The Unlevel Knowing Field: An Engagement with Dotson’s Third-Order Epistemic Oppression, Alison Bailey

Author Information: Alison Bailey, Illinois State University,

Bailey, Alison. “The Unlevel Knowing Field: An Engagement with Dotson’s Third-Order Epistemic Oppression.” Social Epistemology Review and Reply Collective 3, no. 10 (2014): 62-68.

The PDF of the article gives specific page numbers. Shortlink:

Please refer to:

9497392072_779d192d6f_zImage credit: arbyreed, via flickr

We and you do not talk the same language. When we talk to you we use your language: the language of your experience and your theories. When we try to use it to communicate our world experiences, we only succeed in communicating our experience of exclusion. We cannot talk to you in our language because you do not understand it (Lugones and Spelman 1983, 575).

Social justice demands that we think carefully about the epistemic terrain upon which we stand and the epistemic resources each of relies upon to move across that ground safely. Epistemic cartographies are politically saturated. Broadly speaking these terrains are unlevel playing fields—I think of them as unlevel knowing fields— that offer members of socially dominant groups an epistemic home turf advantage.[1]  Members of marginalized groups must learn to navigate this field creatively.

Imagine living in an epistemic twilight zone,[2] a world where many of your lived experiences are regularly misunderstood, distorted, dismissed, erased, or simply rejected as unbelievable. Perhaps you can’t find words to capture an experience that you know to be very real. Or, perhaps there is a local vernacular, but it is rendered nonsensical by listeners outside of your community.  Are the ‘shared’ epistemic resources that structure the unlevel knowing field so resilient in the long run that they absorb, erase, ridicule and repel your words?

Understanding how the unlevel knowing field produces and maintains epistemic oppression (and privilege) requires a set of nuanced conceptual tools for explaining the impact epistemic exclusion has on marginalized knowers’ ability to produce knowledge. This project is at the heart of Kristie Dotson’s “Conceptualizing Epistemic Oppression” (2014). Her distinction between first-, second- and third-order epistemic exclusions offers us a means of pulling third-order cases from the unlevel knowing field, so that we may describe their unseen resilient dimensions.  Isolating third-order epistemic exclusions points to a heartbreaking conclusion: the largest obstacle to overcoming epistemic oppression the origins and structure of the unlevel knowing field itself.

My engagement with Dotson’s essay begins with an overview of first- and second-order epistemic exclusions. I use examples from the epistemic injustice literature, and some of my own,  to highlight the important distinction she makes between reducible and irreducible forms of epistemic oppression.   Next, I turn my attention to her account of third-order epistemic exclusions. I offer a brief explanation of why her sketch of at this level makes an important contribution to the literature on epistemic injustice.  In closing I suggest that Dotson’s account of third-order epistemic exclusions has a cognitive bend that limits the resources she might use for bringing about change, and suggest opening up the conversation to include affective, aesthetic or mystical resources.

First-Order Epistemic Oppression: Testimonial Injustice

Dotson’s account of the irreducible nature of third-order epistemic oppression is best understood against the background of first- and second- order oppressions. I want to remind readers of some of these examples with an eye toward Dotson’s central claim that these expressions are reducible to social and political factors.[3]  In Epistemic Injustice: Power and the Ethics of Knowing (2007) Miranda Fricker makes an important distinction between testimonial and hermeneutic injustice. Testimonial injustice is an example of what Dotson calls first-order epistemic exclusion. It happens when “prejudice causes a hearer to give a deflated level of credibility to a speaker’s word” (Fricker 2007, 1).  Examples of these injustices are tragically common. Tom Robinson’s character in Harper Lee’s To Kill a Mockingbird is innocent of assaulting a white woman, but the all white jury can only hear his courtroom testimony as the words of a “lying Negro.”[4]  The all white panel of men on the Senate Judiciary Committee is so skeptical of Anita Hill’s workplace sexual harassment testimony that they never call in her corroborating witnesses. Senator Howell Heflin makes sense of her testimony by insinuating that she is a ‘scorned woman’ who is crying sexual harassment to cover up a soured love affair. Or, consider columnist George Will’s recent claim that women cry rape so that they get “special privileged survivor status.”[5] These cases illustrate how the epistemic agency of knowers is compromised by a credibility deficit.  If we think about epistemic credibility as a resource, then it is a resource that is unevenly distributed along gendered and racialized lines.

