What’s in a Name? The Hidden Historical Ideologies Embedded in the Black and African American Racial Labels

Abstract

History can inconspicuously repeat itself through words and language. We explored the association between the “Black” and “African American” racial labels and the ideologies of the historical movements within which they gained prominence (Civil Rights and Black Power, respectively). Two content analyses and two preregistered experimental studies (N = 1,204 White American adults) show that the associations between “Black” and “bias and discrimination” and between “African American” and “civil rights and equality” are evident in images, op-eds, and perceptions of organizations. Google Images search results for “Black people” evoke more racially victimized imagery than search results for “African American people” (Study 1), and op-eds that use the Black label contain more bias and discrimination content than those that use the African American label (Study 2). Finally, White Americans infer the ideologies of organizations by the racial label within the organization’s name (Studies 3 and 4). Consequently, these inferences guide the degree to which Whites support the organization financially.

Keywords

race racial bias discrimination prejudice intergroup processes social cognition preregistered

The past is never dead. It’s not even past.

—William Faulkner (1951, p. 73)

While covering an unprecedented rise in racial activism in summer 2020, media outlets toiled over how to respectfully label Americans of African descent (AADs). The Associated Press and The Wall Street Journal announced that they would capitalize “Black” but not “white” to honor AAD culture (Bauder, 2020). The Washington Post declared that it would allow the subjects of articles to choose African American, Black, or a more specific racial label (WashPostPR, 2020). Amid this typographical activism to respect AADs, the semantic consequences of using these different labels were not discussed and remain unclear. For example, does the racial label used to portray AADs alter observers’ interpretations of AADs’ advocacy for racial progress? We suggest that the two most commonly used labels—“African American” and “Black”—have each retained the valences and meanings associated with the different historical movements in which they were popularized. Further, we argue that these valences and meanings critically influence onlookers’ perceptions of AADs’ beliefs as well as support for AADs’ advocacy.

Two Social Movements, Two Labels, One Goal

The Black and African American racial labels gained prominence within separate social movements and were likely originally used in ways that supported those movements’ distinct ideologies of racial progress (i.e., beliefs about how to best achieve equality; Martin, 1991). Specifically, the Black and African American labels were championed within the Black Power and Civil Rights Movements, respectively. We theorize not only that these racial labels took on particular valences and meanings when used in the context of these movements but that when these labels are currently applied to targets (e.g., images, texts, groups), those targets will take on the valences and meanings consistent with those labels.

Many disciplines demonstrate that objects or institutions can retain meanings from the contexts in which they emerged (e.g., zoology, sociology; Lorenz, 1935). Sociological research on organizational imprinting shows how organizations retain vestiges from the historical environments in which they were founded (Johnson, 2007). Because of inertia and the constant reproduction of culture, this historical environment continues to define the organization even after that particular historical context has ended (Stinchcombe, 1965). For example, the ideological orientations held by U.S. fraternities often align with the historical movements in which they were birthed (Marquis & Tilcsik, 2013; Stinchcombe, 1965). Fraternities founded between 1840 and 1850, when the United States experienced a broad secular, anti-evangelical Christian movement, have more secular goals than those founded during other times.

Specific words may also retain the valence of historical periods because of the other words that often surrounded them in conversation. Indeed, psycholinguistic work demonstrates that words absorb the valence of other words that frequently follow them in natural language (i.e., semantic prosody; Hauser & Schwarz, 2018; Sinclair, 1991). For example, in common conversation, the word cause is most frequently followed by negatively valenced words such as death, problems, and damage (https://www.english-corpora.org/coca/), whereas the synonym produced is frequently followed by more neutral terms. Consequently, people perceive a person who caused changes more negatively than a person who produced changes (Hauser & Schwarz, 2018). For example, a fictitious candidate who was labeled as “causing” budget changes received fewer reelection votes than one labeled as “producing” budget changes (Hauser & Schwarz, 2018).

Similarly, we theorized that words can become imbued with the concepts and ideologies from the historical context in which they emerged and were frequently used. Specifically, if the terms Black and African American gained prominence within the rhetoric of social activists’ movements, their meanings would be derived from those ideological contexts. Stokely Carmichael championed the term Black in the 1960s within the context of the Black Power Movement (Martin, 1991). This movement adopted a relatively negative focus on the unequal status that AADs possessed, emphasizing that AADs were victimized by racial bias and subject to poor socioeconomic conditions (Joseph, 2009; Van Horne, 2007). Alternatively, Civil Rights leader Jesse Jackson championed the term African American in the late 1980s (Martin, 1991). The leaders of the Civil Rights Movement adopted a relatively positive focus on the status they hoped AADs could achieve, stressing political activism through equality and inclusion of AADs in American social and political spheres (e.g., voting; Aiken et al., 2013; Valocchi, 1996). We suggest that when AADs are labeled Black, or organizations have the Black label in their title, people are likely to believe that they support a bias and discrimination ideology consistent with the Black Power Movement. In contrast, when AADs or organizations are labeled African American, people are likely to believe that they promote a civil rights and equality ideology consistent with the Civil Rights Movement.

