I wrote this text as a part of the examination on a PhD course I took on Immanuel Kant’s moral philosophy. Basic knowledge is presumed.
“Many contemporary proponents of ‘Kantian’ ethics,” O’Neill writes, “want the nicer bits of his ethical conclusions without the metaphysical troubles” (p. ix). In Constructions of Reasons (Cambridge University Press 1989), she shows how many of our time’s moral issues can be treated within Kant’s own theoretical framework. Kant’s accounts of reason, action, and freedom are not “metaphysical extravaganza,” and his moral theory is “neither pointlessly empty nor relentlessly nasty” (ibid). O’Neill argues that philosophy “must begin with the task of showing why any standards or procedures for orienting our thinking should have authority for us and count as standards of reason” (ibid). The Categorical Imperative (CI) is Kant’s answer to that task. According to O’Neill, the CI is central not only to Kant’s ethics but to his whole philosophy.
Constructions of Reasons is divided into three parts. The first part, “Reason and critique,” offers “an antifoundationalist, constructivist account of the Kantian enterprise” (p. x). In the second part, “Maxims and obligations,” O’Neill tries to show that “the Categorical Imperative is basic not only to reason, but to action and to ethics as well” (ibid). The essays in the last part, “Kant’s ethics and Kantian ethics,” explore “differences between Kant’s ethics and some recent, would be Kantian ethical positions” (p. xi).
In this short review, I comment briefly on one chapter from each part of the book. Thereafter, I discuss (just as briefly) O’Neill’s Kantianism, and how it is oriented around the notion of universalization.
In chapter 2, “The public use of reason,” O’Neill offers a vindication of liberal toleration. It is a common view in liberalism that a diversity of beliefs and expressions should be tolerated, and that toleration requires only noninterference. If, for instance, a person’s freedom to express her opinions is not interfered with, that person’s freedom of expression is tolerated. O’Neill opposes this view: “We do not tolerate others’ communications if we are merely passive and noninterfering” (p. 31). Among other things, real communication fails to be public if it fails to be interpretable by others. “A genuine debate needs some mutual comprehension” (p. 41). O’Neill argues that reason’s authority and toleration are interdependent. A liberal society should develop rational practices for public use of reason that its members are subjected to in public discourses. “Practices of toleration help constitute reason’s authority” (p. 39).
In chapter 6, “Between consenting adults,” O’Neill explores what it is to treat another person as means to an end. “Merely not to be used is not enough for being treated as a person” (p. 105). The most basic notion explored in the chapter is that of consent. If a person consents to an action, she is treated as a person. O’Neill offers three preliminary points that distinguish between real consent and spurious or morally trivial consent. First, every action can be described in different ways. Under some act description, a person will be wronged although she has given her consent. Therefore, morally significant consent is to “the deeper or more fundamental aspects of another’s proposals,” which—presumably—are identified by the CI (p. 110). Second, real consent can only be given by someone who has the possibility to both consenting and dissenting (pp. 110–1). Third, “we need to understand what makes genuine consent to the more fundamental aspects of action possible” (p. 111). It is unclear how the third point is different from the first two in conjunction. The upshot of the argument in the chapter is that persons should not be treated abstractly “ideal” human beings, but as particular individuals with capacities and characteristics that are unique for them. To fully treat another person as a person is to act on maxims that share her ends (p. 115).
In Chapter 11, “Constructivism in ethics,” Rawls’s constructivism is discussed as “a route into an account of a variant constructivism that may be less likely to be absorbed either into moral realism or into relativism” (p. 206). Rawls’s method of reflective equilibrium is known to be able to yield multiple solutions to moral dilemmas. Therefore, critics have objected that its constructivist foundation allows for moral relativism. O’Neill argues that the relativism is due to Rawls’s instrumentalist conception of rationality. Rawls’s ideal of a person is abstract, not real. A more Kantian constructivism, O’Neill proposes, starts from the least determinate conceptions of the rationality of agents (p. 212). She suggests that, behind the veil of ignorance, the question should be: “What principles can a plurality of agents of minimal rationality and indeterminate capacities for mutual independence live by?” (p. 213). Constructivism, O’Neill argues, must “look for a set of entitlements that can be consistently held by all,” where the notion of “all” does not denote ideal abstractions (p. 214).
The sole guiding principle in Constructions of Reasons is universalization. In all matters, O’Neill argues that we should make and follow maxims that can be universalized. This is sometimes expressed as “consistency” or “coherence.” In chapter 5, for instance, O’Neill distinguishes between conceptual and volitional consistency, in defense of the CI: “The interest of a Kantian universality test is that it aims to ground an ethical theory on notions of consistency and rationality rather than upon considerations of desire and preference” (p. 103). And, in chapter 7, she argues that three different formulations of the CI— the Formula of Universal Law (FUL), the Formula of the End-in-Itself (FEI), and the Formula of the Kingdom of Ends (FKE)—are coherent: The “FUL and FEI are complementary and FKE combines both” (p. 144).
But, universalization is central to O’Neill also in that she favors an ontology of human beings that is non-ideal. The ideal image of a human being (as, for instance, an ideally rational creature) is just that: Ideal. But no person is ideal. Instead, she persistently argues that ethics is for real people. Ideal ontology is not universalizable; only a non-ideal ontology is that. Therefore, universalization—essentially, the CI—is central to O’Neill’s thinking on three levels: In how to do philosophy (method), in how philosophy’s object should be understood (ontology), and in practical matters (ethics). Philosophers must choose “between standards of reason that can govern both practice and theory, and disoriented consciousness” (p. x). Thus, the CI is the standard for both thinking and acting.