1. Introduction

The microaggression concept has received much attention—both scholarly and popular—over the last decade. Microaggressions are, roughly, acts (often but not exclusively speech acts) that exhibit prejudice or neglect toward members of oppressed groups, or states of affairs that exclude or denigrate members of oppressed groups. There appears to be a consensus that microaggressions are “subjective,” among both the concept’s scientific proponents, such as Derald Wing Sue, and its critics, such as Scott O. Lilienfeld. Presumably, the claim that microaggressions are “subjective” means that there is no perspective-independent matter of fact regarding whether an act or state of affairs is a microaggression. That is, whether an act or state of affairs counts as a microaggression depends on how it is perceived by some subject. We disagree with this consensus, distinguishing between “explanatory” and “hermeneutical” microaggression concepts. We argue that there is no a priori reason to regard explanatory microaggressions as “subjective” and that there are compelling phenomenological reasons to regard hermeneutical microaggressions as objective.

2. Microaggressions and Their Effects

The term “microaggression” was coined by African American psychiatrist Chester Pierce (Reference Pierce and Barbour1970) as a label for subtle forms of hostility or disdain commonly exhibited by White Americans against Black Americans. The term was subsequently amplified by psychologist Derald Wing Sue and colleagues (Sue et al. Reference Sue, Capodilupo, Tonino, Bucceri, Holder, Nadal and Esquilin2007), who generalized the concept to encompass many subtle forms of racism. Their oft quoted gloss is that “Racial microaggressions are brief and commonplace daily verbal, behavioral, and environmental indignities, whether intentional or unintentional, that communicate hostile, derogatory, or negative racial slights and insults to the target person or group” (273). The term is now understood broadly, both in critical theory and in psychology, as including not only racial slights but also those related to gender (Capodilupo et al. Reference Capodilupo, Nadal, Corman, Hamit, Lyons, Weinberg and Sue2010; Barthelemy, McCormick, and Henderson Reference Barthelemy, McCormick and Henderson2016), LGBTQ oppression (Nadal, Rivera, and Corpus Reference Nadal, Rivera, Corpus and Sue2010), disability (Keller and Galgay Reference Keller, Galgay and Sue2010; Gonzalez et al. Reference Gonzalez, Davidoff, Nadal and Yanos2015), socioeconomic status (Smith and Redington Reference Smith, Redington and Sue2010), religion (Nadal, Issa, et al. Reference Nadal, Issa, Griffin, Hamit, Lyons and Sue2010), or indeed any form of structural oppression (Sue Reference Sue and Sue2010b), including intersectional forms of oppression (Crenshaw Reference Crenshaw1989; Nadal et al. Reference Nadal, Davidoff, Davis, Wong, Marshall and McKenzie2015; Olkin et al. Reference Olkin, Hayward, Abbene and VanHeel2019). A person or social group that is demeaned or alienated by a microaggression is called a target. For microaggressions that are acts, the agent of the microaggression is generally called a perpetrator or performer.

A commonly cited example of a verbal microaggression (e.g., in Sue Reference Sue2010a; Lilienfeld Reference Lilienfeld2017) is a remark made by John McCain during his 2008 presidential campaign against Barack Obama. A woman at a town hall event said to McCain that she does not trust Obama because “He’s an Arab.” McCain replied, “No ma’am. He’s a decent family man, citizen, that I just happen to have disagreements with on fundamental issues. … He’s not.” McCain’s reply carries the unfortunate (and probably unintentional) conversational implicature that being of Arab descent counts in some way against being “a decent family man” or a “citizen.” As such, it is an ethnic microaggression. Many microaggressions carry such implicatures, which are referred to in the microaggression literature as hidden messages (Sue et al. Reference Sue, Capodilupo, Tonino, Bucceri, Holder, Nadal and Esquilin2007). One of the challenges for researchers who generalize the microaggression concept to new domains of oppression is the identification of the relevant hidden messages (Johnston and Nadal Reference Johnston, Nadal and Sue2010). Microaggressions can also be nonverbal acts, for example, tightly clutching one’s purse or crossing the street when encountering a Black man. And microaggressions can be states of affairs, such as the persistence of a problematic monument. Sue and colleagues (Reference Sue, Capodilupo, Tonino, Bucceri, Holder, Nadal and Esquilin2007) call these latter states of affairs environmental microaggressions.