Statement of Relevance

Americans of African descent (AADs) have long fought for equality, but their advocacy is often misunderstood or misrepresented by White Americans. We highlight how communicators’ choice to use the “Black” or “African American” racial label may alter White Americans’ perceptions of AADs’ advocacy intentions. We show that when organizations are labeled as Black, Whites perceive that their intentions are consistent with the themes of the Black Power Movement (e.g., focused on bias and discrimination). In contrast, when organizations are labeled as African American, Whites perceive that their intentions are consistent with the themes of the Civil Rights Movement (e.g., focused on civil rights and equality). Because of these associations, application of these labels in traditional and social media may affect Whites’ support of AADs’ advocacy for racial progress. For example, we found that if Whites personally endorse bias and discrimination ideologies, they donate more money to AAD organizations labeled as Black than AAD organizations labeled as African American.

These associations may explain work showing that Whites view AADs labeled as Black more negatively than AADs labeled as African American (Hall et al., 2015). For example, Whites perceived a Black perpetrator more negatively than an African American perpetrator, and crime-related news articles that used the term Black contained more negative rhetoric than those that used the term African American. It is possible that the greater negativity associated with the Black than the African American label was due to the association of the Black label with the relatively negatively oriented focus on AADs’ unequal status promoted by the Black Power Movement in its push for racial progress.

We extended Hall and colleagues’ (2015) focus on valence to examine whether the Black and African American labels continue to carry meanings consistent with the historical contexts in which they were championed. We conducted two content-analysis studies and two preregistered experiments to examine the ideological content embedded in these racial labels. We made three predictions for how labels’ retained meanings from their respective movements may currently affect society. First, we predicted that, compared with the African American label, the Black label would be more associated with poverty, victimhood, and racial disadvantage, in addition to the more general negative connotations found by Hall et al. (2015). Additionally, we hypothesized that the Black label would be associated with more bias and discrimination and less civil rights and equality ideological content than the African American label. Finally, we predicted that White Americans’ own preferences for these two ideologies would guide their financial support for Black organizations over African American organizations. The four studies we conducted to test these hypotheses were reviewed and approved by the Emory University Institutional Review Board.

Study 1: Depictions of Blacks Versus African Americans

In Study 1, we used Google Images to examine the cultural associations that society imbues in racial labels. Recent research suggests that algorithms, such as those used in Google Images, may reflect the same cultural ideals espoused in society (Noble, 2018). Because the algorithm tags photos on the basis of the captions applied to those photos (Benjamin, 2019; Noble, 2018), we predicted that a Google Images keyword search for “Black people” would return images that were more negative and depicted more victimized targets than a keyword search for “African American people”).

Method

Participants

We recruited 387 adult participants from Amazon Mechanical Turk; however, following prior research (e.g., Hall et al., 2015), we retained only White American participants for the final analysis (N = 292; 99 women; age: M = 35.87 years, SD = 11.63). Survey data for this study were collected in March 2015. A sensitivity power analysis (1 – β = 0.95, α = .05, two tailed) indicated that our sample size of 292 would allow us to detect a Cohen’s d effect size of 0.19 with a probability of .95 (Faul et al., 2009).

Procedure

First, our research team conducted two separate Google Images searches for the phrases “African American people” and “Black people” and downloaded the first 100 images returned for each of the two searches. The 200 images were saved and numbered according to their search tag (e.g., “African American photo #1”).

Of note, Google searches are time sensitive (Pitkow et al., 2002), and they retrieve the most current cultural artifacts of the time. We searched for and downloaded all images before the Black Lives Matter Movement gained popularity in late 2014 (Pew Research Center, 2018). This provides evidence that the Black Lives Matter Movement was not the initial cause of the association between the Black label and the bias and discrimination ideology. Further, we tasked a research assistant, blind to the study’s hypothesis, to download all 100 images from each search on the same day. The searches for “African American people” and “Black people” were conducted on January 14 and 15, 2014, respectively.

Google searches are often personalized for the user (Lawrence, 2005) such that images and articles that fit the user’s interests rank higher within the search results. To investigate the plausibility of personalization affecting our findings, the first author took a screenshot of the first page of each search (“African American people” vs. “Black people”) using her profile. Then, the first author compared these images with those that the research assistant downloaded using his profile. Of note, the first author and the research assistant differed on a number of factors that may have affected Google’s personalization of their profiles (e.g., age, sex, race, geographic location, and research interests). Nonetheless, approximately 81% of the images that the first author captured also appeared in the research assistant’s search, indicating that the majority of the search result content was consistent across user profiles.

Next, we loaded the 200 photos onto an online survey platform for participants to evaluate. Each participant evaluated a total of 20 photos: 10 randomly selected from the “African American” photo set and 10 randomly selected from the “Black people” photo set. Participants were unaware of the search tags associated with each photo. Three questions measured how negative the photos were on a scale ranging from 1 (not at all) to 7 (extremely): “How negative is this depiction of this group member (or these group members)?” “How stereotypical is this depiction of this group member (or these group members)?” and “How derogatory is this depiction of this group member (or these group members)?” (α = .96). Three additional measures assessed victimization on a scale ranging from 1 (strongly disagree) to 7 (strongly agree): “This photo depicts a person or people of low socioeconomic status,” “This photo depicts a person or people who are disadvantaged,” and “This photo depicts a person or people who are victimized” (α = .96). A principal components factor analysis with varimax rotations indicated that items loaded onto two independent factors: The three negativity items loaded onto the first factor at .89 and above, and the three victimization items loaded onto the second factor at .92 and above. The rotation converged in three iterations.

