Date of Award

June 2017

Degree Type


Degree Name

Doctor of Philosophy (PhD)




Ben Bradley


ethics, moral character, moral psychology, moral responsibility

Subject Categories

Arts and Humanities


Here is a simple observation about moral character: Moral virtue apparently consists, at least in large part, in caring about the right things. When we imagine a virtuous agent, we find that she cares about particular considerations, and that her caring is at least part of what makes her virtuous. One cannot be fully virtuous, for example, unless one cares at least somewhat about the welfare of others. Here is a corollary: At least sometimes, agents are morally vicious because they do not care about the right things. An agent who just doesn't care whether others live or die should, for example, strike us as severely vicious.

And here, from Hume, is a simple observation about moral responsibility: In order for an agent to be blameworthy or praiseworthy for an action, that action must reflect something about that agent. This observation, too, is supported by common intuitions. Agents seem to be blameworthy when and because their actions reflect something bad about their moral character, and they seem praiseworthy when and because their actions reflect something good about their moral character. And we are generally reluctant to attribute blameworthiness in cases in which circumstances prevent an agent's character from being reflected in his actions – we typically excuse agents whose bad actions result from delusions or uncontrollable impulses, for example.

Here, finally, is an appealing synthesis of these observations. Agents are blameworthy for actions that reflect their moral vices, and moral vices consist, at least in large part, in having the wrong attitudes towards certain considerations. Therefore, it seems that agents are blameworthy when their actions reflect such attitudes. And, since virtues consist, at least in large part, in having the right attitudes towards certain considerations, agents will be praiseworthy for actions that reflect these attitudes. This synthesis is also intuitively plausible. An agent who stands idly by and watches a child drown seems not only vicious in virtue of his indifference to human life, but blameworthy in virtue of the fact that this indifference is reflected in his action. And an agent who makes significant sacrifices to help others is not only virtuous in virtue of her great concern for others, but also praiseworthy when she exercises her virtue.

The preceding observations raise two obvious questions: Which considerations are relevant to virtue and moral worth, and which attitudes are the “appropriate” ones to have towards these considerations? A recently-influential family of views (Arpaly 2002, 2003, 2006; Markovits 2010, 2012, Arpaly and Schroeder 2014a) offers a procedure for answering these questions. The considerations relevant to virtue and moral worth, according to these views, are the considerations that the correct normative theory identifies as relevant to determining the deontic status of an action, and the appropriate attitude towards a particular consideration is determined by that consideration's role as right-making or wrong-making. Thus, a virtuous agent will have positive or pro- attitudes towards those considerations that make actions good or right, and negative or anti- attitudes towards those considerations that make actions bad or wrong. Call accounts of this kind actual good (AG) accounts. A number of considerations count in favor of AG accounts. As noted, they do an excellent job of accommodating several intuitively plausible observations about character and moral worth. They are also equipped to provide intuitively plausible attributions of moral worth in a range of important cases.

But there are additional desiderata for an account of virtue and moral worth. Attributions of moral worth are not merely of theoretical interest but also of practical importance, as they are likely to have implications for which agents we should reward or punish. And while the correct attributions of virtue and moral worth seem to be obvious in some cases, they are not obvious in others. In particular, there are a number of socially, legally, and morally important cases of wrongdoing in which it is not intuitively clear how we should evaluate the agents in question. These include the case of the psychopath; they also include cases of ideologically-motivated agents who act badly as the result of false moral beliefs. Preferably, our account of virtue and moral worth would be useful in guiding our judgments of moral worth in these difficult, real-world cases. Ideally, it would be complete, in the sense that it would offer a generalized procedure for assessing moral worth in all cases: Our account would take the correct normative theory as an input, along with the attitudes reflected in an agent's action, and then act as a function that outputs an unambiguous judgment of moral worth.

I argue that existing AG accounts are not complete in this sense, as there are realistic problem cases in which these accounts struggle to provide an unambiguous judgment of moral worth. That there are such cases at all means that there is a theoretical problem, and that we do not yet have a complete account of moral worth. That some of these cases are realistic means that there is also a practical problem, as these are precisely the cases in which we may need to rely on our account to guide our judgments. The reason that certain cases are problematic, briefly, is that normative theories identify a range of features of actions as right-making and wrong-making. Because an action can reflect appropriate attitudes towards some of these features while reflecting inappropriate attitudes towards others, our account will produce different attributions of moral worth depending on which of an agent's attitudes we evaluate him against.

Fortunately, I argue, this problem can be solved. It will require us to develop a further procedure for determining which attitudes, towards which right- and wrong-making features, we should use to evaluate agents. This in turn will require us to address a further substantive question as to which kinds of attitudes count, for the purposes of assessing character and moral worth, as appropriate or inappropriate attitudes towards that which is actually good or bad. Once this work has been done, however, we will have an account of moral worth that is much more powerful, and that is able to provide unambiguous judgments in the cases which were previously problematic. This strengthened account has potentially surprising consequences when applied to the real world, implying, for instance, that psychopaths are morally blameworthy, and that many seemingly well-meaning agents are morally vicious.


Open Access