Some (e.g., Greg Lukianoff and Jonathan Haidt and other critics of “campus culture”) suggest that the proper response to microaggressions is to toughen up or “grow a thicker skin.” As Rini (Reference Rini2018) notes, this may be an appropriate response to mere insults, but it is an insufficient response to microaggressions because microaggressions are components of larger patterns of systematic oppression. The targets of microaggressions are necessarily oppressed groups or their members. Of course slights can target privileged social groups or their members (e.g., “White people can’t dance”), but such slights are not called “microaggressions” because they are not likely to have the same negative effects.Footnote ¹ The relevant difference between microaggressions and other slights is that microaggressions are congruent with oppressive systems, in Liao and Huebner’s (Reference Liao and Huebner2021, 100) sense, and therefore are smaller extensions of larger power structures. Slights that target privileged social groups go against the grain of oppressive social systems rather than being congruent with them.

Rini’s reply is underappreciated in many skeptical discussions of microaggressions, including Lilienfeld’s (Reference Lilienfeld2017), which raises doubts about whether the acts called microaggressions are always performed with malicious motivations. Performers’ motivations may be relevant for assigning blame (see Washington and Kelly [Reference Washington, Kelly, Saul and Brownstein2016] for discussion) but not for understanding the effects of microaggressions on their targets. Much of the psychological literature on microaggressions should be understood as part of what Branscombe, Schmitt, and Harvey call the “psychology of the historically disenfranchised” (Reference Branscombe, Schmitt and Harvey1999, 135, 146): empirical investigations that focus on the psychology of oppressed social groups rather than, like much of the implicit bias literature, the mental states of those who are privileged.

And it is hypothesized that the aggregate effect of microaggressions—perceived or otherwise—on their targets is significant, and not only because they cause gratuitous pain or discomfort. Perceived discrimination regarding race, gender, and sexual orientation predicts psychological and somatic health outcomes (Carter Reference Carter2007; Mays, Cochran, and Barnes Reference Mays, Cochran and Barnes2007; Herek Reference Herek2009). Racial gaps in health outcomes in the United States are not fully explained by differences in socioeconomic status or self-esteem (Gee et al. Reference Gee, Spencer, Chen, Yip and Takeuchi2007a, Reference Gee, Spencer, Chen, Yip and Takeuchi2007b). Plausibly, microaggressions play a role in explaining these recalcitrant health gaps, and many discussions of microaggressions are motivated by appeal to various outcome gaps (in health, academic or professional achievement, etc.). The detailed mechanism by which microaggressions contribute to such outcome gaps is not known (Okazaki Reference Okazaki2009; Torres, Driscoll, and Burrow Reference Torres, Driscoll and Burrow2010), but stress seems to be a mediating factor (Harrell and Taliaferro Reference Harrell and Taliaferro2003), complicated by in-group identification, which seems to have a protective effect (Crocker and Major Reference Crocker and Major1989; Branscombe et al. Reference Branscombe, Schmitt and Harvey1999). The scientific situation is made more complicated by the multiplicity of experimental protocols (Sullivan Reference Sullivan2009): since microaggression incidence is measured in a variety of ways, experimental inference about microaggressions is complicated in ways that are played down in published literature. And in some discussions, “microaggression” may function as a catchall term referring to any manifestations of structural oppression that are relatively difficult to measure independently.

So in the interest of promoting a little more clarity, we distinguish two microaggression concepts. The explanatory microaggression concept refers, ex hypothesi, to some factor that explains recalcitrant gaps in desirable outcomes (e.g., good health, professional success) between members of privileged and oppressed social groups, such as those that remain after other factors like wealth, income, and legal discrimination are accounted for. Microaggressions in this sense may turn out to be a variety of diverse factors (they may be “lumpy”; see Feest Reference Feest2020); we will not know exactly what they look like until we have a more sophisticated causal understanding of recalcitrant outcome gaps. But the term “microaggression” functions in some discourse as a more determinate label for concrete experiences of slights and invalidations. So, let the hermeneutical microaggression concept be what is invoked in such contexts. The hermeneutical microaggression concept is a hermeneutical resource (Fricker Reference Fricker2007) that helps people to make sense of their lived experiences, and the popularity of the microaggression concept outside of the behavioral and social sciences is probably largely due to its hermeneutical role. It is an open empirical question whether the explanatory and hermeneutical microaggression concepts are largely coextensive.