Finally, participants responded to an experimental-check question to confirm the presence of AADs in the photo. We conducted our primary analysis on all 200 photos, but if we restricted our analysis to only those for which at least 90% of participants noted an AAD (i.e., 163 photos), the significance and direction of our effects were unchanged (see the Supplemental Material available online).

Results

As indicated in Table 1, participants perceived images from the “Black people” Google Images search to be significantly more negative than the images from the “African American people” Google Images search. Further, participants perceived that the “Black people” Google images depicted people who were significantly more victimized, more disadvantaged, and of lower socioeconomic status than the images from the “African American people” search.

Table 1.

Results of Google Images Search (Study 1)

Measure	Black(n = 100)	African American(n = 100)	Comparison
Measure	M (SD)	M (SD)	t	df	p	95% CI	d
Negativity	3.39 (1.53)	1.67 (0.60)	10.50	198	< .001	[−2.044, −1.307]	1.48
Victimization	2.76 (1.09)	2.30 (0.77)	3.44	198	.001	[−0.7214, −0.1951]	0.49

Note: Confidence intervals are shown for the difference between means.

Discussion

Consistent with the findings of Hall and colleagues (2015) and our predictions, the results of Study 1 showed that the top image-search results for “Black people” reflected more negativity and victimization themes than the top image-search results for “African American people.”

Study 2: Racial-Bias Imprints in Written Media

We predicted that, in Study 2, editorial text that uses the Black label would have a more negative emotional tone than editorial text that uses the African American label (replicating the findings of Study 1) and also be more likely to mention bias and discrimination themes than civil rights and equality themes. We analyzed op-eds for the presence of each label and measured the degree to which the text evoked each ideology and negative emotion.

Method

In December 2019, we used Factiva, an international news database, to collect all op-eds from a 40-year period (1980–2019) that had at least five mentions of one or more of the terms “Black,” “African American,” or “African-American” (in either their singular or plural forms). We limited our analysis to the top 10 online and print op-ed outlets as determined by The OpEd Project (2015): USA Today, The Wall Street Journal, The New York Times, New York Daily News, Los Angeles Times, The Washington Post, New York Post, Chicago Tribune, Houston Chronicle, and The Philadelphia Inquirer. The search found 6,183 op-ed results.

We conducted a pretest to develop a custom dictionary that best represented each ideology. We recruited 124 adult participants from an online survey platform and randomly assigned each of them to list the first five words that came to mind when they thought of either “bias and discrimination” or “civil rights and equality.” We compiled the results into a more easily manipulated list. We began with a free-response list of 493 traits and retained traits that had the consensus of at least two or more participants. We deleted words that emerged in both the “bias and discrimination” and “civil rights and equality” lists and any repeated traits. Finally, we deleted any label used to identify AADs (e.g., African American or African Americans) from both lists because these labels were the dependent variables of interest. The resulting list consisted of 30 traits to describe bias and discrimination and 31 traits to describe civil rights and equality (see Table 2).¹

Table 2.

List of Terms Used in Linguistic Inquiry Word Count (LIWC) Custom Dictionary (Study 2)

Bias and discrimination		Civil rights and equality
Unfair	Differen*	Freedom	Gay
Hate	Disabilit*	Justice	Hope*
Prejudic*	Disgust*	Fair*	Human*
Gender*	Disrespect*	Equal*	Important
Bad	Dumb	Right	KKK
Judg*	Evil	Democra*	Law*
Wrong*	Housing	March	Libert*
Favoritism	Media	Protest*	Love
Racis*	Men	Slavery	Minorit*
Intoleran*	Nepotis*	America	Peace
Republican	Preference	Good	Respect
Rude*	Sexuality	Martin	Rosa
Sexis*	Stereotype*	Equal*	South
Anger*	Terrible	Equit*	Struggle
Bigotry	Trump	Evolution	United*
			War*

Note: Asterisks acted as wild cards that prompted LIWC to search for all words beginning with that stem (e.g., “racis*” would yield results such as “racist” and “racism”).

Using these terms, we created a custom dictionary within the Linguistic Inquiry Word Count (LIWC2015) program to measure the existence of “bias and discrimination” and “civil rights and equality” terminology within the op-eds. Following Hall and colleagues (2015), we used the internal dictionary codings within LIWC2015 for the negative emotion terminology. The LIWC program systematically calculates the proportion of terms within a text (Pennebaker et al., 2015). To analyze the text at the paragraph level, we segmented blocks of text that were separated by at least two space delimiters and were more than 50 words in length. Further, we analyzed only paragraph segments that used either African American (n = 2,259, 12.3%) or Black (n = 16,046, 87.7%), and we eliminated paragraph segments that used both. This resulted in 18,305 paragraph segments.

