By: Ava Malkin
Volume X – Issue I – Fall 2024
I. INTRODUCTION
In 1947, psychological experts Drs. Kenneth and Mamie Clark conducted an investigation, colloquially deemed the “doll test,” which played an integral role in the Brown v. Board of Education (1954) case and the future legal system, particularly in combination with social science research.
Conducted years before Brown made its way to the Supreme Court, this social scientific study used identical dolls of different races. These scholars explained that their results indicated decreased self- esteem, racial awareness, and internalization of value judgments in African American children, as young participants responded to questions in a way that outwardly favored white dolls over black ones. During Brown, Dr. Kenneth Clark utilized these findings in his testimony as evidence that African American students endured psychological harm—an impaired self-image—due to segregation, thereby arguing for a legal and psychological need for change. [1] The Court then cited this evidence as part of their decision in the Brown case; thus, the Clarks’ findings contributed to Brown’s overturning of the “separate but equal” doctrine from Plessy v. Ferguson (1896). [2]
The impacts of this research extended far beyond Brown, as the use of the “doll test” in this landmark case marked the first time the Supreme Court utilized a psychological investigation, thereby altering the evidence type and procedure in many future cases. [3] However, this psychological study and its applications in the Court faced much criticism, as academic scholars disapproved and legal scholars opposed the Clarks’ methodological errors. Skeptics also condemned the Court’s reliance on the “doll test” for the Brown ruling, demonstrating the implications of this study for other discrimination-based cases. [4]
To fully comprehend the importance of the Clarks’ contributions to the realms of research, law, and psychology, this paper explores what exactly their investigation entailed, how lawyers employed the study during Brown, why this evidence faced criticism, and how the “doll test” impacted social scientific evidence in future court cases.
II. DESCRIPTION OF THE “DOLL TEST”
Dr. Kenneth Clark and Dr. Mamie Clark were the first two African American individuals to earn a Ph.D. in psychology from Columbia University in the early 1940s. [5] In 1947, the couple engaged this psychological expertise to conduct a series of academic analyses surrounding racial identification, particularly in young African American students amidst segregation in schools and constant discrimination. The most notable of these experiments is known as the “doll test.”
In this investigation, the researchers used four dolls, identical except for their skin and hair colors: They included two white dolls with light hair and two black dolls with dark hair. (Interestingly, researchers actually had to paint the African American dolls brown because toy companies did not manufacture Black dolls at the time. [6]) Black children aged three to seven participated in this experiment, around half of which attended segregated schools. After placing the dolls in front of the children, the scholars asked them eight questions, prompting the students to point to a doll in an effort to determine the young pupils’ racial perceptions, preferences, and attribution associations. [7] Specifically, the Clarks asked which doll the children would like to play with, which doll looked “nice,” which doll looked “bad,” which doll looked white, which doll looked colored, and which doll looked like them (i.e., the child participant). [8]
Most children responded in a way that favored the white doll, associating positive attributes with it. They specifically answered that the white doll looked “nicer” and that they preferred to play with it. [9] When asked to select the doll that looked like them, many children cried or left the testing room, while other participants smiled and referred to the black doll using a racial slur. Dr. Kenneth Clark noted these responses as upsetting and “disturbing.” [10] The researchers then employed these results to identify a damaged self-esteem among young African American pupils, which they attributed to internalized racist messages due to the widespread segregation and discrimination during this time. [11] “These children saw themselves as inferior and they accepted the inferiority as part of reality,” Dr. Kenneth Clark explained. [12] This psychological harm became a legal context for the Brown case.
III. IMPORTANCE OF THE “DOLL TEST” FOR THE BROWN OUTCOME
Dr. Kenneth Clark testified as an expert witness in one of the lower court cases synthesized into the Brown case. In this testimony, he explained his knowledge of the psychological research surrounding racial preferences, particularly relying on his “doll test” as evidence of the harms facing African American children due to school segregation. This assisted the 1954 Warren Court in comprehending the effects of segregation on its closest, most malleable subjects: children. [13] With this newfound understanding of the mental consequences of separating young pupils based on their race, the Court reviewed four state cases and chose to overturn Plessy and its “separate but equal” doctrine, declaring that segregation was a violation of the Equal Protection Clause of the Fourteenth Amendment. [14]
The Supreme Court cited the Clarks’ work in its decision. Chief Justice Earl Warren incorporated social science evidence into Footnote 11 of the opinion, which legal scholars explain “contributed to an increasingly empirical equal education opportunity doctrine.” [15] While the opinion did not explicitly mention the Clarks by name, J. Warren wrote, “To separate [children in grade and high schools] from others of similar age and qualifications solely because of their race generates a feeling of inferiority as to their status in the community that may affect their hearts and minds in a way unlikely ever to be undone.” [16] Hence, he argued that, as the “doll test” pointed out, children in segregated schools experienced irrevocable insecurity and internalized racial narratives that could only begin to be repaired by integration and the elimination of the “separate but equal” doctrine. J. Warren further defended this notion under the claim that this lower self-confidence “is amply supported by modern authority.” [17] In other words, without outright referencing Drs. Clark and their experiment, he recognized that sufficient academic research existed to illustrate the psychological harm facing the young African American population.