3. Two Senses of “Subjective

”So are microaggressions objective? The answer depends on whether we are talking about explanatory or hermeneutical microaggressions. But clarification is also in order regarding the terms “objective” and “subjective.” Philosophers tend to reserve the term “subjective” for propositions whose truth values vary according to a perspective (MacFarlane Reference MacFarlane2014; a “context of assessment”). For example, a dress may look blue and black to me and may look white and gold (i.e., not blue and black) to you. There is a perspective-independent fact about what color the dress is but no such fact about how the dress looks; it looks different to different people. Let us call such claims alethically subjective, and claims that have perspective-independent truth values can be called alethically objective.

By contrast, in common parlance a claim is often said to be “subjective” if reasonable people disagree about its truth value, even if the claim has a perspective-independent truth value. We may call claims that are controversial in this manner discursively subjective. The claim that Shakespeare’s works were written by William Shakespeare is discursively subjective—some folks believe the plays and poems were written by someone else. But there is a perspective-independent fact of the matter about who wrote Shakespeare’s works, so the claim is not alethically subjective.Footnote ² Both alethic and discursive subjectivity are properties of claims rather than concepts, words, or acts, but for ease of expression we will talk about “microaggressions” as subjective or objective, meaning that classifying an act or state of affairs as a microaggression is subjective or objective.

Now, obviously claims about microaggressions can be discursively subjective—there is often disagreement about whether a particular act or state of affairs is a microaggression. Nevertheless, it is commonly held that microaggressions are also alethically subjective. Lilienfeld criticizes the microaggression concept on the grounds that microaggressions are thought to be “necessarily in the eye of the beholder” (Reference Lilienfeld2017, 143), and Sue claims that “microaggressions are about experiential reality” (Reference Sue2017, 171). Lilienfeld regards the subjectivity of microaggressions as a source of confusion: “If Minority Group Member A interprets an ambiguous statement directed toward her … as patronizing or indirectly hostile, whereas Minority Group Member B interprets it as supportive or helpful, should it be classified as a microaggression? The [microaggressions] literature offers scant guidance in this regard” (Reference Lilienfeld2017, 143). Generally speaking, that a claim is discursively subjective does not imply that it is alethically subjective (e.g., the Shakespeare case above is discursively subjective but not alethically subjective). So even if there is reasonable disagreement about whether a particular act or state of affairs is a microaggression, that does not imply that the microaggression is alethically subjective.

Lilienfeld continues: “it is unclear whether any verbal or nonverbal action that a certain proportion of minority individuals perceive as upsetting or offensive would constitute a microaggression. Nor is it apparent what level of agreement among minority group members would be needed to regard a given act as a microaggression” (Reference Lilienfeld2017, 143). Such questions are unmotivated. No serious proponent of the microaggression concept holds that poll results should determine which acts are microaggressions. While “focus groups” and similar methods are sometimes used to determine which kinds of acts should be regarded as microaggressions (e.g., the use of Consensual Qualitative Research methods in Nadal et al. [Reference Nadal, Davidoff, Davis, Wong, Marshall and McKenzie2015]), researchers do not assume that intersubjective agreement among participants is a criterion for being a microaggression. Rather, “focus group” methods are generally employed as techniques for discovering new varieties of microaggression while minimizing the role of researcher biases (see, e.g., Nadal et al. Reference Nadal, Davidoff, Davis, Wong, Marshall and McKenzie2015, 150–51).

Furthermore, as we argue below, there is no a priori reason to regard microaggressions as alethically subjective. Regarding explanatory microaggressions, it is an open empirical question whether outcome gaps are explained by perceived microaggressions or by microaggressions regardless of how they are perceived by their targets (i.e., microaggressions ascribed according to an alethically subjective or objective criterion) or by some other factor. Regarding hermeneutical microaggressions, the concept fails to serve as an adequate hermeneutical resource unless microaggressions are regarded as objective.

