This is an open access article distributed under the terms of the Creative Commons
Attribution License (In the Netherlands, the increase in ethnic diversity has been accompanied by growing awareness about aspects of Dutch culture that are insensitive towards people of color (such as racial slurs in baked goods; Deems, 2020, or referring to the 17th century as the Dutch ‘Golden Age’ — a time when the Dutch were a major colonial power and player in the Transatlantic slave trade; Kiers et al., 2019). The public discourse around changing insensitive Dutch traditions tends to be met with a lot of pushback. Many of these discussions end up touching on bigger questions regarding the Dutch national identity (The Netherlands Institute for Social Research, 2019) and the dominant position of the White majority group (Gargard, 2020; Rodenberg & Wagenaar, 2016). The most controversial of these traditions has been the popular Black Pete figure that is part of the Sinterklaas festivities. The figure represents a racial caricature of Black people and is typically portrayed by White volunteers in blackface (e.g., black or brown face paint, red lips, golden hoop earrings, and afro wigs).
Regardless of the controversies surrounding the Black Pete figure and the Sinterklaas festivities, most ethnic Dutch people still hold positive attitudes towards the figure. For example, most Dutch people deny that the Black Pete figure can be considered racist and prefer not to change the figure’s appearance (Klapwijk, 2019; Lubbe, 2020). Research suggests that part of this resistance stems from people strongly identifying as Dutch (Tjew-A-Sin & Koole, 2018b). However, public discourse suggests that for many ethnic Dutch people, the resistance to changing Black Pete stems from wanting to maintain cultural dominance (Cankaya, 2016). With that in mind, we will investigate the following research questions: Can subgroups of people be differentiated whose attitudes on the Sinterklaas festivities and Black Pete cluster with either a) high Dutch national identification or b) a strong preference for social hierarchy, also known as social dominance orientation (SDO; Pratto et al., 1994)? Additionally, are attitudes on the Sinterklaas festivities and Black Pete more strongly associated with Dutch national identification or SDO?
Sinterklaas Festivities and the Black Pete Figure
Sinterklaas festivities in Western Europe date back to the late Middle Ages. Sinterklaas is a figure in Dutch culture that resembles Santa Claus — an old man in a red costume with a white beard who brings gifts to children if they have been good or punishes them if they have been naughty. During the festivities in the late Middle Ages, Sinterklaas was often portrayed as being alone but at times accompanied by a devilish helper (Huismans & Wayenberg, n.d.-b). The current version of the holiday, as celebrated in the Netherlands and the Dutch-speaking parts of Belgium, can be traced back to the 1850s (Huismans & Wayenberg, n.d.-a). In this telling of the story, Sinterklaas and his servant Black Pete arrive yearly by steamboat from Spain to bring children gifts on the 5th of December. In the first half of the 1900s, Sinterklaas was typically accompanied by a single Black Pete servant. Over the years, however, increasing numbers of Black Pete figures accompanied Sinterklaas (Huismans & Wayenberg, n.d.-a). The controversy of the Black Pete figure stems from his blackface style appearance. According to Dutch folklore, Black Pete has dark skin because of going down the chimney to bring gifts to children — no explicit justification is made for the rest of the figure’s racially insensitive appearance. In addition, he has a subservient role, serving Sinterklaas by making sure all the presents are in order and delivered to the children.
Because of Black Pete’s subservient role and racially stereotypical looks, many people see the figure as a form of racism that fosters discrimination against Black people in the Netherlands. This sentiment was famously captured on Dutch national television in 1987 when during a broadcast of the Dutch version of the children’s show Sesame Street (VPRO, 2014), the Black Surinamese-Dutch actress Gerda Havertong explained that “Sinterklaas is not even in the country yet and Black people, adults and children, are being insultingly called Black Pete… For many Black people, adults and children, this is not a festivity at all”. Recent polls back up her claim; a majority of the Surinamese-Dutch people (60%) and half of the Antillean-Dutch people (49%), many of whom are Black, have experienced prejudice linked to the Black Pete figure (Klapwijk, 2019).