IV. CRITICISMS OF THE “DOLL TEST”
i. Academic Concerns surrounding the “Doll Test”
After its recognition in Brown via Footnote 11, immediate controversy surrounded the Clarks’ “doll test.” The main criticisms began in the 1960s and 1970s, specifically focusing on the study’s methodological and statistical downfalls. [18]
The technical critique mostly emphasized how the Clarks’ research contained a small sample size of 16 participants and lacked a control group, representing a lower level of quality and complexity than the standard. [19] Many also alleged that the researchers themselves could have influenced the participants’ answers because the Clarks were both African American. Additionally, because the Clarks painted the black dolls due to a lack of diverse dolls within the market, they could have appeared unusual or unrecognizable to the children, further altering their responses.
Some even expressed concern that the question order may have affected participants’ answers. Considering the fact that the children’s responses to the first few questions generally reflected negative perceptions of the black doll, the children may not have wanted to identify with the black doll at the end of the experiment. [20] Finally, scholars were hesitant to agree with the Clarks’ causal claim between lower self-esteem and segregation among African American children because research revealed that participants who did not experience segregation carried the same inferiority perceptions. Therefore, many legal experts render the Clarks’ conclusions illegitimate, claiming the evidence “did not approach the level of and stating that the “doll test” did not accurately establish the negative relationship scientific proof” between self-esteem and segregation. [21] One legal scholar even wrote that “a competent cross-examination might have neutralized the testimony by revealing the fallacies in the test.” [22]
Furthermore, future investigations found that the “doll test” might not have measured self-esteem at all, implying that the Clarks’ perception and the Court’s argument of low self-confidence might not have been accurate during Brown. [23] Because critics worried that Brown relied too much on imperfect social scientific evidence, they worried about the implications of the decision as social science research evolved. This caused many individuals to reject the psychological implications described in Footnote 11, believing that the Court should make conclusions solely based on traditional constitutional values. [24]
ii. References to the “Doll Test” in Other Cases”
Denunciation of the “doll test” and its use in Brown was not limited to academic perspectives, as it expanded into legal realms. Other court cases also cited and attempted to delegitimize the Clarks’ research and its impact on the Brown decision. In Stell v. Savannah-Chatham County Board of Education (1963), for example, the plaintiffs protested desegregation in schools (i.e., disputing the Brown ruling). In their arguments against integration, plaintiffs claimed that Dr. Kenneth Clark was the sole witness for the “doll test” in Brown. This meant that Clark would have explained his own study, thereby calling into question the validity of his research and claims. Judge Francis Muir Scarlett wrote in his opinion that “[n]ot only was Dr. Clark the first authority cited by the Supreme Court in the Footnote (11) but he was the only witness in that case who testified on the basis of tests conducted by himself.” Scarlett continued by highlighting the aforementioned flaws in the experiment:
Apart from various experimental fallacies described as scientifically invalidating Dr. Clark's conclusions as to injury from segregation, the Court considers most significant testimony that the test on which Dr. Clark relied for much of his oral testimony, and conclusion in Brown, was a test on only 16 children in a segregated school area showing a result directly contrary to an earlier test of 300 children in both separate and mixed schools. It was further stated that Dr. Clark had made the later test and limited it to 16 selected children for the purpose of affording a basis for his conclusions in Brown. [25]
Despite these claims, and although the federal district court ruled in the plaintiffs’ favor, the Fifth Circuit Court of Appeals rejected this argument, stating that the Brown decision decisively deemed segregation in schools unconstitutional.