4. Explanatory Microaggressions: Measures and Constructs

Lilienfeld observes that most microaggression studies rely on self-report measures and takes this to be a consequence of the fact that microaggressions are alethically subjective (Reference Lilienfeld2017, 151). For example, many studies of microaggressions against African Americans use an instrument called the Daily Life Experiences (DLE) scale (e.g., Scott Reference Scott2003; Seaton, Yip, and Sellers Reference Seaton, Yip and Sellers2009; Torres et al. Reference Torres, Driscoll and Burrow2010), developed by Jules Harrell. The instrument consists of 17–20 items describing discriminatory experiences, such as “overhearing or being told an offensive joke” or “being left out of conversations or activities” (Seaton et al. Reference Seaton, Yip and Sellers2009, 417). Study participants rate how often they have each kind of experience on a scale from “never in the past year” to “once a week or more.” Their responses are analyzed (in various ways, depending on the study) to obtain a quantity representing how often participants experience racial microaggressions. The DLE scale is a so-called self-report or subjective measure, since study participants more or less transparently report information in which experimenters are interested for its own sake (in contrast to behavioral measures or other indirect measures). Self-report measures are common in psychological research on “subjective” constructs like subjective well-being (Alexandrova Reference Alexandrova2008) or conscious visual experience (Boone Reference Boone2013), where a “construct” in psychology is a theoretical term whose quantity can be measured (Stone Reference Stone2019, 1250n2).

However, the connection between subjective constructs and so-called subjective measures is not straightforward. An experimental measure will generally differ from its associated construct in various ways. For example, a Stroop test may be administered as a measure of cognitive depletion (as in, e.g., Richeson and Trawalter Reference Richeson and Trawalter2005). But Stroop performance is a temporal measure (a relative delay, measured in milliseconds), whereas cognitive depletion is theoretically something more abstract: it may manifest as a temporal delay or as poorer performance or in various other ways. So here a temporal measure is used to approximate, for the purposes of experimental analysis, the quantity of a more abstract construct of interest (cognitive depletion).

More to the current point, self-report measures may be used to gather information about constructs whose values are alethically objective. Consider, for example, the Perceptual Awareness Scale (PAS), a graded measure of visual awareness (Rams⊘y and Overgaard Reference Rams⊘y and Overgaard2004). Study participants are briefly shown an image (often for less than 250 ms) and asked to classify their visual experience as “clear image,” “almost clear image,” weak glimpse,” or “not seen.” One may think that this is a subjective measure for a subjective construct, since visual experience is often said to be “subjective,” but alethic subjectivity is a property of claims, so we must be precise about what claim is at issue. Visual experience is subjective in that the content of two visual experiences may differ for various judges (or one judge at different times), although those experiences are of the same object in the same conditions. We often characterize the contents of such experiences using clauses with “seem” or “look” as the main verb, and such clauses are alethically subjective. A dress may look blue to Derrick and at the same time look white (i.e., not blue) to Mari: the truth value of an utterance like “This dress looks blue” may vary depending on the judge. But the PAS does not measure what the content of a visual experience is; the PAS measures whether a visual experience of a stimulus occurred for a particular observer and how clear that experience was. This is an alethically objective state of affairs. The truth value of “Derrick had a clear visual experience of the stimulus” does not vary according to who evaluates it. If Derrick and Mari disagree about the truth of such a sentence, then one of them must be wrong (and it is probably not Derrick).

Similarly, instruments like the DLE scale, which purport to reveal rates of microaggression incidence in a participant’s life through self-report, may be fallible measures of an alethically objective quantity. We say “may” because much extant microaggression research does not distinguish clearly between alethically objective and subjective interpretations of microaggression incidence. Instruments like the DLE scale may be used either to measure the frequency of a participant’s exposure to demeaning incidents (an alethically objective quantity) or to measure the participant’s perception of how often she experiences demeaning incidents (an alethically subjective quantity). Microaggressions in the explanatory sense are some factor that explains recalcitrant gaps in desirable outcomes between members of privileged and oppressed social groups, such as those that remain after other factors like wealth, income, and legal discrimination are accounted for. It is an open question whether this factor is (1) mere exposure to demeaning incidents, regardless of how they are perceived by their targets, or (2) the perception of one’s experiences as demeaning, or (3) something else. That is, it is an open empirical question whether explanatory microaggressions are alethically objective or subjective. Further empirical study is needed to assess the relative merits of these hypotheses.

As a matter of verbal hygiene, it seems reasonable to us to treat explanatory microaggressions as alethically objective and then to examine whether outcome gaps are caused by exposure to microaggressions per se or by the perception of events as microaggressions. By analogy, the standard for whether an act is a sexual assault is not whether the survivor characterizes the act as “sexual assault” or even as harmful. But the matter of which way to speak can only be settled by the community of speakers (in this case, the community of social and behavioral scientists), not by fiat, and it seems to us that the matter has not yet been settled.

