We agree that the self is constructed through a collaborative dialog. But hostile interlocutors could use various cognitive techniques to hijack the dialog, resulting in beliefs, values, and even selves that are out of line with reality. The implications of this problem are dire, but we suggest that increased metacognitive awareness could help guide this process to a truthful conclusion.

Doris suggests thought-provoking directions for rehabilitating moral agency within a self that is unaware and incoherent. These directions suggest more radical proposals. First, moral reasoning may serve many different functions beyond merely expressing a person's values. Second, social collaboration may not focus on moral reasoning as much as it does on the “defeaters” of that agency. Ultimately, moral agency may not reside in the individual but in social communities or within external situations.

Doris (2015b) develops a theory of moral agency to avoid a skeptical challenge arising from psychology studies indicating that (im)moral behavior is caused by trivial situational factors. His theory is flawed in attending only to situational influences on behavior and neglecting individual differences such as moral identity and virtue. A focus on individual differences in resilience to influence from trivial situational factors defangs the skeptical challenge and offers a better account of moral agency.

We welcome Doris's dual systems, social account of agency and self. However, we suggest that a level of affordances regarding agency is interpolated between those dual systems. We also suggest a need to consider joint (“we”) agency in addition to individual (“I”) agency, and we suggest a more fundamental role for culture in configuring both the values entering the dialogue that generates the sense of agency and self, and the nature of the dialogue itself.

Although anti-reflectivism seems to preclude a role for reflection, this dichotomy could be synthesized in a Piagetian developmental framework. Development integrates a role for error and ignorance in reflection, and supports Doris's espousal of valuation, collaboration, and pluralism, and the importance of extrinsic factors to the self.

This commentary extends Doris's approach of agency by highlighting the importance of responsibility attributions by observers. We argue that (a) social groups determine which standards are relevant and which actors are responsible, (b) consensus about these attributions may correct individual defeaters, and (c) the attribution of moral responsibility reveals agency of observers and may foster the actors' agency.

The cases that Doris chronicles of confabulation are similar to perceptual illusions in that, while they show the interstices of our perceptual or cognitive system, they fail to establish that our everyday perception or cognition is not for the most part correct. Doris's account in general lacks the resources to make synchronic assessments of responsibility, partially because it fails to make use of knowledge now available to us about what is happening in the brains of agents.

We offer two kinds of constructive criticism in the spirit of support for Doris's socially scaffolded pluralism regarding agency. First: The skeptical force of potential “goofy influences” is not as straightforward as Doris argues. Second: Doris's positive theory must address more goofy influences due to social processes that appear to fall under his criteria for agency-promoting practices. Finally, we highlight “arms race” phenomena in Doris's social dynamics that invite closer attention in further development of his theory.

John Doris argues that, when behaviors are caused by processes that we would not endorse, our agency is defeated. I argue that this test for defeaters is inappropriate. What matters is not what we would but what we should endorse. The subpersonal mechanisms he identifies as defeaters enable us to track and respond to reasons. They realize agency, rather than defeating it.

I question whether psychological effects that an agent is unaware of can express her values and, if they can, whether this allows us to hold her responsible in the range of cases that we would like to.

In this commentary, I highlight the importance of a proper discussion of the pragmatic implications of John Doris's paradigm for allocation of personal responsibility proposed in his new book Talking to Our Selves. By employing some classic concepts of the American common law tradition, I discuss why Doris's valuational understanding of agency fails to provide an adequate framework for moral responsibility, social accountability, and legal liability.

Doris argues that our choices are heavily influenced by forces that we wouldn't count as genuine reasons. This unsettling conclusion is motivated by a debunking argument so wide-ranging that it isn't foisted upon us by the sciences. Doris sometimes seems to lower his ambitions when offering instead a skeptical hypothesis argument, but that conflicts with his aims in the book.

This comment introduces the interactionist perspective on reason that Dan Sperber and I developed. In this perspective, reason is a specific cognitive mechanism that evolved so that humans can exchange justifications and arguments with each other. The interactionist perspective significantly aligns with Doris's views in rejecting reflectivism and individualism. Indeed, I suggest that it offers different, and maybe stronger arguments to reject these views.

Doris argues that an agent is responsible for her behavior only if that behavior expresses (a relevant subset of) the agent's values. This view has problems explaining responsibility for mistakes or episodes of forgetfulness. These problems highlight a conceptual problem with Doris's theory of responsible agency and give us reasons to prefer an alternative (non-valuational) theory of responsible agency.

Doris proposes that the exercise of morally responsible agency unfolds as a collaborative dialogue among selves expressing their values while being subject to ever-present constraints. We assess the fit of Doris's account with recent data from psychology and neuroscience related to how people make judgments about moral agency (responsibility, blame), and how they understand the self after traumatic events.

Two misguided ideas dominate philosophical thinking on moral responsibility: (1) the idea that it obviously exists, and (2) the idea that even if it does not, it is nevertheless needed for the society to function properly. In his book, Doris (2015b) discusses the first illusion, while uncritically accepting the second. In this commentary, I question the utility of such endeavors.

We characterize Doris's anti-reflectivist, collaborativist, valuational theory along two dimensions. The first dimension is social entanglement, according to which cognition, agency, and selves are socially embedded. The second dimension is disentanglement, the valuational element of the theory that licenses the anchoring of agency and responsibility in distinct actors. We then present an issue for the account: the problem of bad company.

I argue that John Doris should apply his dialogic or collaborationist approach to agency more fully to questions of moral responsibility. To do so, he must discard his form of pluralism that aims to accommodate a variety of theoretical approaches to responsibility in favor of a pluralism that rejects theorizing about responsibility altogether.

This commentary has two aims. First, I raise a practical challenge for accounts of responsible agency: Provide empirically informed strategies for enhancing responsible agency so that actors can become more resistant to the influence of defeaters. Second, I offer an initial sketch of a solution to this practical challenge. My solution is supported by empirical evidence suggesting that responsible agency can be enhanced via self-regulatory strategies (expertise and implementation intentions).

We describe additional research that expands upon many of Doris's points, focusing on collaboration (Ch. 5), selves, and identity (Ch. 8). We also suggest some elaboration of his treatment of dual process theories (Ch. 3). Finally, we ask whether collaborationist accounts confer logical consistency.