The Inkblots

by Damion Searls

Du Bois stayed in Atimelang for eighteen months. She learned the language, in fact named it Abui and was the first to give it written form. She interviewed the people, collected information on child rearing, adolescent rituals, and family dynamics, and recorded long autobiographies from many of the villagers. She found the Alorese in her community emotionally brittle and short-tempered, with frequent verbal and physical fights inside the family and out: these qualities, among others, made up their “culturally determined personality structuring.”

  But she needed to put her impressions on a more solid objective basis. So, upon her return, she gave the autobiographies and other data to Abram Kardiner at Columbia, prominent at the time as a psychoanalytic theorist (author of The Individual and His Society, 1939). She also, without sharing Kardiner’s analysis or her own observations, handed over Rorschach protocols from seventeen Alorese males and twenty females to another colleague: Emil Oberholzer.

  All three reached the same conclusions about the Alorese. Oberholzer found, for instance, that “the Alorese are suspicious and distrustful….This fearfulness is something that is part and parcel of their natural and normal emotional disposition….They are not only easily upset and frightened, easily startled…but also they easily fly into a passion. There must be emotional outbursts and tempers, anger and rage, sometimes resulting in violent actions,” just what Du Bois had found in person. Their analyses of specific people—whose autobiographies Du Bois had collected, whom Kardiner psychoanalyzed from afar based on the documents, and whose Rorschach tests Oberholzer scored and interpreted—converged as well.

  In a letter to Ruth Benedict in February 1940, Du Bois called the overlap nothing short of stunning: “The crux of matters is whether individual data support Kardiner’s analysis from institutions. The Rorschachs seem to be giving him full confirmation….Oberholzer and I are still working at them and O. is gratifyingly cautious. If the individual stuff really confirms K’s analysis I’m going to be permanently pie-eyed. It will be just too damned good to be true.” She recognized that Kardiner’s analysis was suspiciously consistent with her own impressions; perhaps she had unintentionally cherry-picked or skewed the data she had given him. “But I can’t have tampered with the Rorschachs. Oberholzer, without really knowing the other implications, is as excited as I am. He knows nothing about the culture except what is necessary to explain the responses. He views it as a triumph for Rorschachs and I for the whole business of interpreting from sociology to psychology. Exciting, isn’t it?”

  By then Du Bois was on the faculty at Sarah Lawrence College, where Rorschach testing was already in full swing. In 1941, she finished writing her synthesis The People of Alor, and she published it in 1944 with extensive essays by Kardiner and Oberholzer included. She had brought anthropology and psychology together.

  Like the Bleuler brothers before him, Oberholzer was testing the test. Could anything useful be learned from the Rorschach without already knowing the specific culture of the Alorese—which answers were popular or original, good or bad, which details were normal or rare, all the numerical norms that made scoring possible? The Alorese results certainly looked foreign at first sight: one woman’s responses to Card V (the “bat”) were:

  (1) like pigs’ legs (side projections)

  (2) like goat’s horns (top center, the “bat’s” ears)

  (3) like goat’s horns (bottom center)

  (4) like a crow (dark spot in the large portion)

  (5) like black cloth (half of the large portion)

  Nevertheless, Oberholzer ended up arguing that the test could see past these superficial cultural differences to the kinds of personality profile he then shared with Du Bois.

  The stakes for Du Bois were higher: she wanted to find out whether it was even possible to argue that culture shapes personality. Any such argument would be circular unless there was a source of information about personality that was independent of the behavior that anthropologists studied in describing cultures. This direct access to personality was just what the Rorschach claimed to provide. An EEG had traced its first record of human brain waves onto a roll of paper in 1934, but such neurotechnology had a very long way to go. Being able to actually see through cultural differences to the person within, with inkblots, would be, as Du Bois said, “just too damned good to be true.”