Results

Overall, ordinary least squares regression analyses indicated that paragraphs that used the Black label were associated with higher bias and discrimination terminology than paragraphs that used the African American label. In contrast, paragraphs containing the African American label were associated with higher civil rights and equality terminology than paragraphs containing the Black label. Finally, paragraphs containing the Black label were associated with a more negative emotional tone than paragraphs containing the African American label (see Tables 3 –5).

Table 3.

Results of Ordinary Least Squares Regressions Predicting Bias and Discrimination Terminology in Op-Ed Text (Study 2)

Predictor	Model 1(1980–2019)	Model 2(1980–2019)	Model 3(1988–2019)	Model 4(1988–2019)	Model 5(1988–2013)	Model 6(1988–2013)
Black (vs. African American)	0.081** (0.022)	0.090** (0.022)	0.083** (0.022)	0.089** (0.022)	0.127** (0.025)	0.127** (0.025)
Year		0.003** (0.001)		0.002* (0.001)		0.000(0.001)
R²	.001	.001	.001	.001	.002	.002
N	18,305	18,305	17,320	17,320	15,836	15,836

Note: The regression coefficients are unstandardized; values in parentheses are standard errors.

p < .05. **p < .01.

Table 4.

Results of Ordinary Least Squares Regressions Predicting Civil Rights and Equality Terminology in Op-Ed Text (Study 2)

Predictor	Model 1(1980–2019)	Model 2(1980–2019)	Model 3(1988–2019)	Model 4(1988–2019)	Model 5(1988–2013)	Model 6(1988–2013)
Black (vs. African American)	−0.283** (0.031)	−0.303** (0.031)	−0.228** (0.031)	−0.224** (0.031)	−0.219** (0.034)	−0.219** (0.034)
Year		−0.007** (0.001)		0.002(0.002)		−0.001(0.002)
R²	.004	.006	.003	.003	.003	.003
N	18,305	18,305	17,320	17,320	15,836	15,836

Note: The regression coefficients are unstandardized; values in parentheses are standard errors.

p < .01.

Table 5.

Results of Ordinary Least Squares Regressions Predicting Negative Emotion Terminology in Op-Ed Text (Study 2)

Predictor	Model 1(1980–2019)	Model 2(1980–2019)	Model 3(1988–2019)	Model 4(1988–2019)	Model 5(1988–2013)	Model 6(1988–2013)
Black (vs. African American)	0.193** (0.036)	0.192** (0.037)	0.205** (0.037)	0.208** (0.037)	0.243** (0.042)	0.242** (0.042)
Year		0.000(0.002)		0.001(0.002)		−0.005* (0.002)
R²	.002	.002	.002	.002	.002	.002
N	18,305	18,305	17,320	17,320	15,836	15,836

Note: The regression coefficients are unstandardized; values in parentheses are standard errors.

p < .05. **p < .01.

We conducted our regression analyses over three different periods and included year as a covariate (see Tables 3 –5). Jessie Jackson formally introduced the term African American into the lexicon in 1988, and it is unlikely that the public commonly used the term before then. Thus, Models 3, 4, 5, and 6 restricted the sample to op-eds authored after 1988. Similarly, the hashtag #BlackLivesMatter, a slogan committed to raising awareness about racial injustice, gained prominence in 2014 (Pew Research Center, 2018). To prevent confusion between this popular movement’s association with racial bias and our original hypothesis, we restricted the sample in Models 5 and 6 to op-eds authored before 2014.

Model 6, the most conservative model, remained highly significant for both the bias and discrimination (b = 0.127, p < .001, R² = .002) and civil rights and inequality (b = −0.245, p < .001, R² = .003) ideological themes as well as for negative emotion terminology (b = 0.242, p < .001, R² = .002). Paragraph segments using the Black label contained significantly more bias and discrimination terminology than paragraphs containing the term African American (see Table 3, Model 6). Conversely, paragraph segments using the Black label contained significantly less civil rights and equality terminology than paragraphs containing the term African American (see Table 4, Model 6). The year the op-ed was published lost significance for both ideological domains within the period of 1988 to 2013 (see Tables 3 and 4, Model 6) but was significant for a more extended period (see Tables 3 and 4, Models 1 and 2).

Discussion

Our analysis of a longitudinal corpus of op-ed texts found that writers who used the Black label as opposed to the African American label opined more through a lens of bias and discrimination than a lens of civil rights and equality (consistent with the findings of Study 1 and past research); in addition, they wrote more negative emotional content.

Study 3: Predicting an Organization’s Ideological Intent

Studies 1 and 2 examined the valence and meanings associated with the Black and African American labels. Study 3 explored how these associations affect White perceivers’ perceptions of differently labeled AAD groups and whether perceivers’ ideological values influence which group they would support. We predicted that White perceivers would assume that an organization with Black in its title is more aligned with a bias and discrimination ideology and that an organization with African American in its title is aligned more with a civil rights and equality ideology. Further, we hypothesized that White perceivers who prioritize bias and discrimination would report being more inclined to support an organization that uses the Black label, whereas White perceivers who prioritize civil rights and equality would report being more inclined to support an organization that uses the African American label.

Method

Study 3 was preregistered on OSF (https://osf.io/kxjd9/). Because of the data structure, we opted to change our analytical technique. We created an amendment document with the highlighted changes that we needed to make (see https://osf.io/nc57d/).

