As part of a self-led course on liberal feminist philosophy, I accidentally started reading Situating the Self: Gender, Community, and Postmodernism in Contemporary Ethics, by Seyla Benhabib. I say accidentally because feminism isn’t really the focus of the book. I have found myself instead knee-deep in “discourse ethics” – the book’s main subject.
It’s easiest to describe discourse ethics in contrast to social contract ethics, where we envision some ideal scenario in which ideal individuals come to some kind of ideal agreement about what constitutes justice. (Social contract ethics typically focuses on justice rather than the broader topic of what constitutes the good life). To the social contract theorist, justice is whatever we could agree to if we were perfectly rational and perfectly objective. In a passage that would have been incredibly helpful had it appeared on page 2 instead of page 169, Benhabib trots out the differences between this approach and discourse ethics.
Both the Rawlsian “original position” and the Habermasian model of “discourse ethics” are idealizations intended to make vivid to us the ideal of impartiality or of what it means to assume the moral standpoint. Their differences center around the following points. According to discourse ethics, the moral standpoint is not to be construed primarily as a hypothetical thought process, carried out singly by the moral agent or by the moral philosopher, but rather as an actual dialogue situation in which moral agents communicate with one another. Second, in the discourse model no epistemic restrictions are placed upon moral reasoning and moral disputation, for the more knowledge is available to moral agents about each other, their history, the particulars of their society, its structure and future, the more rational will be the outcome of their deliberations. Practical rationality involves epistemic rationality as well, and more knowledge rather than less contributes to a more informed and rational judgment. To judge rationally is not to judge as if one did not know what one could know (the effect of hanging the “veil of ignorance”), but to judge in light of all available and relevant information. Third, if there are no knowledge restrictions to be placed upon such an argumentative situation, then it also follows that there is no privileged subject matter of moral disputation. In the discourse model, moral agents are not only limited to reasoning about primary goods which they are assumed to want whatever else they want. Instead, both the goods they desire and their desires themselves can become subjects of moral disputation. Finally, in such moral discourses agents can also change levels of reflexivity, that is they can introduce metaconsiderations about the very conditions and constraints under which such dialogue takes place and they can evaluate their fairness. There is no closure of reflexivity in this model as there is in the Rawlsian one. (pp 169)
Emphases in original. There are three ideas here that I’d like to discuss in turn: (1) actual versus hypothetical dialogue; (2) universal particularity versus blind universalism; and (3) the vague rather than strict boundary between the right and the good.
Actual versus Hypothetical Dialogue
The whole point of thinking about justice hypothetically in terms of idealized agents coming together to hammer out the principles of justice once and for all is that this way you can actually reach an agreement and come to some conclusions. Of course, in real life we can never get any group of appreciable size and diversity to agree on anything at all. The consensus achieved by idealized agents is illusory because (1) we ultimately have to return to the world of flesh and blood to test out the ideas arrived at behind the veil; and (2) meaningful differences between the contractors have been idealized away to such an extent that the contractors are identical, so it’s no consensus at all, except that of the philosopher with themself. I’ll save (2) for the next section.
As to (1), the veil of ignorance – whereby we ponder what folks might agree to about how society is organized if they didn’t know what their own place in that society would be – is a useful thought experiment. It challenges us to think objectively, and it can yield valuable insights. But upon lifting the veil, the conclusions must still be justified to real people who can argue back. This becomes evident when you see the real world criticism by other philosophers when a theory of justice derived from ideal theory is published. This might seem like a trivial point, but I think it matters that at the end of the day, there’s no escaping engaging with real folks who keep raising good objections regardless how objective we think we’re being.
Benhabib’s alternative approach of discourse ethics moves the emphasis from the agreement to the process of getting to that agreement and explicitly eschews the requirement to actually reach universal agreement.
We must interpret consent not as an end-goal but as a process for the cooperative generation of truth or validity. The core intuition behind modern universalizability procedures is not that everybody could or would agree to the same set of principles, but that these principles have been adopted as a result of a procedure, whether of moral reasoning or of public debate, which we are ready to deem “reasonable and fair.” (pp 37)
Disagreement and therefore politics are brute facts of our social nature. The unhappy fact is that sometimes we will of necessity – when the stakes are high enough to warrant it – have to coerce cooperation in order to reap the rewards of social coordination. We can’t avoid this, but we can try to make the social and political decision procedures as just and as robust as possible.