Stell was not the only case that attempted to target the illegitimacy of the Clarks’ investigation. Evers v. Jackson Municipal Separate School District (1964) also highlighted the “doll test” as inadequate evidence of lower self-esteem. Even a decade after Brown, some schools, including the Jackson Municipal Separate School District, had not desegregated. A civil rights activist named Medgar Evers filed a lawsuit, marking an initial attempt to promote inclusion, eliminate racial segregation in the South, and enforce the Brown ruling. In Evers, Judge Sidney Carr Mize recognized how the “doll test” highlighted a potential loss of racial identification and therefore personality deficits because of segregation. However, she also recognized that a similar investigation showed no personality injury or lost racial identity, thereby rendering the “doll test” on its own to be “unworthy of belief.” Specifically, the opinion stated:
The same test was shown to have been conducted by a Negro principal of unquestioned integrity on 85 Negro school children in the segregated schools in Jackson. Ninety-five per cent of those in Jackson identified themselves with the Negro doll and showed a complete absence of the personality injury which Dr. Clark testified that he found in his test of 16 which formed the sole basis for his testimony as to personality damage in Brown. In another study by the same Dr. Kenneth B. Clark, not called to the attention of the Supreme Court in Brown, involving many scores of Negro children in integrated and segregated situations in the North and South, it was reported that injury from personality conflict, if any, is suffered primarily by Negro children reared and schooled in integrated classes of the North—not in the segregated schools of the South. From this corroborating evidence, I am forced to find that the principal evidence of injury relied on by the Supreme Court in Brown was unworthy of belief. [26]
Similar to Stell, even though legal officials recognized the downfalls of the “doll test,” the court held for the plaintiffs, claiming that the Supreme Court’s decision in Brown still stood and that schools must begin to integrate.
V. THE “DOLL TEST” AS A LANDMARK CASE FOR SOCIAL SCIENCE IN COURTS
Regardless of these criticisms, the “doll test” still holds extreme value for the field of law in general, even with its flawed methods and conclusions. The “doll test’s” application in Brown marks the first instance that the Court cited a psychological investigation in a decision. The American Psychological Association labels this “scientific psychology's moment in a great spotlight.” [27] Because the Court used the experiment to eliminate school segregation, the Brown decision granted social science more respect and recognition. [28]
Future legal cases even recognized this landmark distinction and authorization of social science research in the courts. For instance, in Sangmeister v. Woodard (1977)—a case that determined that the practice of placing a particular party first on voting ballots violated the Equal Protection Clause of the Fourteenth Amendment, the judge of the Seventh Circuit Court of Appeals cited the Court’s use of social scientific investigations in Brown to validate their own implementation of similar studies. Judge Robert Arthur Sprecher stated, “[T[he Supreme Court has never taken so limited a position on the evidentiary uses of social science studies. In Brown v. Board of Education, 347 U.S. 483, 98 L. Ed. 873, 74 S. Ct. 686 (1954), the Court relied on several general social science studies in concluding that segregation was harmful to black children.” He continued,” “Those studies were cited and relied on notwithstanding the fact that Professor Kenneth Clark had conducted specific doll tests on the black children who were the plaintiffs in one of the consolidated cases and testified as to the results of those tests.” [29] Thus, Sangmeister serves as confirmation that the “doll test” and its application entirely altered evidence processes.
By underscoring the practicality of psychology in legal realms, the “doll test” paved the way for more legal cases beyond Sangmeister to employ social scientific research. Michele Bisaccia Meitl, an assistant professor of criminal justice, investigates the use of social science research in the courts, stating that judges are “experts in law and precedent, but they may not be experts in brain development or [jury] deliberation… They look to the outside world… to inform them of these things.” [30] Bisaccia Meitl discovered that around 40% of “the criminal procedure cases decided by the Court where Fourth, Fifth, Sixth, Eighth, or Fourteenth Amendment issues were decided” between 2001 and 2015 used social science research, which is a significant increase from 14% in 1990 and 10% in 1978. [31]
This increasingly large portion of cases employing social science investigations likely has roots in the “doll test.” Since Brown, the courts now use social science research, including expert testimony, related/existing studies, public opinion polls, and original investigations specific to that case, which assists legal professionals in understanding the contexts and implications of their cases. [32] While the Supreme Court did not establish its standard for the admission of scientific evidence until Daubert v. Merrell Dow Pharmaceuticals, Inc. (1993), the “doll test” clearly set a standard that permitted social science to make its way into courts, and it generated a level of respect for psychological fields in law. [33]
VI. CONCLUSION
Overall, the “doll test” served as an incremental investigation to justify the claim that segregation actively and irrevocably harmed young African American children, playing a crucial role in the decision of the well-known, impactful Brown case. With such influence came many objections, and the “doll test’s” integral function in desegregation yielded many critical interpretations. However, it is essential to note that some of these denunciations in court cases (e.g., Stell) were rooted in deeper racism and a refusal to integrate despite the Brown ruling. However, hesitation surrounding the Clarks’ academic methodology, namely their sample size, biases, question order, and causal claims, are valid and necessary; yet, that does not detract from the fact that this experiment was monumental for many legal cases to come, as it warranted and promoted future use of psychological investigations in courts.