Participants

We recruited 503 White adult participants from the Prolific Survey platform in July 2020. Following our preregistration, we omitted participants who failed to answer an attention check correctly (23.5%). Our final sample consisted of 385 White American participants (178 men, 197 women, seven nonbinary or other, three preferred not to answer; age: M = 34.71 years, SD = 12.35). We sought to recruit a sample size sufficient to detect a small to medium effect size (Cohen’s d = 0.2–0.5). A sensitivity power analysis (1 – β = 0.95, α = .05, two tailed) indicated that our sample size of 385 would allow us to detect a Cohen’s d effect size of 0.19 with a probability of .95 (Faul et al., 2009).

Procedure

Participants completed a survey on which they guessed the intentions of three minority organizations. As a cover story, we suggested that many people can accurately guess the platform and political stance of organizations using only the organization’s name. Then, we instructed participants to guess the ideological platform, goals, idolized historical figure, and preferred charity of three minority organizations.

We listed three fictitious organizations that used the racial labels Black, African American, or people of color, and we coupled these labels with the monikers “alliance,” “coalition,” or “union” (e.g., The Black Alliance, The African American Coalition). We randomly alternated monikers between racial labels for every participant (e.g., The Black Alliance vs. The Black Union). Following our preregistration, we added the people-of-color condition primarily as a filler group but also for use in exploratory analyses (for further information, see the Supplemental Material).

For each guessing question, we instructed participants to review three selections and match each selection to the organization that it best epitomized (see Table 6). We presented the options in random order. We predicted that participants would sort selections related to bias and discrimination under the organization donning the Black label. In contrast, we predicted that participants would sort selections related to civil rights and equality under the organization donning the African American label. Following our preregistration, we included the “diversity and culture” issue as a filler and made no a priori predictions regarding it.

Table 6.

Typology of Ideological Associations (Studies 3 and 4)

Question	Question prompt	Bias and discrimination	Civil rights and equality	Diversity and culture
Question 1: ideological platform	“Guess the ideology and platform that each group most likely endorses.”	Eradicating bias, discrimination, and racial injustice	Promoting civil rights and equal opportunity	Celebrating customs and culture within schools and other institutions
Question 2: goal	“Guess the goal that each group is aiming for.”	To redirect the budget from the police to community support services	To stop voter suppression and redlining	To recognize a wider diversity of ethnic holidays in schools
Question 3: historical figure	“Guess the historical figure that each group idolizes.”	Malcolm X	Martin Luther King, Jr.	Nelson Mandela
Question 4: charity	“Guess the charity that each group donates to.”	Anti-Racism and Bias Alliance	National Association for the Advancement of Colored People	Multicultural Unity Association

Finally, we measured participants’ likely financial support for the groups. Specifically, participants read and responded to the following: “Based on your estimations, if you had to financially support one of these groups with a donation, which one would it be?” After participants completed demographic questions, we tasked them with rank ordering each issue by importance, for example, “Please rank order each of these issues according to their importance to you,” ranging from 1 (most important) to 3 (least important).

Results

We analyzed the data using Wilcoxon signed-rank tests to detect differences in the sorting patterns between Black and African American organizations. Following our preregistration, we do not present responses related to people of color or our filler “diversity and culture” theme, but we include them in the Supplemental Material. Over all four questions, participants were more likely to sort selections related to bias and discrimination under Black than African American organizations.

Ideological platform

The majority of participants estimated that the Black organization’s ideological platform was related to eradicating bias and discrimination (57.3%) rather than civil rights and equality (28.8%). However, they believed that the African American organization’s ideological platform was related to civil rights and equality (41.0%) rather than eradicating bias and discrimination (17.4%; z = −9.41, p < .001, r = .41).

Goal

The majority of participants estimated that the Black organization’s goal was to defund the police (55.1%) rather than to stop voter suppression (35.6%); however, they believed that the African American organization’s goal was to stop voter suppression (38.2%) rather than to defund the police (27.3%; z = −7.74, p < .001, r = .33).

Historical figure

The majority of participants estimated that the Black organization idolized Malcolm X (75.6%) rather than Martin Luther King, Jr. (12.5%); however, they believed that the African American organization idolized Martin Luther King, Jr. (55.1%) rather than Malcolm X (12.7%; z = −11.28, p < .001, r = .49).

Charity

The majority of participants estimated that the Black organization would donate to the Anti-Racism and Bias Alliance (58.2%) rather than the National Association for the Advancement of Colored People (NAACP; 33.0%); however, they believed that the African American organization would donate to the NAACP (50.1%) rather than the Anti-Racism and Bias Alliance (28.1%; z = −6.80, p < .001, r = .30).

Choice of organization to financially support

Finally, we predicted that the issue that a participant declared as most important would guide their choice of which organization to support financially, but our results did not reach significance, Wald χ²(1, N = 385) = 3.11, p = .078, exp(b) = 1.70, 95% confidence interval (CI) = [0.942, 3.067]. Nonetheless, the pattern and direction are consistent with our hypothesizing: Participants who ranked bias and discrimination over civil rights and equality as the most important issue were 70% more likely to express an intent to support an organization with the Black racial label than the African American racial label. Conversely, participants who ranked civil rights and equality as the most important issue were 41.2% less likely to choose to support a Black than an African American organization.