Nevertheless, calls to change racially insensitive aspects of the Black Pete figure have been met with substantial resistance by many ethnic Dutch people. These calls to change the figure are not new — they date all the way back to the 1930s. However, it is only in recent years that calls for changing Black Pete have become a yearly recurring part of the public debate leading up to the Sinterklaas festivity (Heilbron et al., 2018). Since 2011, protests against the Black Pete figure have become increasingly polarized. On the one hand, large crowds of protestors from various ethnic backgrounds want to change the figure’s appearance. However, on the other hand, increasingly large crowds of predominantly White Dutch protestors do not want to change the figure’s appearance. This resistance is also reflected in national polls; the majority (55%) of Dutch people prefer not to change the appearance of Black Pete, and an even larger proportion (78%) do not see the figure as being a racist phenomenon (Lubbe, 2020). This resistance tends to be justified in a number of ways. Many emphasize that the Sinterklaas festivities are an important part of Dutch culture (Strouken, 2010) and that it is an innocent children’s holiday (Cankaya, 2016). In contrast, most Surinamese- and Antillean-Dutch people (69% and 58% respectively) prefer to change the appearance of Black Pete and see the character as racist (Surinamese-Dutch: 65%, Antillean-Dutch: 53%; Klapwijk, 2019).
Identity and SDO
Research suggests that resistance to changing Black Pete is strongest among those who strongly identify as Dutch (Tjew-A-Sin & Koole, 2018b). Based on social identity theory (Cameron, 2004; Hogg, 2021; Tajfel & Turner, 1979), one could argue that people with high national identification in the Netherlands hold on to the traditional Sinterklaas festivities and the Black Pete figure because it is a) considered a core aspect of Dutch identity (the festivities are considered the most important Dutch holiday; Strouken, 2010) and b) changing core aspects of the festivities would have a negative effect on people’s (identity-derived) self-esteem. However, the public discourse on changing Black Pete suggests an important role for the notion that the majority group has to change traditions because members of a marginalized ethnic group are offended by it (Cankaya, 2016; Gargard, 2020). This suggests that national identification may not be the only psychological factor at play because national identification is not consistently associated with negative attitudes towards marginalized ethnic groups (Anderson & Ferguson, 2018; Thomsen et al., 2008). Some people with high national identification may consider marginalized ethnic groups with the same nationality part of the ingroup (Anderson & Ferguson, 2018). Therefore, it is likely that other psychological constructs may help explain why resistance to changing Black Pete is so common in the Netherlands.
Among right-wing Dutch politicians in the Netherlands, it is common to refer to changing the appearance of Black Pete as a symbol of losing Dutch culture (Vliet, 2020). This sentiment is also common among the general public. For example, a popular pro-Black Pete Facebook group (which was later deleted) with over 18,000 followers (as of March 2020) included the following post: “you came to the NETHERLANDS…I will not change my flag, religion or culture or habits for you. If you do not like what the Netherlands is and what it stands for, you are FREE TO LEAVE” (Ik ben Zwarte Piet, 2019). Clearly, the public sentiment regarding the Sinterklaas festivities and the Black Pete figure suggests that more is at stake for many ethnic Dutch people than lowering one’s (identity-derived) self-esteem and that there is fear about losing their dominant position in society. A preference for social hierarchy, or social dominance orientation (SDO; Pratto et al., 1994), may play a role besides national identity.
