His methodological thoroughness and industry I have never seen surpassed." Since Dodge, under Benno Erdmann's direction, had been a major originator of tachistoscopic research, perhaps the most demanding of psychological methods, this is very high praise.

The praise was mixed with a judgment that Müller could be inordinately critical. Rosa [Heine] Katz described the exactness thus, "Müller was very critical, one became doubtful, believed the work was not worth anything at all ... In spite of the crude things which Müller could say to you, you didn't hold it against him, but never forgot what a unique schooling in [psychological] methods one received from him". (Müller war sehr kritisch, man wird verzweifelt, glaubte die Arbeit tauge ueberhaupt nichts. ... Trotz der Grobheiten, die Müller einem sagen konnte, trug man es ihm nicht nach, sondern vergaß niemals welch einzigartige methodische Ausbildung man bei ihm bekam.; Katz, R. 1972, p. 106). In a personal communication, Werner Traxel (a tale told by Heinrich Düker, who had studied in Göttingen) described Müller as flying into a rage if he saw a loose bolt lying on the floor. Müller's paucity of praise was represented in understanding that David Katz was one of a few with whom Müller shook hands after the completion of an oral examination. While we might wish for a kinder approach to students, this combination of precision and criticism is far from unknown today.

SUMMARY

In the preceding pages, I have sketched some critical ways G. E. Müller influenced early experimental psychology. His students formed a presence, though not a large one, in American psychology--and the neglect of Lillie Martin often shocks me.

Müller's work formed the basis of two widely known components of American experimental psychology: the ascendance in psychophysics of constant stimulus methods and the verbal learning methods of serial anticipation and paired associates, interference theory, the standardization of methods and important fundamental findings. While Müller's efforts clearly formed the basis of American work, many American texts show gross ignorance of the detailed contents of this work.

Müller's work, clearly advanced and important at the time, also included a well-developed two-stage theory of color vision and extensive work on cognitive processes which was intended to apply to both thought processes and perception, his Komplextheorie. Both of these topics failed to penetrate American psychology, except in very limited ways; much of this rejection can be traced to the limitations (which supposedly made it more like a natural science) which American psychology placed on itself in the 1920s and 1930s. Much of this has to do with the manner in which American psychology constituted itself in the 1920s as well as the dominant psychological ideology, which rejected any concepts which hinted at mentalism, since they might well be contaminated with spiritualism. In fact, since Müller's work exemplified the natural science tradition informed by subtle use of introspection, the effect was to deny many people knowledge of an important range of equally-scientific research.

Finally, I have used some of the descriptions of Müller by his associates to point out the way in which he, as no other contemporary, exemplified the spirit of using meticulous natural science methods to create an experimental psychology. Let us hope that the guiding spirit of the experimental practice which Müller exemplified will continue to guide psychological science.

References

Ach, N. (1905). Über die Willenstätigkeit und das Denken [On the activity of the will and thought]. Göttingen: Vanderhoek und Ruprecht.

Boring, E. G. (1950). A history of experimental psychology, 2nd Ed. New York: New York: Appleton-Century-Crofts.

Brown, E., & Deffenbacher, K. (1975). Forgotten mnemonists. Journal of the History of the Behavioral Sciences, 11(1), 312-319.

Dewsbury, D. (1993). On publishing controversy: Norman R. F. Maier and the genesis of seizures. American Psychologist, 48(8), 869-877.

Dodge, R. (1961). Raymond Dodge. In C. Murchison (Ed.), A history of psychology in autobiography, Vol. I (pp. 99-121). New York: Russell and Russell.

Ebbinghaus, H. (1964). Memory: A contribution to experimental psychology. (H. A. Ruger & C. E. Bussenius, Trans.). New York: Dover Publications. Original work published 1885)

Fechner, G. T. (1889). Elemente der Psychophysik (2. Unveränderte Auflage) [Elements of psychophysics. Second unaltered edition]. Leipzig: Breitkopf & Härtel.

Franklin, C. L. (1973). Colour and colour theories. New York: Arno Press. (Reprint of 1929 edition published by Harcourt Brace, New York.)