Universal particularity versus blind universalism
As I said above, consensus behind the veil is really the consensus of the contract theorist with themself. This criticism has been lobbed at Rawls from multiple angles, including from feminists and non-contractarian liberals. If we abstract away all our socially defining features and our substantive commitments, we erase the particular, vivid experience of, for example, what it’s like to be a member of a marginalized ethnicity or religion, or a woman in a society where men have been calling the shots for centuries. Individual experience, in all its granular particulars, is a requisite epistemic resource for any meaningful discourse on justice.
The conditions for a just and robust discourse are these:
(1) that we recognize the right of all beings capable of speech and action to be participants in the moral conversation – I will call this the principle of universal moral respect; (2) these conditions further stipulate that within such conversations each has the same symmetrical rights to various speech acts, to initiate new topics, to ask for reflection about the presuppositions of the conversation, etc. Let me call this the principle of egalitarian reciprocity. (pp 29)
Emphases in original. Everyone can speak, raise issues, and contest common assumptions and narratives. And each person owes it to others to listen with charity.
The right and the good are distinct, but not cleanly separable
Liberal neutrality is a critical feature of a peaceful civilization, as it prohibits those in power from enforcing their visions of the good life onto their dissenting fellows, thus eliminating a source of irreconcilable conflict. But liberal neutrality shouldn’t be construed to shield conceptions of the good from critical inquiry. We derive the same epistemic advantages from opening our conceptions of the good to outside critique as we do from opening our theories of justice to critique.
If in discourses the agenda of the conversation is radically open, if participants can bring any and all matters under critical scrutiny and questioning, then there is no way to predefine the nature of the issues discussed as being public ones of justice versus private ones of the good life. Distinctions such as between justice and the good life, norms and values, interests and needs are “subsequent” and not prior to the process of discursive will formation. As long as these distinctions are renegotiated, reinterpreted and rearticulated as a result of a radically open and procedurally fair discourse, they can be drawn in any of a number of ways. (pp 110)
Justice and theories of the good shouldn’t be strictly separated because this is likely to implicitly privilege status quo power relations to the detriment of the marginalized. Without critical challenge from diverse perspectives, these relations can easily be understood as natural or inevitable, rather than socially constructed and changeable. An example of this is the common understanding throughout pre-feminist history that domestic family matters are private and not open to public scrutiny, and further that a woman’s natural role is within the domestic sphere while the man’s natural role to engage in civic life.
As an aside, I suspect one can fruitfully compare this ethical approach to Hayek’s defense of freedom as a discovery process: the trusting of “the independent and competitive efforts of many to induce the emergence of what we shall want when we see it.” (Constitution of Liberty, pp 29) For both Hayek and Benhabib, the core problems of living together are the epistemic ones of effectively harnessing dispersed knowledge.
Reflective equilibrium and discursive disequilibrium
Reflective equilibrium is the process by which we achieve harmony between our moral intuitions and our rational theories of ethics. We reason about ethics in order to construct some kind of machinery we can trust to guide us through novel moral terrain. We calibrate our theories against our intuitions in well-understood cases, and we go back to the drawing board if our theories throw up catastrophic moral horrors for these intuitive cases. And we go back and forth, sometimes becoming so confident with our theoretical machinery that we accept its verdict in central cases over our intuitions.
Ethical discourse extends this procedure beyond the individual to the moral community (construed as the whole of humanity in some cases). We must strive not only reach agreement within ourselves, but with our fellows. But moral pluralism is a brute political fact, and the socio-moral landscape is continually evolving as cultural, political and market arrangements, technology, and scientific understanding all change with the times. New critiques and new ways of relating to one another are always emerging. We’re thus always in a state of disequilibrium, seeking consensus where none is actually possible. On this view, moral understanding is a continual process and not something that can ever be figured out once and for all.