This is in line with social dominance theory, which asserts that high SDO people are opposed to hierarchy-attenuating changes, or changes that make relations between dominant and less dominant groups more equal and less hierarchal (Sidanius & Pratto, 1999). Similar to members of the dominant ethnic group who feel strongly attached to their nationality, having high SDO may help members of the dominant ethnic group justify their group's position in comparison to other groups (Guimond et al., 2003; Morrison & Ybarra, 2008). Conceptually, opposing changes that increase equality between groups implies more resistance to making a cultural tradition less stigmatizing towards a marginalized ethnic group. This is part of the debate regarding the offensiveness of the Black Pete figure (Wieringa, 2019). Research provides support for this assumption in various ways. For instance, SDO is associated with resistance to having an intercultural dialogue with an ethnic outgroup member regarding racism (Cargile, 2017), unwillingness to apologize to ethnic outgroup members for harm suffered during colonial times (Mifune et al., 2019), a lower tendency to classify a potentially prejudiced situation as being racist (Miller & Saucier, 2018), tolerance of racism (Hunt et al., 2021), and a greater likelihood of framing social progress made by marginalized ethnic groups as a ‘loss’ for the dominant ethnic group (Eibach & Keegan, 2006). Additionally, although national identification and SDO are associated with one another, SDO is consistently linked with negative attitudes towards marginalized groups cross-culturally, while national identification is not (Anderson & Ferguson, 2018; Pratto et al., 2013; Thomsen et al., 2008). Taking these findings into consideration, the fact that a large subgroup of the ethnic Dutch population still holds positive attitudes towards Black Pete may be due to high SDO, given that the debate touches on topics such as the unwillingness of many White Dutch people to even discuss changing the figure, dismissal of the notion that the figure is a racist phenomenon, and the sentiment that the White Dutch are losing their dominant position because ‘newcomers’ are demanding changes to cultural traditions.
In sum, many ethnic Dutch people resist changing racially insensitive aspects of Dutch culture, especially the figure of Black Pete. Considering the debate regarding Black Pete and the literature, we expect national identity and SDO to play a role in driving this resistance. We plan to use both a person-centered and variable-centered approach to test our hypotheses. The person-centered approach is used to identify if subgroups of people exist in which positive attitudes towards the Sinterklaas festivities and the Black Pete figure cluster with identity concerns or preference for social hierarchy (Howard & Hoffman, 2018). Variable-centered analyses on the other hand provide broad general patterns between variables within a sample (e.g., is variable A related to variable B?). For a person-centered approach, we hypothesize that people with positive attitudes towards the Black Pete figure and the Sinterklaas festivities can be differentiated into at least two distinct groups: a subgroup characterized by high national identification and a subgroup characterized by high SDO. Secondly, we hypothesize that subgroups with high SDO people will also be characterized by having the strongest positive attitudes towards the Black Pete figure, while subgroups with high national identification will be characterized by having the strongest positive attitudes towards the Sinterklaas festivities in general. Lastly, considering that men tend to have higher SDO than women (Pratto et al., 1994; Sidanius et al., 2000), we will also examine if subgroups with high SDO have a higher proportion of men than low SDO subgroups. For the variable-centered approach we hypothesize that national identification will be more strongly related to positive attitudes towards Sinterklaas than SDO, and that SDO will be more strongly related to positive attitudes towards the Black Pete figure than national identification.
Method
Participants and Procedure
The current study uses a secondary dataset made publicly available through the Data Archiving and Network service repository (Tjew-A-Sin & Koole, 2018a). The original experimental study examined the influence of mortality salience on attitudes regarding Sinterklaas and Black Pete (Tjew-A-Sin & Koole, 2018b), and the data were collected in the last quarter of 2013. The sample consists of 174 ethnically Dutch university students (139 women; Mage = 19.87, SD = 3.39). Participants received monetary compensation or partial course credit. First, the participants filled out measures of mood. Next, the participants were randomly assigned to either the high (n = 83) or low (n = 91) mortality salience condition. In the high mortality salience condition, participants filled out a brief 7-item version of the Fear of Personal Death Scale (FPDS; Florian & Kravetz, 1983), with higher scores representing a higher fear of death. In the low mortality salience condition, participants filled out similar items on fear of going to the dentist. Subsequently, participants filled out a second questionnaire on mood, followed by questionnaires on national identification, self-esteem, SDO, self-stereotypes referring to the Dutch, Theory of Mind, Dutch caricatures, and general background information. See Tjew-A-Sin and Koole (2018b) for more information on measures not included in the current study.
