Skip to main content

Open for Debate

Socializing Intellectual Autonomy

10 August 2020

Two key ideas which frame our thinking about autonomy are self-governance and self-reliance. The problem with using these arguably modern ideas to frame our thinking about autonomy is that it can easily make it look like autonomy and community are in tension; that securing one’s autonomy means withdrawing from one’s community.

These modern ideas about autonomy also suggests a certain way of thinking about intellectual autonomy, one according to which the autonomous thinker regulates their own thinking (intellectual self-governance) and doesn’t rely on the epistemic efforts others (intellectual self-reliance).[1]

It’s easy to see that this idea of intellectual autonomy is at odds with social epistemology. Social epistemology tells us that, just as we are reliant on others for our material well-being, so too we are reliant on others for our intellectual well-being. A lot of the knowledge and true belief that matters to us comes to us from others [see Emily Sullivan’s Depending on Others for Knowledge].[2] Think, for example, of all the scientific, historical, and current event beliefs that you have. It’s hard to see how you might have gotten any of those beliefs—let alone good reason to adopt them—without trust and reliance on others.[3] And this isn’t simply some unideal fact of life. For it’s hard to see how you could, even ideally, get such knowledge through your own intellectual efforts, without relying on the past or present intellectual labor of other people. (Intuitively, not everyone can jointly be an anthropologist, an historian, a mathematician, a psychologist, an investigative journalist, and so forth). So this way of conceptualizing the ideal of intellectual autonomy seems to be at odds with getting knowledge from others and, in this way, at odds with social epistemology. It is no wonder, then, that intellectual autonomy gets a bad wrap in contemporary epistemology where social concerns are increasingly (and rightly) taking center stage.

Rather than relegate intellectual autonomy to the dustbin of bad philosophical ideas, I think we should instead rethink what intellectual autonomy could be in an increasingly connected world. This task doesn’t call for reengineering the concept as such but rather drawing upon ideas that, for the most post at least, have been neglected by contemporary epistemologists (at least in this context). These are ideas outside the liberal paradigm: ideas from feminist, critical and leftwing social theory.

  1. Intellectual Autonomy: Two Ideas

Let’s compare two broad ways of thinking about intellectual autonomy. The first way says that only the internal is what counts: that critical self-reflection and belief-management is what makes one’s thinking or inquiry autonomous. It is to manage and rely upon that which is your own. What someone tells you, or the inquiry one recommends to you, stops with you. You need to consider it, to reflect on it; not doing that is failing to be autonomous. Simply trusting another person is to flout one’s autonomy.

Crucially, this idea doesn’t take the social world to make any essential difference to one’s intellectual autonomy. If the world doesn’t give you labs or other people to talk to, this is not autonomy reducing because whether there are labs or people willing to inquiry with you is not (not normally) something within your internal, executive control. It is not something you can, as it were, impose on yourself in order to guide your thinking.

The second way says that what’s external counts. There is a weaker and stronger version of this idea. The weaker one says that our capacity to be autonomous depends on being socialized and, crucially, how we are socialized.  For example, this is what lies behind that the thought that “what actually enables people to be autonomous …. is not isolation, but relationships—with parents, teachers, friends, loved ones” (Nedelsky 1989, 12). So this thought takes social relationships to be important for intellectual autonomy by enabling it. A stronger version of the idea is that social relationship are not just enabling conditions for intellectual autonomy, but constitutive conditions. That intellectual autonomy requires social relationships, and is not simply caused by it.

I think this idea is basically correct. To see why, consider an interesting case discussed by Marina Oshana:

Subservience: Harriet is a ‘stay at home mom’ in the 1950’s, and she values her role as homemaker. She has a very traditional relationship with her husband, who expects Harriet to cook, clean, raise the children, and do other domestic activities, which is what is ordinarily expected of women in her society. Unlike most women, however, Harriet has a few other options available to her besides being a ‘stay at home mom’, but she has reflectively decided to lead her ‘stay at home mom’-life, having considered seriously the alternatives, each of which were open to her (Oshana 1998, 89).