  The figure usually credited with bringing Lawrence Frank’s projective methods into the heart of American anthropology did so for just that reason. After studying with Ruth Benedict under Boas in 1922, A. Irving Hallowell (1892–1974) turned to problems of culture and personality. Between 1932 and 1940, he spent summers along the Berens, a small Canadian river flowing into Lake Winnipeg from a source almost three hundred miles to the east.

  This area was one of the last in North America to be explored by Europeans, “a country of labyrinthine waterways, swamps, glacier smoothed rocks, and unbroken forests.” Since the Berens linked no major lakes or rivers to Lake Winnipeg, the area had remained isolated. Nomadic hunters and fishermen, part of the Ojibwe or Chippewa group, lived in three separate communities reaching east into the wilderness: one on Lake Winnipeg at the mouth of the river, one about a hundred laborious miles inland—fifty portages if traveled by canoe, and there were no roads—and one even farther inland.

  This geography put the communities at three different stages along the “culture gradient” from precontact to full assimilation. The lakeside group, with white residents in the community and a semiweekly steamer from the city of Winnipeg during the summer, lived in European-style log houses with few obvious signs of traditional culture—no native ceremonies or dances, no sound of drums. Inland, though, where few traders or missionaries had ever gone, there were “birchbark-covered tipis…sharply defined against a background of the dark and stately spruces that line all horizons,” and a “flavor of old Indian life remained.” The Ojibwe there might wear woven clothes from the company store, cook with iron pots and pans, drink tea, chew gum, and eat candy bars, but men were still bringing moose carcasses to shore in their canoes while the women made moccasins, stitched birchbark with spruce roots, or chopped and hauled wood with babies strapped into the cradleboards on their backs. There were medicine men, conjuring lodges, midsummer wabanówīwin dances. “In this atmosphere,” Hallowell wrote, “one could not help but feel that, despite many outward appearances, much of the core of the aboriginal thought and belief still remained.” One could not help but feel, but how could one be sure?

  Hallowell first heard “the strange word Rorschach” in the mid-thirties, uttered by Ruth Benedict in a meeting of the National Research Council’s Committee on Personality in Relation to Culture. His Winnipeg fieldwork had led him into what he called the “newly emerging area of research in anthropology: the psychological interrelations of individuals and their culture,” and this new technique for revealing individual psychology beneath culture was just what Hallowell was looking for. He cobbled together enough information to try his hand at it, combining elements of Beck’s, Klopfer’s, and Hertz’s approaches and improvising a procedure to administer the test through a translator:

  “I am going to show you some cards one after another. These cards have marks”—here the interpreters inserted an Ojibwe term, ocipiegátewin, which meant “picture”—“on them, something like you see on this paper (trial blot shown). I want you to take each card in your hand (trial blot given to subject). Look at it carefully and point out what you see there with this stick. (Handing an orangewood stick to the subject.) Tell me everything that the marks on the card make you think of or what they look like. They may not look like anything you have seen but if they resemble something closely, mention whatever it is.”

  He returned from his next summer in Canada with dozens of Ojibwe Rorschach protocols in hand.

  Hallowell felt that the different stages of Ojibwe assimilation to white Canadian culture provided a perfect way to study the interrelations of individual psychologies and culture because by definition these stage
s meant subjecting the same psychology to different cultural forces. “If, as has been assumed, there are intimate connections between the organization of personality and culture patterns,” Hallowell wrote, “it follows that changes in culture might be expected to produce changes in personality.”

  Like Oberholzer with the Alorese, Hallowell claimed to detect in his Rorschachs “an Ojibwe personality constellation…clearly discernible through all levels of acculturation thus far studied.” While outward cultural practices may have been borrowed from white Canadians, there was “no evidence at all” of changes to the “vital core of native psychology.” He then argued that since the three groups along the Berens River shared the same heredity and cultural background and had been given their Rorschach tests under the same conditions, any differences between the groups’ results could only be due to their different level of acculturation. The Rorschach could reveal how different Ojibwe individuals were adapting, or not, to the new cultural pressures.