Measures
National Identity
The degree to which a person feels a sense of belonging to their Dutch identity was assessed with three items developed by Verkuyten (2005) that were inspired by the Identity and Membership subscales of the Collective Self-Esteem Scale (Luhtanen & Crocker, 1992). The three items (“I feel connected to the Netherlands,” “I identify with Dutch people,” and “I feel like a Dutchman”) were rated on a scale ranging from 1 (strongly disagree) to 7 (strongly agree). The three items were averaged. The internal consistency of the scale was good (Cronbach’s α = .84).
Social Dominance Orientation
The degree of preference for social hierarchy between groups was measured with the four-item Short Social Dominance Orientation (SSDO; Pratto et al., 2013). The four items (“In setting priorities, we must consider all groups”, “We should not push for group equality”, “Group equality should be our ideal” and “Superior groups should dominate inferior groups”) were rated on a scale ranging from 1 (disagree strongly) to 7 (strongly agree). After reverse-scoring two items, the items were averaged. The internal consistency of the scale was acceptable (Cronbach’s α = .66).
Attitudes Regarding Sinterklaas and Black Pete
Attitudes regarding the traditional celebration of Sinterklaas and the controversial Black Pete figure were measured with an eight-item questionnaire. The items were rated on a scale from 1 (do not agree at all) to 7 (completely agree). A principal component analysis (with a varimax rotation constrained to two factors) was used to examine if the items could be split into reliable subscales that capture attitudes towards a) Sinterklaas and the cultural importance of the festivities and b) the controversy surrounding the Black Pete figure. Four items measuring attitudes towards Sinterklaas in relation to Dutch culture (i.e., “Celebrating Sinterklaas is part of the Netherlands”, “The Sinterklaas festivities and Black Pete are an important part of Dutch tradition”, “Only the Dutch understand the value of the Sinterklaas tradition”, and “Sinterklaas is a harmless activity for children”) loaded on the first factor and had factor loadings > .60. These items were averaged to form the subscale “Positive attitudes towards Sinterklaas”, and showed acceptable internal consistency (Cronbach’s α = .66). Four items measuring attitudes towards Black Pete (i.e., “Black Pete should remain a part of the Sinterklaas tradition”, and the reverse-coded “Black Pete is racist”, “The appearance of Black Pete should be adjusted”, and “Black Pete is really not done in today’s society”) loaded on the second factor and had factor loadings > .70. These items were averaged to form the subscale “Positive attitudes towards Black Pete” with good internal consistency (Cronbach’s α = .88).
Results
Data-Inspection and Preliminary Analyses
The missing value analysis found no missing data points. Standardized z-scores greater than 3.29 above and below the mean were defined as outliers (Kim, 2013). Three univariate outliers on the SDO, national identification, and positive attitudes towards Sinterklaas variables were winsorized. The winsorized variables were normally distributed. Additionally, two multivariate outliers were detected. Due to these multivariate outliers, we used K-median cluster analyses which are less sensitive to the influence of outliers than the popular K-means method (Cardot et al., 2012). We excluded these multivariate outliers from our multiple regression analyses. We also tested if there were mean differences between experimental conditions for the SDO, national identification, positive attitudes towards Sinterklaas and positive attitudes towards Black Pete variables. No significant differences were found (p-values ranged from .44 to .89). See Table 1 for descriptive statistics.