Harriet’s non-autonomy is inexplicable in individualist terms because “she has the “right” psychology” (Oshana 1998, 89). After all, the way that the world is and the way that she wants the world to be are congruent. No one in fact interferes with the way she wants to live her life. She lives the life the way she wants to live it, unimpeded by other people making her live that way. We might even say: it is her choice. But she still lacks autonomy: “not because she wants to be subservient, but because she is subservient. Her lack of autonomy is due to her personal relations with others and to the social institutions of her society” (ibid). On this picture, then, certain external, structural features of Harriet’s relationship to her husband and her relationships within a broader patriarchal society is what makes it unsuitable for the promotion of her autonomy. Her autonomy requires different relationships—and thereby different social arrangements—rather than having the sort of psychological states which harmonize with her situation.[4]

  1. Intellectual Domination

This goes some distancing to showing that autonomy is essentially relational in character. But how does it translate over to intellectual autonomy?  To show this, I think we need to distinguish between two key ways of thinking about the realization of autonomy. One way says that autonomy requires a certain kind of freedom, the freedom not to be interfered with by others. This is the kind of freedom that Harriet had. In the epistemic domain, we can translate this idea as follows: intellectual autonomy requires a certain kind of intellectual freedom, namely, the freedom to explore your own thoughts or inquiries without outside interference. Call this the non-inference requirement for intellectual autonomy. In Harriet’s case, she could openly explore her ideas with her husband, her friends, and so forth. For society is patriarchal, but luckily for Harriet she has the right friends and family who don’t actively repress her thinking.

A second way of thinking about autonomy requires not simply not being interfered with, but to be in a condition in which no one has the arbitrary ability to interfere with you; otherwise, you are dominated. In the epistemic domain, we can translate this idea as follows: intellectual autonomy requires a certain kind of intellectual freedom, namely, the freedom to think and inquire outside of the arbitrary interests or decisions of others who could impose on you. Call this the non-domination requirement.[5] To this idea in action, imagine a researcher who wants to pursue some project—perhaps it’s something we’d regard as fairly innocuous—but the government has the ability to change or even halt their project at a whim. More specifically, they even have the power to skew the research to fit their ideas. Maybe the government never does this, but they can do it and there is no serious recourse to an appeal—for they’d just defeat it. In this situation, our researcher is not truly intellectually autonomous. This is because their inquiry is dominated. For someone has the arbitrary power to dictate what they should be researching (or how they do it).

This is also true of Harriet’s situation. For her husband could, if he wanted, make her do more chores, see her friends less, and so on. It is ultimately up to his whims. Of course, he might never do this. He might be a beneficent husband, so to speak, and so Harriet would satisfy the non-interference requirement but not the non-domination requirement. Intuitively, we should say: she isn’t free. If we turn our attention to Harriet’s discussions, or her personal educational projects and ideas—perhaps the books she reads or the lectures she attends at a nearby college—her husband could bring that all to an end. Her husband could easily dismiss her questions or tell her how she should be thinking. In such a case, while Harriet might still have the ability to think the way she wants internally, her inability to do so openly would be intellectually dominating. This is because she’s in a society which doesn’t take women seriously as thinkers.

Intellectual domination can also function less directly through ideology. People might not only internalize ideas which dominate their and other people’s behavior but systems of thinking which promote such domination (cf. Mary Douglas 1986, 18). For example, consider the following example discussed by Sharon Krause:

 Ladylike: According to her mother, Sharon walks ‘like a boy’. It’s Sharon’s typically way of walking, but her mother insists that she take smaller, more ‘ladylike’ steps (Krause 2013, 114)

 Krause takes the Ladylike case to reveal the limitations of non-domination requirements on autonomy but I think they help us to see that domination can be at play in intellectually as well as physically coercive cases. It is ideas or ways of thinking which dominate her, not someone’s threat. An ideology dominates by guiding thoughts and patterns of thinking that are socially enforced and obeyed.[6] They have an almost unlimited ability to interfere with the way we think. It is because of the widespread internalization of these ways of thinking that someone could easily say to Harriet that her preoccupation with mathematical physics is ‘weird’, and no one would think anything of it.