Dotson argues that distributing credibility along these lines is inefficient. It creates differences in epistemic power that compromise the epistemic agency of marginalized knowers by reducing their ability to participate in the production of knowledge.  Testimonial injustice is an expected epistemic by product of larger social and political systems; that is, we can appeal to larger social and historical patterns to explain the cultural failure to believe people of color. For example, the historical branding and re-branding of women and men of color as thieves, delinquents, or criminals is what maintains their current credibility deficit: no one believes a crook.

If testimonial injustice springs from inefficiencies within shared epistemic resources, then how we might we apply these resources more efficiently? Dotson suggests that first-order changes are “alterations made to address a given problem without changing the underlying schemata” (2014, 11). This means testimonial injustices can addressed simply by using ‘off-the-shelf’ shared epistemic resources from the unlevel knowing field to redistribute credibility along more equitable lines. The inefficiencies in the credibility economy are correctable by prompting epistemic agents to pursue minimal reforms in their interactions with other epistemic agents.  For example, epistemic credibility might be more efficiently distributed through cognitive reform projects (Mills, 1997). We can administer an implicit bias test to help jury members and police officers become more mindful of how prejudiced associations between a person’s race/gender and their credibility can hurt (and kill) people of color. We can compensate for our epistemic prejudices by over-believing members of groups whose epistemic credibility is deflated (Fricker, 2007). We might reduce our implicit biases by interacting and hanging out with folks most unlike us (Fridell, 2008 and Lugones, 2003).[6]

Second-Order Epistemic Oppression: Hermeneutical Injustice

Hermeneutical injustice occurs when a knower is unfairly disadvantaged in her capacities to make sense of an experience. Here, unequal social power relationships skew shared hermeneutical resources that favor dominant groups. Think of it this way, hermeneutical injustice happens when powerful groups colonize the knowing field’s schemata. That is, they assign meaning to phenomenon in ways that reflect their understandings and their experiences of the world, leaving the rest of us to work awkwardly with the conceptual vocabulary they have crafted.  Standard examples of second-order epistemic oppression include instances where new phrases have been created to name those experiences that are difficult to see on the unlevel knowing field. For example, before the terms ‘sexual harassment’, ‘date rape’, and ‘marital rape’ were coined, women had no exact public language to name the one-sided unwanted sexualized attentions they received at work, on dates, and in their own bedrooms.  This does not mean that women were struck silent before the experiences were named. It means that our experiences were obscured from the collective understanding (rendered unintelligible) due to gaps in shared epistemic resources (Fricker 2007, 155).  If there is a cultural assumption that marriage grants a husband unlimited sexual access to his wife, then ‘marital rape’ is an oxymoron.

Hermeneutical injustice is also reducible. It is an expected epistemic by product of larger social and political systems, but the nature of the injustice is expressed differently. Unlike testimonial injustice, which is the product of inefficiencies in the credibility economy, hermeneutical injustice is the product of insufficient shared epistemic resources. In other words, the unlevel knowing field’s existing epistemic resources are scanty when it comes to communicating women’s shared experiences with sexual violence. If rape myths and boys-will-be-boys explanations count as the shared epistemic resources for making sense of sexual assault, then they obscure girl’s and women’s experiences with sexual violence. Hermeneutical injustice results from insufficiencies within shared epistemic resources. Changing this requires more than monitoring our biases. To address insufficiencies, Dotson argues that “groups have to be willing to change their ways of thinking and their values” and this requires a conceptual revolution (2014, 14).  So second-order change requires a shift in the deep structures that generate our shared understandings. It requires recognizing that many of the shared epistemic resources we rely upon need revision. Like testimonial injustice reforms can be accomplished by tweaking the existing shared epistemic resources.  If there is no ‘off-the-shelf’ terminology that accurately captures girls and women’s experiences with sexual violence, then new vocabularies can be coined and circulated until they motivate a shift in meaning that hopefully will become a part of the shared epistemic landscape.