Table 1
Means, Standard Deviations, and Correlations Between National Identification, SDO, Positive Attitudes Towards Sinterklaas, and Positive Attitudes Towards Black Pete
| Variable | M | SD | 1 | 2 | 3 | 4 | 5 |
|---|---|---|---|---|---|---|---|
| 1. National Identification | 5.39 | 1.26 | — | ||||
| 2. SDO | 2.62 | 0.96 | -.13 | — | |||
| 3. Positive attitudes towards Sinterklaas | 5.62 | 0.85 | .25** | .12 | — | ||
| 4. Positive attitudes towards Black Pete | 5.69 | 1.39 | .17* | .23** | .56** | — | |
| 5. Gender | — | — | -.12 | .15 | .08 | .04 | — |
Note. M and SD represent the mean and standard deviation. Gender was coded 0 for female, and 1 for male.
*p < .05. **p < .01. ***p < .001.
The NbClust package in R recommended the number of clusters for the current dataset according to 30 popular indices (Charrad et al., 2014). Manhattan distance was used as the dissimilarity measure. A two-cluster solution was supported by eight indices, a three-cluster solution was supported by five indices, and a four-cluster solution was supported by four indices. All other solutions were supported by three or fewer indices. The two-cluster solution formed clusters that scored either high or low (above or below the mean) on all variables. The four-cluster solution formed two clusters that scored either high or low on all variables, a third cluster that scored low on all variables except for SDO, and a fourth cluster that scored high on all variables except for SDO. Because the goal of cluster analyses is to find clusters in the data that a) reflect natural groupings in the data that b) are theoretically interpretable (Greene & Cunningham, 2005), we proceeded with a K-median cluster analysis of the three-cluster solution (using the Gmedian package in R; Cardot, 2016), which better reflected our expectations based on the national identity and social dominance theory literature.
Main Analyses
Table 2 shows the results of the cluster analysis. To interpret the clusters, we examined the standardized and non-standardized median scores of each cluster on the questionnaires to show both the relative and absolute position of participant clusters on the key variables.
The first cluster of participants (n = 73), labeled the high national identification group, scored above average on positive attitudes towards Sinterklaas, positive attitudes towards Black Pete, and national identification, and below average on SDO. The non-standardized scores on the questionnaires indicated that the high national identification group had positive attitudes towards Sinterklaas (mean close to scale descriptor “agree”), and Black Pete (mean between scale descriptor “agree” and “strongly agree”), identified as Dutch (mean close to scale descriptor “agree”), while scoring relatively low on SDO (mean close to scale descriptor “disagree”). The second cluster of participants (n = 51), labeled the high SDO group, scored above average on positive attitudes towards Sinterklaas, positive attitudes towards Black Pete, and SDO, but below average on national identification. The non-standardized scores on the questionnaires indicated that the high SDO group had positive attitudes towards Sinterklaas (mean close to scale descriptor “agree”) and Black Pete (mean between scale descriptor “agree” and “strongly agree”), somewhat identified as Dutch (mean close to scale descriptor “somewhat agree”), while scoring close to neutral on SDO (mean close to scale descriptor “neutral”). The third cluster of participants (n = 50), labeled the neutral-indifferent group, scored below average on positive attitudes towards Sinterklaas, positive attitudes towards Black Pete, national identification, and SDO. The non-standardized scores on the questionnaires indicated that the neutral-indifferent group had slightly positive attitudes towards Sinterklaas (mean close to scale descriptor “somewhat agree”), neutral attitudes towards Black Pete (mean close to scale descriptor “neutral”), somewhat identified as Dutch (mean close to scale descriptor “agree”) while scoring relatively low on SDO (mean close to scale descriptor “disagree”).