Thinking about ideology and how it can have a dominating impact on the way we think can also help us to see the limitations of individualist conceptions of autonomy. Recall that individualist conceptions take it that your thoughts are autonomous insofar as they are self-directed, something you can change or evaluate. More specifically, you are intellectually autonomous only if your thinking is not interfered with by anyone else—at least, not in a way that that doesn’t align with what you actually want for yourself. That fact that you are apt to interfere with your own thinking—rather than other people—is a mark of autonomy. For the critically minded, belief ‘manager’ is apt to do exactly that. But this overlooks the reality that one’s own thought can dominate as well. Imagine a person who’s thinking is littered with ideology. They are unable to grasp ideas or understand arguments without seeing them in the light of their ideology. Even when they think critically, they always surreptitiously shelve their ideological commitments, making it seem to them that they are open-minded people. Their ideology dominates them.

Lao Tzu said the “the mark of a moderate person is freedom from their own ideas”. This reveals what I think is autonomy reducing in the case of ideology: it binds you to certain ways of thinking that aren’t easily constrained. It can also overly influence the way you conduct your inquiries. Imagine a person who is so caught up in being intellectually ‘self-made’ that only their answers to their questions matter to them. They want their epistemic successes to be attributable only to their own intellectual efforts. Determined, as they are, to learn everything for themselves will easily lead them to an impoverished state. After all, by taking it for granted the technology they use to learn what they do — from algebra, history books, or the periodic table of elements, to microscopes, computers, and Python — they forget that they are relying on others epistemically. It is our accumulative epistemic inheritance, the result of humanity’s intellectual efforts.[7] We are all ‘shareholders’ of this epistemic inheritance. When such a person takes it that they learned what they did ‘all on their own’ without ‘anyone else’s help’, they not only think something which is false but disrespectful, for it fails to credit those who are long since passed for their intellectual efforts.



Elizabeth Fricker (2006). “Testimony and epistemic autonomy”, in Jennifer Lackey & Ernest Sosa (eds.), The Epistemology of Testimony. Oxford University Press. pp. 225-253.

Alvin Goldman and Cailin O’Connor, “Social Epistemology”, The Stanford Encyclopedia of Philosophy (Fall 2019 Edition), Edward N. Zalta (ed.), URL = <>.

Sharon R. Krause (2013). “Beyond Non-domination: Agency, inequality, and the meaning of freedom”, Philosophy and Social Criticism, vol. 39, 2.

Peter Kropotkin (1902). The Conquest of Bread. London: Chapman and Hall, LTD.

Jennifer Lackey (2006). “Knowing from Testimony”, Philosophy Compass, vol. 1, issue 5: 43-448.

Frank Lovett (2016). “Non-domination”, Oxford Handbook of Freedom, (eds.) David Schmidtz and Carmen E. Pavel. Oxford University Press.

Jennifer Nedelsky (1989). “Reconceiving Autonomy: Sources, Thoughts and Possibilities,” Yale Journal of Law and Feminism, 1: 7-36.

Charles Mills (2017). “Ideology”, In The Routledge Handbook of Epistemic Injustice, (eds.) Ian James Kidd, José Medina, and Gaile Pohlhaus, Jr. Routledge: pp. 100-111.

Marina A. L. Oshana (1998). “Personal Autonomy in Society”, Journal of Social Philosophy, vol. 29, No. 1, Spring 1998: 81-102.

Linda Zagzebski (2013). “Intellectual autonomy”, Philosophical Issues 23 (1):244-261.




[1] These two key ideas about autonomy are found in the work of Immanuel Kant and Ralph Waldo Emerson, respectively. For their expression in epistemology, see Linda Zagzebski (2013) and Elizabeth Fricker (2006).


[2] See, for example, Goldman and O’ Conner (2019).


[3] See Lackey (2006). She writes: “Were we to refrain from accepting the testimony of others, our lives would be impoverished in startling and debilitating ways” (pg. 432).


[4] It’s important to note that autonomy is not an all-or-nothing feature. One can be more or less autonomous. It is gradient feature. Certain situations might reduce your autonomy without removing it.


[5] See Lovett (2016) for an extensive overview and defense.


[6] See, for example, Mills (2017) pp. 102-103.


[7] This type of argument is inspired by one that Kropotkin (1902) makes for collective ownership. See Chapter 1, §2 and Chapter 13, §4.


Picture by Aarón Blanco Tejedor    on unsplash


1 comment

Comments are closed.