Therefore, not only did this doll study highlight a critical implication of racial segregation plaguing the lives of many living in the segregated world of the early 20th century, but it also helped begin to counteract discrimination and legalized racism while also authorizing the use of social scientific evidence in courts.
Endnotes
[1] Legal Defense Fund. 2024. “Brown v. Board: The Significance of the ‘Doll Test.’” NAACP Legal Defense and Educational Fund. 2024. https://www.naacpldf.org/brown-vs-board/significance-doll-test/.
[2] Bergner, Gwen. 2009. “Black Children, White Preference: Brown v. Board, the Doll Tests, and the Politics of Self-Esteem.” American Quarterly 61 (2): 299–332. https://doi.org/10.1353/aq.0.0070
[3] Benjamin, Ludy T., Ellen M. Crouse, and American Psychological Association. 2002. “The American Psychological Association’s Response to Brown v. Board of Education. The Case of Kenneth B. Clark.” The American Psychologist 57 (1): 38–50. https://pubmed.ncbi.nlm.nih.gov/11885301/.
[4] Heise, Michael. 2005. Review of Brown v. Board of Education, Footnote 11, and Multidisciplinarity. Cornell Law Review 90 (2): 279–320. https://heinonline.org/HOL/Page?handle=hein.journals/clqv90&id=293&collection=journals&index=.
[5] Legal Defense Fund. “Brown v. Board.”
[6] Tjandra, Kristel. 2021. “Science, Civil Rights, and the Doll Test.” Peaceful Science, February 2021. https://peacefulscience.org/articles/science-civil-rights-and-the-doll-test/.
[7] Legal Defense Fund. “Brown v. Board.”
[8] Tjandra. “Science, Civil Rights.”
[9] Bergner. “Black Children, White Preference.”
[10] Blakemore, Erin. 2018. “How Dolls Helped Win Brown v. Board of Education.” HISTORY. September 2018. https://www.history.com/news/brown-v-board-of-education-doll-experiment.
[11] Legal Defense Fund. “Brown v. Board.”
[12] Blakemore. “How Dolls Helped Win Brown v. Board.”
[13] Ibid.
[14] Brown v. Bd. of Educ., 347 U.S. 483, 74 S. Ct. 686 (1954).
[15] Heise. Review of Brown v. Board.
[16] Brown 347 U.S. 483.
[17] Ibid.
[18] Bergner. “Black Children, White Preference.”
[19] Heise. Review of Brown v. Board.
[20] Tjandra. “Science, Civil Rights.”
[21] Heise. Review of Brown v. Board.
[22] Cahn, Edmond. 1956. Review of Jurisprudence: 1956 Survey of American Law. NYU Law Review 31 (1): 182–96. https://heinonline.org/HOL/Page?handle=hein.journals/nylr31&id=192&collection=journals&index=.
[23] Bergner. “Black Children, White Preference.”
[24] Heise. Review of Brown v. Board.
[25] Stell v. Savannah-Chatham Cty. Bd. of Educ., 220 F. Supp. 667 (S.D. Ga. 1963).
[26] Evers v. Jackson Mun. Separate Sch. Dist., 232 F. Supp. 241 (S.D. Miss. 1964).
[27] Benjamin, Ellen, and American Psychological Association. “The American Psychological Association’s Response.”
[28] “The Doll Study – the Legacy of Dr. Kenneth B. Clark.” n.d. CUNY Academic Commons. https://kennethclark.commons.gc.cuny.edu/the-doll-study/.
[29] Sangmeister v. Woodard, 565 F.2d 460, 466 (7th Cir. 1977)
[30] Collier, Caroline. 2022. “U.S. Supreme Court Justices Use Social Science Research in Rulings.” TCU Magazine. May 10, 2022. https://magazine.tcu.edu/summer-2022/supreme-court-uses-social-science-michele-meitl/.
[31] Bisaccia Meitl, Michele, Nicole Leeper Piquero, and Alex Piquero. 2019. “The Gradual Warm-Up: The United States Supreme Court’s Reliance on Social Science Research in Constitutional Criminal Law and Procedure Opinions, 2001-2015.” Deviant Behavior 41 (12): 1575–84. https://doi.org/10.1080/01639625.2019.1635290.
[32] Collins, S M. 1978. Review of Use of Social Research in the Courts. Knowledge and Policy: The UncertainConnection. https://www.ojp.gov/ncjrs/virtual-library/abstracts/use-social-research-courts-knowledge-and-policy-uncertain.
[33] Monahan, John, and Laurens Walker. 1990. Social Science in Law.