Table 2
Z Scores of the Cluster Centers, Means (M), Standard Deviations (SD), and Between-Group Differences for the Three Clusters
| Variable | Z; M (SD)
|
Between-group differences | ||
|---|---|---|---|---|
| High national identification group (n = 73) | High SDO group (n = 51) |
Neutral-indifferent Group (n = 50) | ||
| Positive attitudes towards Sinterklaas | 0.44; 6.01 (0.58)a | 0.48; 5.96 (0.54)a | -1.14; 4.70 (0.76)b | F(2, 103) = 57.74, p < .001, = .47 |
| Positive attitudes towards Black Pete | 0.47; 6.35 (0.79)a | 0.29; 6.41 (0.67)a | -1.09; 3.98 (1.15)b | F(2, 102) = 92.92, p < .001, = .61 |
| National Identification | 0.38; 6.10 (0.74)a | -0.71; 4.84 (1.46)b | -0.69; 4.90 (1.16)b | F(2, 89) = 31.08, p < .001, = .23 |
| SDO | -0.59; 2.15 (0.64)a | 0.78; 3.60 (0.84)b | -0.30; 2.30 (0.70)a | F(2, 101) = 56.73, p < .001, = .44 |
Note. Different superscript letters refer to significant between-group contrasts.
aGroup means with the superscript “a” are not significantly different from one another.
bGroup means with the superscript “b” are not significantly different from one another.
Next, we tested between-group differences between clusters. Because the positive attitudes towards Black Pete and national identification variables violated the homogeneity of variances assumption and were unequal in size, we used Welch’s ANOVA tests on all variables followed up by Games-Howell post-hoc tests (see Table 2). Firstly, significant between-group differences were found for positive attitudes towards Sinterklaas. The scores of the high national identification group and the high SDO group did not significantly differ on positive attitudes towards Sinterklaas, but both scored significantly higher than the neutral-indifferent group. Secondly, significant differences were also found for positive attitudes towards Black Pete. The scores of the high national identification group and the high SDO group did not significantly differ on positive attitudes towards Black Pete, but both groups scored significantly higher than the neutral-indifferent group. Thirdly, significant between-group differences were found for national identification as well. The high national identification group scored significantly higher on national identification than the high SDO group and the neutral-indifferent group, but the scores of the high SDO group and the neutral-indifferent group did not significantly differ from one another. Lastly, significant between-group differences were found for SDO. The high SDO group scored significantly higher on SDO than the high national identification group and the neutral-indifferent group, but the scores of the high national identification group and the neutral-indifferent group did not significantly differ from one another.
Next, exploratory analyses were performed to examine if the proportion of men in the different clusters was comparable to their proportion in the total sample. Exploratory analyses using binomial tests revealed that men were not significantly over- or underrepresented among the high national identification group (10.0%, p = .112), the high SDO group (30.0%, p = .071), or the neutral-indifferent group (20.0%, p = .071). Similarly, we examined if the experimental conditions had any effect on the likelihood of participants being classified in different clusters. Binomial tests revealed that participants assigned to the mortality salience condition were not over-or underrepresented among the high national identification group (47.0%, p = .64), high SDO group (51%, p = 1.00), or neutral indifferent group (52%, p = .88).
Lastly, we ran two hierarchical multiple regression models. The first model had positive attitudes towards Sinterklaas as the dependent variable. Gender was added in the first block as a covariate; national identification and SDO were added in the second block as predictors. Gender was not significantly associated with positive attitudes towards Sinterklaas in the first (β = .103, p = .179) and second block (β = .111, p = .132). In the second block, higher national identification was significantly associated with stronger positive attitudes towards Sinterklaas (β = .297, p < .001), but SDO was not (β = .106, p = .154). The final model significantly predicted stronger positive attitudes towards Sinterklaas, R2adj = .081, F(3, 170) = 6.11, p = .001.