Fröbes, J. (1961). Joseph Froebes. In C. Murchison (Ed.), A history of Psychology in Autobiography, Vol. 3 (pp. 121-152). New York: Russell and Russell.

Guilford, J. P. (1954). Psychological measurement (2nd ed.). New York: McGraw-Hill.

Haupt, E. J. (In press). G. E. Müller as a source of American psychology: The Zur Grundlegung der Psychophysik and Untersuchung des Gedächtnisses. In R. W. Rieber & K. Salzinger (Eds.), Psychology: Theoretical and historical perspectives (rev. ed.). Washington DC: American Psychological Association. Heine, R. (1914). Über Widererkennen und rückwirkende Hemmung. [On recognition and retroactive [massed] inhibition] Zeitschrift für Psychologie, 68, 161-236.

Herbart, J. F. (1895). A text-book in psychology; an attempt to found the science of psychology on experience, metaphysics and mathematics of 1828 (M. K. Smith, Trans.; Lehrbuch der Psychologie; ein Versuch die Wissenschaft der Psychologie auf Erfahrung, Metaphysik, und Mathematik of 1828) (Copyright, 1891). New York: D. Appleton and Company.

Hische, W. (1935). Gedenkrede, gehalten am Sarge von Georg Elias Müller [Memorial speech, made at the coffin of Georg Elias Müller] Zeitschrift für Psychologie, 135(3), 145-149.

Hovland, C. R. (1951). Memory. In S. S. Stevens (Ed.), Handbook of experimental psychology (pp. 613-689). New York: Wiley.

Humphrey, G. (1951). Thinking. London: Methuen.

Hurvich, L. M., & Jameson, D. (1955). Some quantitative aspects of an opponent-colors theory. II. Brightness, saturation, and hue in normal and dichromatic vision. Journal of the Optical Society of America, 45, 602-616.

Jaensch, E. R. (1935). Was wird aus dem Werk? Betrachtungen aus dem Gesichtspunkt der Kulturwende über G. E. Müllers Wesen und Werk und das Schicksal der Psychologie [What will become of the efforts? Observations from the viewpoint of the turnabout of the culture on the essence of G. E. Müller's works and the fate of psychology]. Zeitschrift für Psychologie, 135(3), 191-218.

James, W. (1890). Principles of psychology. New York: Holt.

Jost, A. (1897). Die Assoziationsfestigkeit in ihrer Abhängigkeit von der Verteilung der Wiederholungen [The strength of associations in their dependence on the distribution of repetitions]. Zeitschrift fur Psychologie und Physiologie der Sinnesorgane, 16, 436-472.

Judd, D. (1949). Response functions for types of vision according to the Müller theory. (Research paper RP 1946). Journal of Research of the National Bureau of Standards, 42, 1-16.

Judd, D. B. (1951). Basic correlates of the visual stimulus. In S. S. Stevens (Ed.), Handbook of experimental psychology (pp. 811-867). New York: Wiley.

Katona, G. (1940). Organizing and memorizing. New York: Columbia University Press.

Katz, D. (1935a). The world of colour. London: Kegan Paul, Trench, Trubner.

Katz, D. (1935b). G. E. Müller+. Acta Psychologica, 1, 234-240.

Katz, R. (1972). Rosa Katz. In L. J. Pongratz, W. Traxel, & E. G. Wehner (Eds.), Psychologie in Selbstdarstellung, Bd. 1 [Psychology in self-portrayal, Vol. 1]. (pp. 103-125). Bern: Verlag Hans Huber.

Kirk, S. A. (1940). Teaching reading to slow-learning children. Boston: Houghton-Mifflin Company.

Kroh, O. (1935). Georg Elias Müller. Ein Nachruf [Georg Elias Müller. A eulogy]. Zeitschrift für Psychologie, 135(3), 150-190.

Massof, R. W. (1985). Color-vision theory and linear models of color vision. Color Research and Application, 10, 133-146.

McGeoch, J. A. (1942). The psychology of human learning. New York: Longmans, Green and Co.

Misch, G. (1934-1935). Georg Elias Müller. Nachrichten der königlichen Akademie der Wissenschaften und der Georg Augustus Universität zu Göttingen in Hannover, 45-59.