The Importance of Understanding Third-Order Epistemic Oppression

The last section of Dotson’s essay offers readers a preliminary account of the deepest and most resilient form of epistemic exclusion. Dotson’s account of third-order epistemic oppression is less complete than her descriptions of first- and second-order epistemic oppressions.  To be fair, her closing task is just to sketch out this basic idea and to connect the dots as best as she can and her sketch is clear enough to meet the goal of bringing the irreducible features of third-order epistemic exclusion into focus.[7]

Dotson’s attempt to pull third-order epistemic exclusion from the unlevel knowing field is important for a variety of reasons. First, highlighting the irreducible features of third-order epistemic oppression reveals the limits of Fricker’s earlier account of epistemic injustice. Fricker clearly articulates the ways epistemic agency is compromised on the unlevel knowing field, but her work fails to engage deeply the resilience of the field itself. In other words, the lenses through which she examines the unlevel knowing field are fashioned from the very underlying schemata that gives rise to the epistemic exclusions in the first place. Fricker’s account of epistemic injustice is confined to cases that are reducible to social and political factors. It remains neutral with respect to the ways underlying schemata uphold and preserve both the insufficiency and inadequacy of shared epistemic resources.

Complete epistemic liberation, however, does not lie in making endless corrections to first- and second-order epistemic exclusions. You can invent tool after tool, alter behavior after behavior, and apply strategy after strategy to remedy testimonial and hermeneutical injustices.  But in the end, like the farthest left-fettered person in Dotson’s example, your testimonies and gap-filling hermeneutical solutions may never gain the momentum necessary to counter the unlevel knowing field’s deep epistemological resilience.[8] This resilience is so powerful that even the most thoughtful revelations and strategic moves are readily absorbed into an epistemological system in ways that leave little or no trace of their impact.  Reforms made from within the unlevel knowing field, in Audre Lorde’s immortal words, may “temporarily beat [the master] at his own game, but they will never enable us to bring about genuine change (1984, 112). The biggest obstacle to epistemic liberation (the one Fricker misses) is that our shared epistemic resources are themselves inadequate for understanding their inadequacy.[9] Epistemological systems contain the seeds of their own preservation: the means for preserving and legitimating inadequate epistemic resources are built into the epistemic system itself.  Understanding this point requires piecing together a clear account of the irreducible nature third-order epistemic exclusions and this is the heart of Dotson’s task.

Why Third Order Change Requires A Broader Account of Epistemic Resources:

Dotson’s attention to third-order epistemic oppression offers readers a deeper understanding of the unlevel knowing field. Recall that unlike testimonial and hermeneutical epistemic oppressions, third-order forms are ‘irreducible’: that is, the source of their resilience cannot be explained as the simple by-products of social and political factors. This does not mean that third-order forms are apolitical. What makes third-order epistemic oppression irreducible is not the absence of epistemic power (2014, 18). The unlevel knowing field is always politically saturated. The difference between reducible and irreducible epistemic oppression does not lie in the presence or absence of social or political influences. It concerns “the character of the resistance to change, or in other words, differing causes of inertia (2014, 3).   So, the sheer force of resistance is what makes these resources inadequate for taking up the task of identifying the epistemic exclusions that perpetuate epistemic oppression in the first place.