The second model had positive attitudes towards Black Pete as the dependent variable. Gender was added in the first block as a covariate; national identification and SDO were added in the second block as predictors. Gender was not significantly associated with positive attitudes towards Black Pete in the first (β = .017, p = .826) and second block (β = .004, p = .961). However, in the second block, higher national identification (β = .222, p = .003) and higher SDO (β = .239, p = .002) were significantly associated with stronger positive attitudes towards Black Pete. The final model significantly predicted stronger positive attitudes towards Black Pete, R2adj = .078, F(3, 170) = 5.87, p = .001. To test if these standardized coefficients were significantly different from one another, we calculated their confidence intervals using a bootstrap procedure in SPSS (set to 1000 re-samples). Confidence intervals that overlap by less than 50% can be considered to be significantly different from one another (p < .05; Cumming, 2009). As shown in Figure 1, the confidence intervals mostly overlapped, suggesting no significant difference between the standardized coefficients for national identification and SDO in our second model. This means that national identification and SDO are equally important predictors of positive attitudes towards Black Pete.
Figure 1
Lower and Upper Bounds for the Standardized Regression Coefficients for National Identification and SDO
Conclusions
K-median analyses identified three subgroups. The majority of participants clustered into two subgroups that had strong positive attitudes towards Sinterklaas and Black Pete but differed on national identification and SDO. A minority of participants clustered into a subgroup that seemed rather indifferent (and scored low) on attitudes towards Sinterklaas, Black Pete and scored low on national identification and SDO. Hierarchical multiple regression analyses revealed that national identification, but not SDO, was significantly associated with positive attitudes towards Sinterklaas. In contrast, national identification and SDO were both significantly associated with positive attitudes towards Black Pete.
Inspired by social dominance theory (Sidanius & Pratto, 1999), our first hypothesis was that people with positive attitudes towards the Sinterklaas festivities and the Black Pete figure could be differentiated into at least two distinct groups: a subgroup characterized by high SDO and a subgroup characterized by high national identification. We found support for this hypothesis. Two subgroups of participants were identified with positive attitudes towards Sinterklaas and Black Pete. A high national identification group (42.0% of the participants), characterized by high national identification and low SDO, and a high SDO group (29.3% of the participants), characterized by low national identification and high SDO. The high SDO group scored significantly higher on SDO and significantly lower on national identification than the high national identification group. We also identified a third, neutral-indifferent group (28.7%) that scored low on positive attitudes towards Sinterklaas, Black Pete, national identity, and SDO. We discuss the interpretation of these findings alongside the results of our between-group differences tests in the next paragraph.
Secondly, we hypothesized that subgroups with high SDO would exhibit the strongest positive attitudes about the Black Pete figure, while subgroups with high national identification would exhibit the strongest positive attitudes about the Sinterklaas festivities. We found partial support for this hypothesis. The high national identification group and high SDO group scored significantly higher on positive attitudes towards Sinterklaas and Black Pete than the neutral-indifferent group. However, the high national identification group and high SDO group did not significantly differ from one another on positive attitudes towards Sinterklaas and Black Pete. These findings suggest that the strong resistance to changing Black Pete and related attitudes about the Sinterklaas festivities among the high national identification group seems to reflect a strong identification with Dutch culture. On the other hand, among the high SDO group, these attitudes seem to represent a longing for cultural dominance, as reflected by their positive attitudes towards Black Pete and Sinterklaas clustering with high SDO.
Lastly, we hypothesized that national identification would be more strongly related to positive attitudes towards Sinterklaas than SDO, and that SDO would be more strongly related to positive attitudes towards the Black Pete figure than national identification. We found partial support for this hypothesis. Higher national identification was significantly associated with stronger positive attitudes towards Sinterklaas in our regression analyses, but SDO was not. However, higher national identification and SDO were both significantly associated (and similarly in strength) with stronger positive attitudes towards Black Pete. These results corroborate the findings of the person-centered analyses. Overall, the general trend in the data suggests that more strongly identifying as Dutch is uniquely related to positive attitudes towards Sinterklaas, while a mix of more strongly identifying as Dutch and higher SDO is related to positive attitudes towards Black Pete.
