1 Introduction

Recent years have seen increasing debates regarding the depiction of monster races in games such as D&D, or in movies and books based in fantasy, such as Lord of the Rings (LotR). Much of the debate has focused on orcs, a creature based in Irish and English folklore [1] Orcs subsequently had been incorporated into Tolkien’s Lord of the Rings mythos, as well as the D&D role-playing game system. Monsters are often portrayed as inherently evil, and this raises the question of whether there are parallels to real-life racism or whether monster race essentialism may promote race essentialism in real life. There are also concerns that monster depictions are based on actual stereotypes of real-life human groups. This study sought to address the second question, particularly regarding visual depictions. Specifically: do visual depictions of orcs repeat visual stereotypes, particularly of African or Asian ethnic groups?

Many of these debates have taken place in online communities rather than in academic journals (e.g., [2]). Given this non-academic context of much of the debate, hypotheses were sometimes presented vaguely, but appeared to fall into several main areas. First, that playing D&D might be related to racism in real life due to exposure to race essentialism in evil monster races. Second, that evil monsters were offensive or hurtful to people of color. Third, that stereotyped race depictions were used in the presentation of evil monster races such as orcs. For instance, Hoffer [2] expressed that “Over time, various artists and authors used elements of various real-world non-European cultures when depicting orcs in books and other media, thus solidifying the idea that orcs were a reflection of non-European races, which is incredibly problematic.” Hoffer also made the empirical claim, “Many people see games like Dungeons & Dragons as codifying and normalizing the use of racial stereotypes, which leads to both racist situations in the gaming table and encourages racist attitudes getting adopted by some players.” Barber [3], speaking of the release of a recent D&D product with optional rules regarding race stated, “It wouldn’t start to push back against the weight of decades of psychological anchoring on highly problematic, racist stereotypes and narratives.” As D’Anastasio [4] wrote, following the protests and riots of the Black Lives Matter movement, even the developers of the D&D product appeared to accede to concerns about racism, releasing a statement describing the depiction of orcs and other monster races as “painfully reminiscent of how real-world ethnic groups have been and continue to be denigrated.” No less than National Public Radio [5] released an article titled “'Dungeons & Dragons' Tries To Banish Racist Stereotypes”, but how common were these racist stereotypes in the first place? The questions regarding whether playing D&D is associated with racism in real life, or whether there were widespread concerns about the depiction of D&D monsters among people of color were recently addressed in a survey study of over 300 participants of various ethnic backgrounds [6]. Briefly, the results found no association between D&D playing and racist attitudes, nor were the majority of participants including participants of color offended by the depiction of orcs as evil in the D&D game. As such, claims such as expressed by Hoffer about players adopting racist attitudes from the game are not supported by what data is currently available.

1.1 Theoretical concerns about race in games

Aside from the Ferguson [6] study mentioned above, relatively little empirical work has examined the issue of race depictions in games. However, several scholars have examined this issue more theoretically. Not all such articles focus on the D&D game or Lord of the Rings movies specifically, but rather focus on other related games and media such as World of Warcraft (WoW), a video game which features monster races such as orcs, trolls, etc. Spiecker [7], for instance, claims that depictions of orcs and trolls in the WoW game present African stereotypes such as primitive dances, bone piercings in noses and even Jamaican accents, and references to voodoo. Focusing on elves rather than orcs or trolls, Poor [8] concludes that elves are often depicted as a more idealized western minority group. Others might reasonably conclude that Tolkien had based orcs on Asiatic races given a comment from one of his letters “They are (or were) squat, broad, flat-nosed, sallow-skinned, with wide mouths and slant eyes: in fact degraded and repulsive versions of the (to Europeans) least lovely Mongol-types.” In this context, Mongol appears to refer to then-common use of the term mongoloid to refer to all Asians, not the Mongols specifically (see [9]). However, Tolkien also generally disavowed that the orcs were intended to represent any specific human ethnicity.

Much of this work comes from traditions within the humanities, including post-colonial and related theories, commonly expressing concerns about power dynamics which can function along racial and other identity lines. For instance, Young [10], speaking of Tolkien’s Lord of the Rings (LotR), says, “…significant elements of his world construct race in problematic ways. The novels are soaked in Germanic thinking, under the guise of Anglo-Saxonism, which is very closely linked to that which animated the racial excesses of Nazism” (p. 349). Young focuses her analysis on video games specifically based in the LotR world and considers them replete with racist depictions. Likewise, Monson [11] in analyzing the WoW game once again, expresses concern over biological essentialism, a common concern from many authors writing from the post-colonial tradition, and finds racist depictions to be common. Also assessing WoW, Langer [12] largely agrees with other critiques particularly as related to trolls, more so than orcs, elves, or other races, and does offer a more nuanced analysis, “World of Warcraft is a tricky, complex construction of cultural meaning in this way: it is both racist and antiracist, frequently at the same time” (p. 105).