Dotson’s first pass at this concept offers readers enough insight to bring third-order changes into focus. The basic insight here is that third-order change requires taking a bird’s eye view of the unlevel knowing field, cultivating an awareness of its parameters, becoming attentive to how the underlying schemas orient our perspectives, and grappling with the resilience of the field itself.  The field’s limitations are revealed in moments. We sometimes get an occasional glimpse of worlds beyond those the unlevel knowing field grants us daily access. Dotson’s fettered persons have the ability to detect something larger about the social world.  Extending Plato’s allegory, she explains, “the bound persons have always had some indication that there exists a fuller world than Shadow land epistemic resources may allow.” In the Shadow land, the farthest left-fettered person “has the ability to detect something about the larger social world none of the other members can detect in quite the same way…. she knows that her experience is unique and that it indicates a larger cave than is immediately apparent” (2104, 16).

I treat the sketchiness of Dotson’s first pass as a virtue. When concepts are sketchy their borders and foundations have not solidified and there is room to shift our terrain, to invite new voices, expand our resources, or begin anew. But, here’s my concern. If the epistemic resources of the unlevel knowing field are inadequate for understanding their own inadequacy, then we would do well to think broadly and creatively about what kinds of resources would be most useful for detecting and responding to deep forms of epistemic oppression. Yet Dotson’s orders-of-change approach relies almost exclusively on cognitive resources and responses.  Bartunek and Moch’s  (1994) original discussion of third-order change offers a broader account of the third order and I wonder why Dotson has not engaged the fullness of their account. Bartunek and Moch compare frame-busting epistemic experiences with the mystical. They account for and discuss the affects epistemic shifts have on epistemic agents’ creativity and daily activities.  They also raise the importance of attending to the emotional confusion that accompanies cognitive limitations.  Change agents, they explain, “must be responsive to managing a number of very difficult feelings, such as anger, a strong sense of loss, anxiety or hopelessness…”(1994,38). Why doesn’t Dotson engage these aesthetic and affective resources? If the goal of third-order epistemological change is to “throw large portions of one’s epistemological system into question,” and if resistance to change is as deep as it is tenacious, then we need to consider every single resource we have at our disposal to see, understand, engage, and foment third-order change. My own experience has taught me that you can’t always think, write, or argue your way around intersecting oppressions. Sometimes you need to sing, chant, dance, witness, cry, pray, laugh, read or write poetry, or seek out precious resources from non-dominant knowing fields that offer glimpses of worlds beyond the unlevel knowing field.

Disrupting third-order epistemically oppressive systems demands creativity. These systems are irreducible, so  responses that point to and engage the political origins won’t work here. What would happen if we  thought broadly, imaginatively, and courageously about the epistemic resources available to us? We may find that our understandings of third-order oppression and change are enriched when we come at them through performance art, spoken word, mystical insights, mindfulness about affective resources, or by applying non-Western cosmologies and epistemologies critically to the unlevel knowing field. Audre Lorde, for example, treats poetry as a source of illumination.  It is through poetry, she says,  that “we give name to those ideas which are—until the poem—nameless and formless, about to be birthed, but already felt” (1984,36). Poetry is a means of processing and naming. It is a means of touching new ways of knowing and revealing new pathways to the worlds we occasionally glimpse.  Since the epistemological resilience at this third level runs strong and deep, it makes sense for us to broadly consider the possibility that there may be affective, creative, or other other non-cognitive responses to epistemic oppression.  I’m curious what it might be like to engage third-order oppressions using these resources.


Bartunek, Jean M. and Michael K. Moch. “Third-Order Organizational Change and the Western Mystical Tradition.” Journal of Organizational Change Management 7, no. 1 (1994): 24-41.

Dotson, Kristie. “A Cautionary Tale: On Limiting Epistemic Oppression.” Frontiers: A Journal of Women’s Studies 33, no. 1 (2012): 24-47.

Dotson, Kristie. “Conceptualizing Epistemic Oppression.” Social Epistemology 28, no. 2 (2014): 115-138.

Fricker, Miranda. “Epistemic Oppression and Epistemic Privilege.” Canadian Journal of Philosophy 25 (1999): 191-210.