We wish to be clear that microaggression research employs a variety of methods that vary in quality and purpose; the DLE scale is only one instrument among many. Our main objective here is not to conduct a methodological review (for which see, e.g., Okazaki Reference Okazaki2009; Lau and Williams Reference Lau, Williams and Sue2010; Wong et al. Reference Wong, Derthick, David, Saw and Okazaki2014) but to argue against a tempting error. It is simplistic to identify a construct with its measure, and it is an error to freely attribute the properties of a measure to its associated construct. So while it is true that microaggression frequency is often measured using participant self-reports, we should not infer from this fact that microaggression incidence is alethically subjective. Existing measurement practices do not settle the question of whether microaggressions are “in the eye of the beholder.”

5. Hermeneutical Microaggressions: Phenomenological Considerations

Whereas it is an open question whether explanatory microaggressions are alethically subjective, there is compelling reason to regard hermeneutical microaggressions as alethically objective. Our argument depends on the commonly reported phenomenology of microaggression targets. Members of oppressed groups often report experiencing confusion and uncertainty about whether an act directed toward them is a subtle expression of prejudice or whether it is no different from an act that would have been directed toward a privileged person. This feature of microaggressions is sometimes called “attributional ambiguity” (Crocker and Major Reference Crocker and Major1989). For example, a woman might be addressed at work by her first name (e.g., “Stephanie”) rather than by her title and surname (say, “Dr. Appiah”). In a context where either form of address is acceptable, and where the base rates are not known (i.e., it is not known how often people in general, or people of various genders, are addressed by their first names vs. by their titles and surnames), it can be difficult to determine whether the address expresses a slight.

Here is an argument that we should consider hermeneutical microaggressions to be alethically objective. Supposing the contrary, that microaggressions are alethically subjective, there are two possibilities. First, perhaps, as in many matters of taste, it is appropriate to allow people their own perspectives. So whoever feels the act of addressing the woman by her first name was a gendered slight regards it as a microaggression, and whoever feels the form of address was not influenced by gender does not regard it as a microaggression. If microaggressions are alethically subjective, as we are currently supposing, then there is no perspective-independent fact of the matter about whether this incident is a microaggression (as in predications of “is tasty” or “looks blueish to me”). A second possibility is that people have their own perspectives, but the target’s perspective is decisive: the act is a microaggression if and only if the addressed woman feels slighted. In both of these possibilities, it makes no sense for the woman to wonder whether the act was really an expression of prejudice, that is, whether it was really a microaggression. On the first option, there is no fact of the matter about whether the act was a microaggression. On the second option, the matter is decided by the woman’s own perspective, so her judgment settles the question.

However, people who experience relatively subtle microaggressions often report wondering precisely about this. Indeed, it is often claimed (e.g., by DuBois Reference DuBois1903; Bartky Reference Bartky1975; Sue et al. Reference Sue, Capodilupo, Tonino, Bucceri, Holder, Nadal and Esquilin2007; and others) that much of the harm of microaggressions is caused precisely by anxiety and paranoia regarding one’s inability to quickly and accurately assess whether an act was indeed a microaggression. Only the objectivist view of microaggressions accounts for this phenomenology. If we seek hermeneutical justice, we have reason to adopt concepts that make sense of rather than obscure common experiences for members of oppressed social groups (Fricker Reference Fricker2007). So we should regard microaggressions as objective, in that there are perspective-independent facts about whether particular acts or states of affairs are microaggressions in the hermeneutical sense.

6. Conclusion

We argued, against the common view, that microaggressions should not be regarded as alethically subjective. For microaggressions in the hermeneutical sense—considered as a category of items that help members of oppressed social groups to make sense of their lived experience—we argue that only an objectivist view rationalizes the distress commonly experienced due to attributional ambiguity. For microaggressions in the explanatory sense—considered as the causes of recalcitrant outcome gaps—we acknowledge that it is an open question whether they are best regarded as alethically objective or subjective. But we argued against a tempting view, expressed by Lilienfeld and others, that self-report measures are especially suited for measuring the value of theoretical constructs that are alethically subjective. People will continue to question whether particular acts or states of affairs count as microaggressions, and we contend that those questions have objectively accurate responses.