  What Hallowell found was the Ojibwe personality “being pushed to the limits.” The tests of the inland Ojibwe showed predominantly introvert results and a significant repression of any extravert tendencies, which made sense in a culture where events were always understood with reference to an inner belief system, dreams were the most important experiences and were processed in private (there was a taboo against sharing one’s dreams in most circumstances), and social relations were highly structured. The lakeside Indians, in contrast, “living in closer rapport with other persons and things in their environment,” showed a greater range of personalities, especially among the women, including significantly more extraverts. People could act extraverted if so inclined, instead of repressing that side of themselves.

  This greater freedom to be different, which Hallowell found especially in women, could be a good thing: 81 percent of the most well-adjusted individuals came from the lakeside group. At the same time, 75 percent of the most maladjusted came from the lakeside group as well. Hallowell concluded that white culture was more psychologically challenging, with a greater chance for failure to adapt as well as for more self-expression. “Some of these deductions might be made without the benefit of the Rorschach technique,” Hallowell wrote, “but it would be difficult to demonstrate them without a method of investigation through which the actual personal adjustment of concrete individuals could be evaluated.”

  As with Du Bois, the overall lesson went beyond the specific findings: if the Rorschach could detect cultural norms within individuals, then it could be used to study how culture shapes personality in general. In two groundbreaking articles, “The Rorschach Method as an Aid in the Study of Personalities in Primitive Societies” written for psychologists, and “The Rorschach Technique in the Study of Personality and Culture” written for anthropologists, Hallowell laid out the test’s unique advantages: it collected quantifiable, objective data; it was portable; people liked taking it; it did not require the test taker to be literate or the examiner to be a professional psychologist, since the protocol could be scored and interpreted by someone else; there was no risk that those who had already taken the test would tell their friends the right answers.

  Most important: Hallowell wrote that the Rorschach was “non-cultural.” The norms were surprisingly stable across different populations—popular responses were almost the same among European Americans and Ojibwe, for instance, except on one card that the former often saw as “an animal skin” while the Ojibwe tended to see “a turtle.” Besides, “since psychological meaning rather than statistical norms is the basis of most Rorschach interpretation,” Hallowell felt that valuable insights were possible even without large samples of data. At the time of his first essay, fewer than three hundred protocols from nonliterate cultures had ever been collected, including the Bleulers’, Du Bois’s, and Hallowell’s own. A few years later, the count had risen to more than twelve hundred, and the possibility of any future nonliterate group being un-Rorschachable, while “conceivable” to Hallowell, seemed “quite unlikely.”

  Even if the Rorschach could provide “non-cultural” information about personality, anthropologists using it faced another problem. Every culture had its “typical personality structure” but left room for individual variations, while cultures were assumed to be different and yet people were fundamentally the same. That meant any given result could be claimed to reveal idiosyncrasy within a society or to support a generalization about cultural differences, whichever the anthropologist wanted.

  A 1942 study of Samoans, inspired by Hallowell, recognized the dilemma. The Samoans gave an unusually high number of pure Color responses, which made the Samoans test as generally extraverted. But the study’s author, Philip Cook, argued that this was because of Samoan color vocabulary: the Samoan language had abstract words only for black, white, and red (mumu, “like fire, like flame,” almost always associated with blood), with rarer color words being closely tied to specific things (the word for “blue” meant “the color of the deep sea,” and referred to green or gray as the sea changed color; the word for “green” meant “the color of everything growing”). So Samoans were unlikely to describe things as colored—fewer FC responses. They also gave far more Anatomical responses, which in Europeans and Americans suggested “sex repression or morbid bodily preoccupation.” But since the Samoans were sexually active from a young age, with minimal if any sexual repression in the culture, their Anatomical responses were probably perfectly normal. Cook accepted that while the Rorschach seemed to reveal aspects of Samoan culture, it was unable to distinguish or diagnose individuals. But for Cook this meant only that massive further research should be conducted on each different culture, since “the Rorschach is undoubtedly an excellent instrument for the study of cultural psychodynamics.”