Fricker, Miranda. Epistemic Injustice: Power and the Ethics of Knowing. New York: Oxford University Press, 2007.

Fridell, Lorie A. “Racially Biased Policing: The Law Enforcement Response to Implicit Black-Crime Association.” In Racial Divide: Racial and Ethnic Bias in the Criminal Justice System, ed. Michael J. Lynch, Britt Patterson, and Kristina K. Childs. Monsey, NY: Criminal Justice Press, 2008: 39–59.

Lugones, Maria C. and Elizabeth V. Spelman. “Have We Got a Theory for You!: Feminist Theory, Cultural Imperialism, and The Demand for ‘The Women’s Voice.” Women’s Studies International Forum 6, no. 6 (1983): 573-581.

Lugones, Maria. “Playfulness, ‘World’ Traveling and Loving Perception.” Pilgrimanges/Peregrinajes: Theorizing Coalitions against Multiple Oppressions. Lanham, MD: Rowman and Littlefield, 2003: 77-103.

Lorde, Audre. Sister Outsider: Essays and Speeches by Audre Lorde. Freedom, CA: The Crossing Press., 1984.

Mills, Charles W. The Racial Contract. Ithaca: Cornell University Press, 1997.

[1] What I’m calling the ‘unlevel knowing field’ is my shorthand metaphor for the complexities of the epistemological landscape Dotson highlights with her use of Plato’s cave allegory. For the purpose of my comments, I was looking for a more direct way to communicate the basic problems Dotson gets at with her  allegory without getting too caught up in the details of the cave.

[2] I’m using the term ‘twilight zone’ to refer to an undefined or intermediate conceptual area where there are insufficient or inadequate epistemic resources. The intermediate nature of this space means that epistemic resources, in the end, are not really shared as much as one would think.

[3] In an earlier piece, Dotson (2012) builds on Fricker’s testimonial and hermeneutical injustice to offer examples of what she then called ‘First-Order and Second-Order Epistemic Injustice.’ In this piece Dotson’s replaces the epistemic injustice terminology with an epistemic oppression (or exclusion) terminology. I believe this is motivated by her belief that Fricker’s injustice-centered framework is too inflexible; that is, she does not consider that the epistemic injustice’s roots are in the closed epistemic framework itself.  She adds third-order epistemic injustice (contributory injustice), to point to the closed conceptual character of epistemic systems.

[4] This is Fricker’s example (2007), p. 23-29.

[5] See his essay, “Colleges become the Victims of Progressivism.” The Washington Post. [15 August, 2014]. Available:

[6] The implicit bias driving testimonial injustice is not just about misjudging character: people’s lives are at stake. For example, a very strong implicit bias related to policing is the regularity with which white officers associate people of color with crime.  Lorie Fridell’s discussion of a “Fair and Impartial Policing” curriculum, suggests that police departments can change their relationships to the communities they police by training officers to recognize their biases. She draws on the basic principles of ‘contact theory,’ the idea that the more you interact with individuals who are different from you, including groups that you hold the most prejudices against, the greater your reduce your conscious and unconscious biases.

[7] Dotson, in conversation, 14 August, 2014.

[8] To get at the depth of epistemic resilience, I like to imagine that the unlevel knowing field has a magnetic field that allows some first- and second- order revisions to stick, but repels any attempt to shift meaning deeply.

[9] As Dotson remarks, “The reason one must look to the epistemological systems to understand the exclusion that follow from inadequate shared epistemic resources is because those resources, themselves, do not shed light on why they are incapable of accounting for the farthest left-fettered person’s insight” (2014,17).

Categories: Critical Replies

Tags: , , , , , ,

2 replies

  1. When I started to read this article, I was preparing to attempt a technical critique of the conceptual approach as uni-dimensional, specifically that of high power being the unstated characteristic of those who oppress those with less. I was assuming good faith by the writer, that this bias was an error that I may presume to influence.