However, I briefly offer several critiques of this scholarly tradition of games critique. First, with several notable exceptions e.g., [8], most analyses rely on the authors’ own speculative assessment of the media products in question rather than a more cautious, preregistered, empirical approach. In other words, it is the authors’ opinion that the content is racist, but this conclusion is not clearly documented in evidence. Related to this, traditions based in post-colonialism and other identity and power focused theories may be incentivized to problematize media in ways consistent with their own narrative worldviews. In other words, postcolonial theorists may find problems in Western based media simply because that is the raison d’etre of what such theoretical traditions do. Such traditions may beg the question wherein asking, “Is D&D racist?” the only acceptable answer is “yes.”

These critiques of identity/power based theoretical traditions (postcolonial or otherwise) is that they may exhibit negativity bias (a focus on negatives, worrying over “harms”, incentivizing scholars and students to “problematize” anything, particularly coming from Western traditions) which is poorly suited to available data. For instance, given much of the discourse focuses on WoW, it is worth noting that the empirical evidence has tended to find that exposure to WoW is associated with expanding and diversifying of communication partners rather than contracting [13]. As noted by Ferguson [8], the conclusion that exposure to the D&D product is “harmful” vis a vis racism, is not supported by the available data, though of course more would be welcome. Nonetheless, the current narrative on race and racism in Tolkien, D&D and other products such as WoW would strongly benefit from more empirical data.

1.2 The present study

At present, less is known about the degree to which visual depictions of orcs rely on racist stereotypes. Such stereotypes can be important as, although there appear to be no direct relationships between D&D playing and racist attitudes, negative depictions of racial minorities in media may promote racist stereotypes [14]. In studies of other media such as television, effects of exposure to stereotyped depictions appears to have less of a direct effect and is rather moderated by the personality characteristics of individual viewers [15, 16]. This can also work in positive directions, with positive exposures to outgroups associated with improved attitudes seen, for instance, in video games [17].

The current paper, however, focuses less on the potential impacts of stereotypical depictions and rather on whether they exist in the first place for D&D and LotR orcs. In particular, this article focuses on depictions of Asian and African peoples, who have often been at the center of discourse on D&D orcs. Both Asian and African people have been the target of visual racist stereotypes dating back to the nineteenth century as print and visual media became common. For African people, minstrel shows or early cartoons often emphasized the supposed exoticness, but also childishness of African people, with features such as exaggerated red lips, bulging, wide eyes, childish clothing, chalky black skin and wide grins [18, 19]. Such depictions were often made by white performers in blackface, though picture and cartoon depictions were also common. Racist Asian depictions likewise emphasized the exoticness of Asian people, but also suggested sneakiness and threateningness. Depictions often included features such as the queue hairstyle, baggy, flowing clothes, exaggerated “slit” eyes, bucked teeth, and yellow skin [20,21,22].

As such the current study will address the question of whether racist stereotypes of Asian or African people were commonly used in either D&D or LotR visual depictions of orcs. It is hoped that this analysis may help inform current debates regarding such depictions.

2 Method

All methods for this study were preregistered prior to data analysis. The preregistration can be found at: https://osf.io/k2cpn

2.1 Image selection

A search for images was made using images.google.com. This approach has been used in other content analyses in different areas (e.g., [23]). For each edition of D&D the search terms used were “Orc”, AND “dungeons and dragons”, AND “1st edition”, as well as “Orc”, AND “dungeons and dragons”, AND “1e”. For each subsequent edition (2–5), 1st or 1e was imply exchanged for 2nd, 2e, and so forth. As much as was possible only official actual source book or module pictures were used as opposed to fan art, or orcs from other sources such as Magic the Gathering, or World of Warcraft.

For the Lord of the Rings movies, separate searches were used for both the 1970s cartoons (which included several technically unrelated cartoons) as well as the 2000s Peter Jackson movies. Search terms took the form of “Orc”, AND “Lord of the Rings” AND “cartoon” or “Orc”, AND “Lord of the Rings”, AND “Peter Jackson movie.” For each movie (Twin Towers, Return of the King, Hobbit, etc.), the movie title was simply exchanged for “Lord of the Rings”.