Discussion

Participants were more likely to associate organizations labeled as Black with characteristics related to bias and discrimination and organizations labeled as African American with characteristics related to civil rights and equality. However, participants’ preferred ideologies were not significantly associated with their intent to support a Black over an African American organization, although the direction of difference was consistent with our prediction.

Study 4: Donation Behaviors as a Function of Racial Label

In Study 4, we included a stronger measure of participant support for AAD organizations: a behavioral measure in which participants could donate a portion of a $1.00 bonus to one of three 501(c)(3) charities committed to racial progress. We predicted that participants who endorsed bias and discrimination issues would likely donate a higher amount to a Black organization and that those who endorsed civil rights and equality issues would likely donate more to an African American organization.

Method

Study 4 was preregistered on OSF (https://osf.io/tf96w/).

Participants

Given that we lost approximately 25% of our sample in Study 3 because of a failed attention check, we overrecruited by approximately 25% in the current study and sought to replicate the associations between issues and racial labels that we found in Study 3. Consequently, we recruited 747 White adult American participants from the Prolific Survey platform in July 2020. Following our preregistration for Study 4, we omitted participants who failed to answer an attention check correctly (24%) and participants who were familiar with at least one of the organizations in question before rendering their financial support (11.4%). Our final sample consisted of 527 White adult participants (242 men, 276 women, nine nonbinary; age: M = 37.78 years, SD = 13.89). A sensitivity power analysis (1 – β = 0.95, α = .05, two tailed) indicated that our sample size of 527 would allow us to detect a Cohen’s d effect size of 0.16 with a probability of .95 (Faul et al., 2009).

Procedure

The materials and procedure in Study 4 were identical to those in Study 3; however, we substituted fictitious names for our three real-life 501(c)(3) charities. We initially chose this set of charities because they each had a different racial label (Black, African American, or person of color). However, we randomized the names with their ending monikers such that we randomly presented each organization with its actual label (e.g., The Black Alliance) or another label (e.g., The African American Alliance, The People of Color Alliance) for any given participant.

Additionally, we added a bonus and donation request. We granted participants a $1.00 bonus at the end of the survey and allowed them to donate 0% to 100% of their bonus to any one of the three charities they evaluated in the survey. After the study, we remitted participants’ donations to each of the three real-life organizations.

Results

Overall, we replicated the significant effects that we found in Study 3 in that participants were more likely to sort selections related to bias and discrimination under Black organizations and selections related to civil rights and equality under African American organizations.

Ideological platform

The majority of participants estimated that the Black organization’s ideological platform was related to eradicating bias and discrimination (51.5%) rather than civil rights and equality (29.1%). However, they believed that the African American organization’s ideological platform was related to civil rights and equality (35.9%) rather than eradicating bias and discrimination (23.8%; z = −7.85, p < .001, r = .34).

Goal

The majority of participants estimated that the Black organization’s goal was to defund the police (51.3%) rather than to stop voter suppression (32.5%); however, they believed that the African American organization’s goal was to stop voter suppression (39.0%) rather than to defund the police (27.9%; z = −6.88, p < .001, r = −.30).

Historical figure

The majority of participants estimated that the Black organization idolized Malcolm X (72.2%) rather than Martin Luther King, Jr. (14.4%); however, they believed that the African American organization idolized Martin Luther King, Jr. (51.3%) rather than Malcolm X (13.5%; z = −12.69, p < .001, r = .55).

Charity

The majority of participants estimated that the Black organization would donate to the Anti-Racism and Bias Alliance (56.2%) rather than the NAACP (30.4%); however, they believed that the African American organization would donate to the NAACP (45.8%) rather than the Anti-Racism and Bias Alliance (27.6%; z = −7.25, p < .001, r = .32).

Choice of organization to financially support

We predicted that the ideology that a participant favors would guide their preference for a Black than an African American organization. We found that participants who ranked bias and discrimination as their most important issue were 99% more likely to choose a Black than an African American organization to financially support, whereas participants who chose civil rights and equality were 50% less likely to choose a Black than an African American organization, Wald χ²(1, N = 527) = 7.02, p = .008, exp(b) = 1.99, 95% CI = [1.197, 3.322].

Donation amount

Finally, we predicted that an alignment between a participant’s most important issue and the racial label applied to their organization of choice would correspond to a higher donation amount. Consistent with this, results showed that participants who ranked bias and discrimination as their most important issue donated significantly more to a Black organization (M = $0.53, SD = $0.40) than to an African American organization (M = $0.41, SD = $0.35), t(159) = 2.13, p = .04, 95% CI for the mean difference = [−0.2434, −0.0092], d = 0.34. However, participants who ranked civil rights as their most important issue did not donate significantly more to an African American organization (M = $0.29, SD = $0.36) than to a Black organization (M = $0.38, SD = $0.36), t(102) = 1.25, p = .21, 95% CI for the mean difference = [−0.2422, 0.0548], d = 0.26.