  These were the assumptions shared by psychology and anthropology. Hallowell had proposed a full theoretical integration of the two fields and had praised the Rorschach as “one of the best available means” to that end. By 1948, Hallowell was president of both the American Anthropological Association and Klopfer’s Rorschach Institute—a concrete sign of the convergence of the two disciplines. In the popular understanding as well, the Rorschach could discover the personality structure in anyone, whether in America or the most alien and exotic culture.

  The Rorschach “seemed like a mental X-ray machine,” recalled a graduate student at the time. “You could solve a person by showing them a picture.”

  On December 7, 1941, the Japanese attacked Pearl Harbor. Within three weeks, Bruno Klopfer had organized a “Rorschach volunteer unit” to coordinate between the Institute back home and its many members who had volunteered for service. He also put himself forward as the point man for information and advice about the test. By early 1942, questions and requests were trickling and then pouring in from the military, and Klopfer was soon working with the army personnel procedures division to see how the Rorschach could help in the US war effort.

  This Rorschach was a very different beast from the nuanced instrument at the forefront of anthropology and personality studies. First and foremost, the military needed efficient evaluations, along the lines of its Army General Classification Test, developed in 1940 and given to twelve million soldiers and marines over the next five years. Ruth Munroe, tester of entering classes at Sarah Lawrence College, had published her Inspection Technique, designed to help examiners scan Rorschach protocols quickly for striking problems. While less subtle, such a scan produced interpretations more uniform across different scorers, and much faster.

  To streamline test administration as well as scoring, Molly Harrower introduced the Group Rorschach Technique, where slides were shown in a semidarkened room and test takers wrote down their own answers. Twenty minutes were enough to test an auditorium of more than two hundred people. It was almost as hard to get the slides right as it had been for Rorschach to get his inkblots printed, especially considering “the great difficulties that arose in getting reliable film during the war years
,” but a photographer who could do it was eventually found.

  Group Rorschach used by the Office of Strategic Services for selection purposes during World War II

  Even with these advances, the Rorschach presented two major obstacles to mass use. Though less-expert staff could administer the test, protocols still had to be scored and interpreted by trained Rorschachers. Worse, the results could still not be boiled down to a simple number for bureaucrats, punch cards, or IBM scoring sheets. So Harrower went one step further, “departing so far from the essence of what Rorschach intended” that she had, by her own admission, really invented “an entirely different procedure,” which she called “A Multiple-Choice Test (For Use with Rorschach Cards or Slides).”

  From a list of ten responses for each card, test takers were asked to check one box for “the suggestion you think is the best description of the blot,” and put a 2 in the box for their second choice (optional). Card I (see this page), for instance, was:

  An army or navy emblem

  Mud and dirt

  A bat

  Nothing at all

  Two people

  A pelvis

  An x-ray picture

  Pincers of a crab

  A dirty mess

  Part of my body

  Something other than the above: _______________

  A top-secret answer key distinguished good answers from bad, and Harrower’s wartime article about the procedure described it with language out of a spy thriller: “Since it is of the utmost importance that this simple key does not fall into the wrong hands, it has not been published here. A copy will be sent immediately on request, however, to psychiatrists and psychologists in the Armed Forces.” Three bad answers or fewer and you passed the test, four or more and you failed.

  If you think this sounds a bit suspect, you’re not alone. “The Group Rorschach procedure was greeted with raised eyebrows,” Harrower later commented, but “the introduction of the Multiple-Choice Test met with an even colder reception.” Yet the need to screen millions called for new approaches. “In the last analysis,” as she had originally pointed out, a screening program is “much less interested in knowing in detail WHY the individual is unfit, provided we can spot him,” and “much less interested in an extremely sensitive instrument which only few persons can handle than in a simple tool which anyone can use, anywhere.”