Note that the intent was to get a reasonably representative sample of visual depictions of orcs in the D&D game and LotR movie systems, not every single visual depiction. However, pictures included were those that were identified by Google Images and all such images found were included. The author did not select from among them. Safesearch was turned off during the search so that racist images wouldn’t be artificially filtered out (this was confirmed from a simple search for “racist cartoon” which had no difficulty returning numerous explicitly racist examples). Ultimately 78 images were located, 30 from the D&D system, 18 from the various LotR cartoon movies and 30 from the various Peter Jackson movies. These reflected all unique examples available via Google Images, copyright laws likely limiting the availability of images.

2.2 Stereotypes coding

Each image was coded for the presence or absence of racial stereotypes as sourced from previous scholarship on African [xviii, xix] or Asian peoples [xx, xxi, xxii]. African visual stereotypes included: bulging white eyes, exaggerated red lips, coal/chalky/burnt cork skin, large grins, and infantilizing costumes (large collars, too-big shoes). Asian visual stereotypes included: yellow skin, “slit” eyes, queue ponytail, flowing robes/clothes, and exaggerated “buck” teeth. Each image was also coded for skin color and an assessment of whether the orc’s armor appeared to be European, Asian, African, or other. European style armors tended to make use of metal rings and plates, often atop leather, becoming progressively heavier as the Middle Ages progressed. By contrast, both African and Asian armor tended to be lighter, more often used quilted or lamellar or laminar (although the Romans also used laminar armor) armor, which tended to be lighter [24].

The issue of skin color was difficult to discern in some samples. Some samples were in black and white while others, particularly from the rotoscoped (placing animation over live action sequences) 1978 Lord of the Rings movie, projected ambient lighting onto vague figures. In the latter case, the resultant skin color contrast was coded for the skin color variable.

3 Results

Results from this study are primarily descriptive. Table 1 presents the frequency of racist stereotypes appearing in samples from the D&D game, the 1970s–80 s cartoon movies and the 2000s Peter Jackson movies. As can be seen, qualities evocative of racist stereotypes were rare in all three formats and nonexistent in the cartoon movies. It’s important to qualify those coding of stereotypes in two ways. First, coding was generally coded generously, meaning if there was any question of a feature evoking a stereotype, it was coded as a yes. However, it was not always clear whether a quality was really intended to evoke racist caricatures. For instance, “coal skin” appeared more commonly in the Peter Jackson movies, but this was mainly due to the Uruk Hai; mutant orcs often portrayed with dark but uneven skin tones and face paint. Such tones did not appear to evoke the minstrelsy of the past. This was true also for other samples where wide grins, for instance, appeared to be intended to conjure menacing intent rather than race. Second, where these features appeared, they were always idiosyncratic, meaning a single feature appeared alone, but not in combination with other caricatured features. For instance, ponytails were present in some D&D depictions of orcs, but never in combination with other stereotyped Asian qualities. This suggests these features were generally incidental, likely intended to project exoticness (similar hairstyles were often used in shows such as Vikings for instance) rather than racist stereotypes. Prevalence of stereotyped depictions that occurred in concert would have been more likely to indicate purposeful or even implicit racist stereotyping. In other words, coding of qualities was generally weighted in the direction of coding “yes” if there was any question of the caricature existing and the sporadic findings here should not be overinterpreted.

Table 1 Proportion of orc samples with racist stereotyped depictions
Full size table

Related to the two other categories coded, skin color and armor history, results are as followed. Across platforms, gray skin color was the most common (30.8%), followed by black and white images where skin color could not be determined (15.4%), followed by black (12.8%), green (11.9%), white (9%), brown (9%), blue (5.1%), orange (3.8%), yellow (1.3%) and red (1.3%). A chi-square analysis across platforms was significant (χ2 = 120.12, p < 0.001), with both gray and black (mostly in Uruk Hai) skin tones more common to the LotR cartoon and movies, with black and white images more common in D&D. Note that this significant outcome is unrelated to the question of stereotyping based on skin color, and reflects mainly the reliance on black and white images in the early versions of D&D.

More subjectively, suggesting that racial features were otherwise represented in the depiction of orcs appeared unlikely to the author. D&D orcs originally were presented almost akin to walking pigs, though these shifted to more humanoid forms in later editions (albeit with a range of blues, greens, and other skin colors). The LotR cartoons featured orcs either as vague shadowy monsters with little humanoid discernment at all (as in the 1978 film), or as gray, toad-like creatures in the later cartoons such as Return of the King. In the Peter Jackson films, orcs appeared to have more European features than anything, with the exception of the Uruk Hai who were presented as bigger, darker and more monstrous (albeit not particularly evocative of African people other than with the darker tone of skin).