In general, participants who ranked bias and discrimination as the most important issue gave a more substantial donation (M = $0.43, SD = $0.38) than those who ranked civil rights and equality as the most important (M = $0.30, SD = $0.33), t(428) = 3.80, p < .001, 95% CI for the mean difference = [−0.2031, −0.0647], d = 0.37. In addition, participants who selected the Black organization gave a more substantial donation (M = $0.47, SD = $0.40) than those who selected the African American organization (M = $0.33, SD = $0.35), t(306) = 3.29, p = .001, 95% CI for the mean difference = [−0.2248, −0.0565], d = 0.38.

Discussion

Consistent with Studies 1 to 3 and our predictions, results showed that participants associated AAD racial labels with the ideologies of important historical movements. Further, participants’ ideologies guided their actual support for organizations. In partial support of our prediction, results showed that White people who prioritized bias and discrimination donated more money to a Black than an African American organization. However, those who prioritized civil rights did not differ in the amount of money donated to an African American than a Black organization. Although Whites who endorse civil rights and equality believe that an African American organization best reflects their values, they may lack the passion necessary for offering fiscal support.

General Discussion

We argue that the two most commonly used AAD labels, African American and Black, retained the valence and meanings associated with the ideologies of the historical movements within which they gained prominence. Using rich archival data and experimental methods, we found that “Black” is associated with a bias and discrimination ideology and “African American” is associated with a civil rights and equality ideology. The ideologies embedded in these labels skewed Google Images search results (Study 1) and the content of op-eds (Study 2). Additionally, whether an organization’s name included the African American or Black labels shaped White participants’ assumptions about the organization’s ideology (e.g., platform, goals; Studies 3 and 4) and their financial support of the group (Study 4). Specifically, White participants who prioritized bias and discrimination donated more money to an organization with the Black label than the African American label.

Theoretically, the current research contributes to the stereotype-content model, which suggests that AADs are perceived as members of two distinct subtypes (i.e., poor AADs and AAD professionals; Fiske et al., 2002). We speculate that the content of the poor-AAD subtype may consist of traits that align with the Black label as well as its ideology and originating historical movement. Likewise, the content of the professional-AAD subtype may consist of traits that align with the African American label and its ideology. The current research underscores the importance of avoiding generalizations about how AADs are perceived and begins to build new theory regarding common routes through which AAD subtypes are triggered in cognition (i.e., the use of racial labels).

Our results also suggest practical, and deleterious, downstream consequences. In particular, the content associated with the racial labels may lead to inaccurate perceptions and polarization even when broad consensus exists. For example, we found that Whites are more likely to associate defunding the police with Black as opposed to African American activists. If an editorial uses the Black label to describe a group of AAD protesters, it could unintentionally (or intentionally) lead White readers to believe that the protestors want to defund the police. Importantly, this may further stymie support for the protestors among White voters who are concerned with law and order (e.g., Baker & Haberman, 2020) and may occur even though only 19% of AADs indicate that they want less police presence (Grzeszczak, 2020).

Our studies represent a snapshot of a particular time, and future work should explore how the meaning of these words continues to be altered as race-based events arise and garner discussion. In addition, we expect our effects to generalize only to American populations who are not of African descent. Specifically, AADs may understand the nuance of racial-label choice and, therefore, be less likely to apply broad-brush ideological stereotypes to AAD targets who are labeled by these terms.

The movement for racial progress continues, and so does the media’s coverage of it. The seemingly small linguistic choices that activists and journalists make can have considerable consequences. For example, the use of the Black label in Black Lives Matter may have connected the movement to the ideology of the Black Power Movement and increased the perception that Black Lives Matter is focused exclusively on racial bias and victimization. Although they may not be consciously aware of the ideologies attached to racial labels, activists and journalists must choose whether to use the labels Black or African American, and these choices carry real consequences for how the public interprets or misinterprets activists’ intentions and journalists’ articles. As society strives toward the goal of racial progress, it is critical to understand how these linguistic decisions can influence the public’s perceptions and divert its attention from this ultimate goal.

Supplemental Material

sj-docx-1-pss-10.1177_09567976211018435 – Supplemental material for What’s in a Name? The Hidden Historical Ideologies Embedded in the Black and African American Racial Labels

Supplemental material, sj-docx-1-pss-10.1177_09567976211018435 for What’s in a Name? The Hidden Historical Ideologies Embedded in the Black and African American Racial Labels by Erika V. Hall, Sarah S. M. Townsend and James T. Carter in Psychological Science

Footnotes

Acknowledgements

We thank Adam Galinsky, Katherine Phillips, and Kareem Hall for helpful feedback on earlier versions of this article.

Transparency

Action Editor: Kate Ratliff

Editor: Patricia J. Bauer

Author Contributions

E. V. Hall developed the overall concept for the studies. All the authors developed the study designs. E. V. Hall and J. T. Carter analyzed the data. All the authors interpreted the data. E. V. Hall and S. S. M. Townsend drafted the manuscript, and all the authors revised the manuscript. All the authors approved the final manuscript for submission.

ORCID iD

James T. Carter

Supplemental Material

Additional supporting information can be found at

Notes

References

Aiken

J. R.

Salmon

E. D.

Hanges

P. J.

(2013). The origins and legacy of the Civil Rights Act of 1964. Journal of Business and Psychology, 28(4), 383–399. https://doi.org/10.1007/s10869-013-9291-z

Baker

Haberman

(2020, June 8). Trump rebuffs protests over systemic racism and calls police ‘great people.’ The New York Times. https://www.nytimes.com/2020/06/08/us/politics/defund-police-trump.html?