Martin, L. J. & Müller, G. E. (1899). Zur Analyse der Unterschiedsempfindlichkeit. Experimentelle Beiträge [On the analysis of sensitivity for differences. Experimental contributions]. Leipzig: J.A. Barth.

Müller, G. E. (1878). Zur Grundlegung der Psychophysik. Kritische Beiträge [On the founding of psychophysics. Critical contributions]. Berlin: T. Grieben. Bibliothek fur Wissenschaft und Literatur, 23. Band. Philosophische Abtheilung. 4. bd. (Habilitation, S. 424; 2. Aufl. Berlin: Hoffmann).

Müller, G. E. (1879). Ueber die Massbestimmung des Ortsinnes der Haut mittels der Methode der richtigen und falschen Fälle [On the determination of measurement of the sense of place of the skin by means of the method of right and wrong cases]. Pflügers Archiv für die gesammte Physiologie, 19, 191-235. [Archiv für die Gesellschaft der Physiologie]

Müller, G. E. (1903). Die Gesichtspunkte und die Tatsachen der psychophysischen Methodik [Viewpoints and the facts of psychophysical methodology]. In L. Asher und K. Spiro (Hrsg.), Ergebnisse der Physiologie, Jahrgang II, 2. Abtheilung, 267-516. Wiesbaden: J. F. Bergmann. (Reprinted 1904, pp. 1-244.)

Müller, G. E. (1923). Komplextheorie und Gestalttheorie; Ein Beitrag zur Wahrnehmungspsychologie [The theory of complexes and the theory of Gestalts; a contribution toward a psychology of perception]. Gottingen: Vandenhoeck & Ruprecht.

Müller, G. E. (1930a). Uber die Farbenempfindungen. Psychophysiche Untersuchungen [On the sensations of color. Psychophysical inquiries]. Leipzig: J.A. Barth. (Zeitschrift für Psychologie. Erganzungsband 17)

Müller, G. E. (1930b). Uber die Farbenempfindungen. Psychophysiche Untersuchungen [On the sensations of color. Psychophysical inquiries]. Leipzig: J.A. Barth. (Zeitschrift für Psychologie. Erganzungsband 18)

Müller, G. E., & Pilzecker, A. (1900). Experimentelle Beiträge zur Lehre vom Gedächtniss [Experimental contributions on the theory of memory]. Leipzig: J.A. Barth. (Zeitschrift für Psychologie. Erganzungsband 1)

Müller, G. E., & Schumann, F. (1894). Experimentelle Beiträge zur Untersuchungen des Gedächtnisses [Experimental contributions on the investigation of memory]. Zeitschrift fur Psychologie und Physiologie der Sinnesorgane, 6(2), 81-190, 257-339.

Murray, D. J. (1995). Gestalt psychology and the cognitive revolution. Hemel Hemstead, Hertfordshire UK: Harvester Wheatsheaf (Simon & Schuster).

Osgood, C. E. (1953). Method and theory in experimental psychology. New York: Oxford University Press.

Schuhmann, K. (1977). Husserl-Chronik: Denk- und Lebensweg Edmund Husserls [Husserl-Chronicle: The thought- and lifeway of Edmund Husserl]. Den Haag: Martinus Nijhoff.

Steffens, L. (1900). Experimentelle Beiträge zur Lehre vom ökonomischen Lernen [Experimental contributions on the theory of economy in learning]. Zeitschrift für Psychologie und Physiologie der Sinnesorgane, XXII, 321-380.

Stevens, S. S. (Ed.). (1951). Handbook of experimental psychology. New York: Wiley.

Titchener, E. B. (1905). Experimental psychology, Vol. II, part II. New York: Macmillan. (pp. 275-318 for Müller's psychophysics; many citations for Müller's Grundlegung der Psychophysik in the Criticism section of the roman-numbered pages).

Titchener, E. B. (1908). Lectures on the elementary psychology of feeling and attention. New York: Macmillan.

Woodworth, R. S., & Schlosberg, H. (1954). Experimental psychology (2nd ed.). New York: Holt, Rinehart & Winston.