    Among the references was # 5 described in the text as : Or, consider columnist George Will’s recent claim that women cry rape so that they get “special privileged survivor status.”[5]
    These cases illustrate how the epistemic agency of knowers is compromised by a credibility deficit. If we think about epistemic credibility as a resource, then it is a resource that is unevenly distributed along gendered and racialized lines.

    I double checked the article and while this paraphrase may be the writers conclusion, it was not what the article said. The quote, meaning exact phrase as written, does not exist in the article. This is not an error, but rather evidence that rather than an epistemic community following precise norms of discourse, there is an ideological element that allows breaches of such clear academic norms as precision of quotations as long as it is advances the consensus of the group.

    The quote reflected an article focusing on the individual woman, while the actual article was focused on a movement codified by federal law, translated into university regulations. (the extensive comments were balanced and analytical) It happened to have great validity in my view, and yet was so demeaned by the this writer, Dr. Baily, that she felt safe to debase Mr. Will’s precise criticism as being a slur on the weaker gender. Ironically, among the academic readership of this article only someone outside of academia, myself, bothered to note the inaccurate quotation.

    The distortion of this article is that it is based on a proffered truism that epistemic knowledge is an accretion provided by formal education so that the apex must be of those with certification of having mastered this at the highest level. It ignores the possibility that a bi-product of such advanced knowledge may be a worldview that like all such “true believing” communities is defined by a self interest that may transcend the pecuniary. The writer and the readers of this academic site are comfortable in a distortion of George Will’s article, since he is among the group that is not respected, Thus confirmation bias is built into the dialog, when the actual words of the enemy do not suffice, no one of the “true believers” will challenge a minor distortion of a quotation. The message in a given article by “the enemy” is comfortably used to solidity one’s membership in a value system that is assumed to be “progressive” tacitly understood as moving to an idealized perfection that is only impeded by conservatives.

    Epistemic “blindness” is universal, and not a product of “oppression” which is a value laden concept that implies a movement of good against evil, and erroneously exclusively equates the “power to oppress” as being such “evil.” I conclude with my own hyperbole, which is that the world view of this article contains a fatal flaw, one that is no more enlightened than that which underlies the most “regressive” movements in the history of civilization. As a most relevant example, while this “enlightened progressivism” currently espouses the celebration of same gender sexuality, it ignores serious investigation of the underlying reasons why other subcultures, African American men for instance and Muslim cultures for another, find it abhorrent. The reality that for the male of the species sexuality has a large element of expression of dominance, with concomitant oppression of the more subordinate, and that this is hard wired into our primate brain is not evaluated, much less acknowledged. . The tools of sociology, ethology and neuroscience that could investigate and possibly support this view are quietly suppressed to the point that this conversation does not even exist. Thus, a perfect example epistemic blindness among the academic social and power elite.

    Academic evaluation of how we know, how we think, “epistemology” — is worthy endeavor that deserves society’s support as a vital function that has been entrusted in the institution of academia. This should impart an obligation to its avatars to be alert to their own biases, and then to transcend them. Just as “Nature abhors a vacuum,” “Humans abhor isolation from a community” and doing violence to reason is a small price for us to pay to avoid this fate. If every social science is to become simply another cult masked in the very pseudo-precision of its arcane terminology, the loss will be much greater than to its practitioners, but to us all.

  2. Wills chose to see an epidemic of fake oppression, nit- and cherry-picking in order to ignore the actual substance of what’s at stake here. The report that Wills mocks is one of two dozen that the reporter examined for the newspaper article. Wills is happy to dismiss a woman’s report of sexual assault, dismiss her pain, and dismiss the documented complicity of the school administration in a pattern of protecting perpetrators of rape and sexual assault over a number of years. The title of the article, “Rape Happens Here,” is amply evidenced in the text. That Wills can read the article and only see a fake victim narrative provides a good case study for the epistemic oppression that this article examines. Your comment, Al Rodbell, provides another.

Leave a Reply