Regarding armor types, in most cases, the armor type was simply unidentifiable, being incomplete, broken, unrelated to historical armor types, etc. (53.8%). Of identifiable armor types, historically European type armor and weapons were most common (39.7%) followed by Asian (1.3%). 5.1% of orc depictions featured no armor at all.

4 Discussion

Concerns that racist stereotypes of real-life races have influenced visual depictions of orcs in the D&D game system or LotR movies have been revived, particularly on social media. The current study sought to examine this with a content analysis of visual depictions of orcs in these systems. Results found that the use of racist caricatures in the visual depictions of orcs were rare and incidental, never occurring in the type of combination that would suggest that stereotyped depictions were intended. There was no particularly consistent pattern to skin color of orcs, with gray skin being most common, and orc armor and weapons were more influenced by European historical designs, than Asian or African types. Taken together, these results suggest that visual depictions of orcs in the D&D and LotR systems are not influenced by racist stereotypes.

This brings up the interesting question of, why then some activists think orcs induce racist stereotypes. It may be simply that any depiction of fearsomeness or evilness is considered to be evocative of African of Asian people. In this sense, it seems likely that such activists have reversed the stereotyping process. It is of the historical record that Western depictions of Africans and Asians (and also other European groups such as Germans) relied on stereotypes of dangerousness and sneakiness. These are undesirable qualities that have been used to portray monsters in indigenous folklores for millennia. In other words, depictions of monsters were portrayed onto human population groups, stigmatizing them. Many modern activists appear to be reversing this process, assuming that qualities of actual human population groups are being superimposed onto monster races such as orcs. This latter does not appear to be the case, and there appears to be little in the depiction of orcs that is influenced by actual human cultural groups. In other words, racist stereotypes sometimes portray human population groups as akin to monsters, but at present monsters such as orcs are not meant to portray human population groups.

One interesting possibility is that individuals who assume orcs are intended to portray African or Asian groups may be projecting their own prejudices or stereotypes related to these groups onto orcs and assuming most people view them the same way. Indeed, empirical evidence has already found the majority of individuals including people of color finding nothing either offensive or racist in depictions of orcs [6]. One interesting future line of research may involve examining why a minority of individuals tend to see stereotypes in images where others do not (including the majority of individuals from the ostensibly stereotyped group). It could be the case that such individuals are projecting their own biases, or it may be that some individuals are attempting to use claims of victimhood for personal influence, a phenomenon already documented in the psychological literature [25]. Further work on this issue would be useful. It would also be worth considering how a minority of such individuals have outsized influence in culture war debates surrounding popular culture, whether on the political left or right.

One other concern is worth mentioning. As is noted, inherently evil monsters are common to many indigenous folklores [1]. If modern progressivism declares such depictions taboo, modern progressivism may ultimately find itself in the practice of exactly the forms of colonial cultural imperialism it tends to decry in history.

5 Limitations

As with all studies, this one has potential limitations. First, it is a purely descriptive study, and no causal implications can be drawn. Second, any search for images inevitably will miss some and it is possible there may differences between images located and those which are not. Third, conclusions from this study cannot be applied to other systems where orcs are depicted such as World of Warcraft. Furthermore, this study only considered visual depictions of orcs and it is possible that racist stereotypes could exist either in the movie dialogue or game description of orcs. Regarding the images, in each case efforts were made to track down and verify the source of each image. However, it is possible that some independent or fan images may have passed through this process. Lastly, rating images for stereotypes inevitably includes some subjectivity. However, true blinding of raters to the hypotheses of the study is unlikely given the fairly direct and obvious nature of the analysis. The methods of this study also make replication efforts fairly easy.

6 Conclusions

Any one study is unlikely to resolve intense culture-war debates. However, in combination with other recent data [6], the current study suggests that empirical support for activist pushes to change the depiction of orcs and other monster races in the D&D system or elsewhere may be weak. The concern expressed by this author is that efforts to address race relations in areas where empirical support is weak and outcomes are fairly peripheral, may actually exhaust efforts better spent toward enacting criminal justice reform, addressing education disparities in the school system, or addressing income inequality. Culture war battles over popular culture may actually backfire, reducing support and enthusiasm for these other worthwhile efforts. Certainly, more studies on race in gaming and movies would be welcome, particularly from preregistered, open science frameworks.