In summary, these findings suggest that among ethnic Dutch people, acceptance of the unequal treatment of minority groups is related to resistance to changing Black Pete. Even though the controversy surrounding Black Pete is specific to Dutch-speaking countries (e.g., the Netherlands, Belgium, the Dutch Caribbean), our findings may explain resistance to changing insensitive traditions in other cultural contexts as well. For example, in the US, the debate regarding sports teams that have mascots that present racially insensitive caricatures of Native-Americans (e.g., as noble savages) or use racial slurs as team names (e.g., the Washington Redskins, a popular American Football team) suggests that group identification and attitudes regarding equal treatment of marginalized groups may also be at play (Gonzalez, 2005). With this in mind, future studies could focus on conceptual replications by examining if similar subgroups as those found in this study could be identified in other countries and pertaining to other racially insensitive traditions. In this way, it may be possible to determine if resistance to other racially insensitive traditions clusters with identity concerns in some groups or SDO in other groups.
We acknowledge some limitations of the current study. This study consisted of a convenience sample of ethnically Dutch university students who were overwhelmingly young and mainly consisted of women. It is therefore advisable to replicate these findings with a larger nationally representative sample. Replication with a larger sample would also allow for potential gender differences to be examined. In contrast to our findings, other studies have found that SDO tends to be higher in men (Pratto et al., 1994; Sidanius et al., 2000). A larger nationally representative sample would also allow researchers to examine potential sociodemographic differences that are often correlated with more conservative attitudes, such as regional (i.e., city vs. rural) or class (higher vs. lower income) differences. Lastly, person-centered analyses are sensitive to the kind of variables used (Osborne & Sibley, 2017). Future studies should test if the subgroups we identified are specific to attitudes on the Sinterklaas festivities and the Black Pete figure or generalizable to broader topics such as nationalism (a belief in the superiority of one’s nation; Kosterman & Feshbach, 1989), fear of change of cultural traditions, and tolerance of racism (Hunt et al., 2021).
In recent years, there have been a number of positive developments regarding the Black Pete figure. From 2013 to 2020, the proportion of people who reported not wanting to change Black Pete decreased from 83% to 55% (Lubbe, 2020). This suggests that a growing number of Dutch people recognize that the Black Pete figure is racially insensitive and should be changed. Considering that repeated exposure to information can lead to attitude change (Moritz et al., 2017; Zajonc, 1968), perhaps mere repeated exposure to public debates on the racially insensitive Black Pete character in recent years has made some parts of the ethnic Dutch population feel less proud of their Dutch identity, the dominant position of the White Dutch, and subsequently, Black Pete. The public debates have also already affected local policies: some major Dutch cities have made the figure less racially insensitive in their local Sinterklaas festivities by introducing the Chimney Pete figure. This figure is portrayed by volunteers that paint their face with just a few swipes of black paint (to resemble soot from a chimney) and who do not paint their lips red or wear golden hoop earrings. In 2019, the nationally broadcasted version of the Sinterklaas festivities followed suit by only having Chimney Pete accompany Sinterklaas.
In conclusion, our findings suggest that part of the reason why resistance to changing Black Pete in the Netherlands is so widespread can be explained by two underlying reasons: identity concerns and a preference for cultural dominance. These findings are not only relevant for understanding a Dutch cultural phenomenon but also how high SDO people react to power relations being challenged, which is an understudied aspect of social dominance theory (Pratto et al., 2006, p. 311). Thus, our study is part of a recent trend in studies that examine how high SDO people deal with specific societal changes (e.g., Cargile, 2017; Holt & Sweitzer, 2020; Mifune et al., 2019). This trend reflects the current political climate in which protests against the biased and sometimes violent treatment of marginalized ethnic groups are particularly salient in the public eye. Social scientists are well equipped to investigate why substantial parts of ethnically dominant populations react negatively to changing traditional power structures. Insights from social dominance and national identity theories can be applied empirically to specific cases in their cultural contexts to understand the causes and consequences of (resistance to) shifting power relations in society.