Bauder

(2020, July 20). AP says it will capitalize Black but not white. AP News. https://apnews.com/7e36c00c5af0436abc09e051261fff1f

Benjamin

(2019). Race after technology: Abolitionist tools for the New Jim Code. Polity.

Faul

Erdfelder

Buchner

Lang

A.-G.

(2009). Statistical power analyses using G*Power 3.1: Tests for correlation and regression analyses. Behavior Research Methods, 41(4), 1149–1160.

Faulkner

(1951). Requiem for a nun. Random House.

Fiske

S. T.

Cuddy

A. J. C.

Glick

(2002). A model of (often mixed) stereotype content: Competence and warmth respectively follow from perceived status and competition. Journal of Personality and Social Psychology, 82(6), 878–902.

Grzeszczak

(2020, August 5). 81% of Black Americans don’t want less police presence despite protests–some want more cops: Poll. Newsweek. https://www.newsweek.com/81-black-americans-dont-want-less-police-presence-despite-protestssome-want-more-cops-poll-1523093

Hall

E. V.

Phillips

K. W.

Townsend

S. S. M.

(2015). A rose by any other name?: The consequences of subtyping “African-Americans” from “Blacks.” Journal of Experimental Social Psychology, 56, 183–190. https://doi.org/10.1016/j.jesp.2014.10.004

10.

Hauser

D. J.

Schwarz

(2018). How seemingly innocuous words can bias judgment: Semantic prosody and impression formation. Journal of Experimental Social Psychology, 75, 11–18. https://doi.org/10.1016/j.jesp.2017.10.012

11.

Johnson

(2007). What is organizational imprinting? Cultural entrepreneurship in the founding of the Paris Opera. American Journal of Sociology, 113(1), 97–127. https://doi.org/10.1086/517899

12.

Joseph

P. E.

(2009). The Black Power movement: A state of the field. Journal of American History, 96(3), 751–776. https://doi.org/10.1093/jahist/96.3.751

13.

Lawrence

S. R.

(2005). Personalization of web search (U.S. Patent No. 10,877,775). U.S. Patent and Trademark Office. https://patentimages.storage.googleapis.com/61/7f/46/3a8ae4d48b382f/US20050071328A1.pdf

14.

Lorenz

(1935). Der Kumpan in der Umwelt des Vogels. Der Artgenosse als auslösendes Moment sozialer Verhaltungsweisen [The companion in the bird’s world. The fellow-member of the species as releasing factor of social behavior]. Journal Für Ornithologie. Beiblatt, 83, 137–213. https://doi.org/10.1007/BF01905355

15.

Marquis

Tilcsik

(2013). Imprinting: Toward a multilevel theory. Academy of Management Annals, 7(1), 195–245. https://doi.org/10.5465/19416520.2013.766076

16.

Martin

B. L.

(1991). From Negro to Black to African-Americans: The power of names and naming. Political Science Quarterly, 106(1), 83–107.

17.

Noble

S. U.

(2018). Algorithms of oppression: How search engines reinforce racism. New York University Press.

18.

The OpEd Project (2015). Submission information: Outlets 1-50. https://web.archive.org/web/20150318084629/http://theopedproject.org/index.php?option=com_content&view=article&id=47&Itemid=65

19.

Pennebaker

J. W.

Boyd

R. L.

Jordan

Blackburn

(2015). The development and psychometric properties of LIWC2015. University of Texas at Austin.

20.

Pew Research Center. (2018, July 10). Use of the #BlackLivesMatter hashtag on Twitter periodically spikes in response to major news events. Pew Research Center: Internet, Science & Tech. https://www.pewresearch.org/internet/2018/07/11/an-analysis-of-blacklivesmatter-and-other-twitter-hashtags-related-to-political-or-social-issues/pi_2018-07-10_social-activism_0-08/

21.

Pitkow

J. E.

Schütze

Cass

T. A.

Cooley

R. W.

Turnbull

Edmonds

Adar

Breuel

T. M.

(2002). Personalized search. Communications of the ACM, 45(9), 50–55. https://doi.org/10.1145/567498.567526

22.

Sinclair

(1991). Corpus, concordance, and collocation. Oxford University Press.

23.

Stinchcombe

A. L.

(1965). Social structures and organizations. In March

J. P.

(Ed.), Handbook of organizations (pp. 142–193). Rand McNally.

24.

Valocchi

(1996). The emergence of the integrationist ideology in the Civil Rights Movement. Social Problems, 43(1), 116–130. https://doi.org/10.1525/sp.1996.43.1.03x0339a

25.

Van Horne

W. A.

(2007). The concept of Black Power: Its continued relevance. Journal of Black Studies, 37(3), 365–389. https://doi.org/10.1177/0021934706290079

26.

WashPostPR. (2020, July 29). The Washington Post announces writing style changes for racial and ethnic identifiers. The Washington Post. https://www.washingtonpost.com/pr/2020/07/29/washington-post-announces-writing-style-changes-racial-ethnic-identifiers/

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.09 MB