What are the bearers of virtue?

(UPDATED August 28th)

I’m doing a lot of writing these days.  Here’s another draft paper, for a collection edited by Hagop Sarkissian and Jen Cole Wright.  The title is, “What are the bearers of virtue?”  The answer I propose is that, contrary to a natural assumption, the bearer of a virtue is not an individual agent, but a complex of object that encompass the agent, her social milieu, and her asocial environment.

As always, comments and objections are quite welcome.


What are the Bearers of Virtues?



Despite the recent hubbub over the possibility that the concepts of virtue and the virtues are empirically inadequate,[1] researchers have only superficially considered the fact that these concepts pick out dispositional properties.[2]  For the first time in this controversy, we need to take the dispositional nature of virtue seriously.  Once we do, one question immediately arises: what are the bearers of virtues?


In this paper, I argue for an embodied, embedded, and extended answer to this question.  It is generally hopeless to try to say what someone would do in a given normative state of affairs without first specifying bodily and social features of the situation.  There’s typically no fact of the matter, for instance, about whether someone would help when there is sufficient reason to help.  However, there typically is a fact of the matter about whether someone in a particular bodily state and social environment would help when there is sufficient reason to help.


If that’s right, it puts some pressure on agent-based theories of virtue, which tend to claim or presume that the bearers of virtue are individual agents (Russell 2009, Slote 2001).  Such theories hold that a virtue is a monadic property of an individual agent.  Furthermore, this objection to agent-based and agent-focused theories suggests a way of reconceiving of a virtue as a triadic relation among an agent, a social milieu, and an asocial environment (Alfano 2013).  On this relational model, the bearers of virtue are not individual agents but ordered triples that refer in addition to objects outside the agent.


Here is the plan for this paper: Section 1 summarizes the relevant literature on dispositions.  Next, section 2 sketches some of the relevant psychological findings, which suggest that how people are disposed to think, feel, deliberate, act, and react is determined to a large degree by both their immediate asocial environment, which in turn influences their moods and other bodily states, and their social milieu.  Finally, section 3 argues that the best response to the empirical evidence is to revise the concept of a virtue.  A virtue is not a monadic property of an agent, but a triadic relation among an agent, a social milieu, and an asocial environment.


1. Virtues as Dispositional Properties


1.1. The Subjunctive Conditional Analysis


The most natural way to approach dispositions is through the simple subjunctive conditional analysis:


(S-SCA) object o is disposed to activity A in condition C if and only if o would A if C were the case (Choi & Fara 2012).


The A-term refers to the characteristic manifestation of the disposition; the C-term refers to its stimulus conditions.  To say that OxyContin is an analgesic for humans is to attribute a dispositional property to it: OxyContin is disposed to relieve pain when ingested by a human.  This statement would then be analyzed as: OxyContin would relieve pain if it were ingested by a human.  According to the standard semantics for subjunctive conditionals, this analysis means that all close possible worlds at which a pained person ingests OxyContin are worlds at which that person’s pain is subsequently relieved.


Many of the dispositions we refer to on a regular basis, such as being fragile, soluble, or poisonous, do not wear their manifestation or stimulus conditions on their sleeves.  Following Lewis (1997), the standard “two-step” strategy for dealing with them is first to spell out these conditions, and then to translate them into the subjunctive conditional schema.


Furthermore, it’s now recognized that most dispositions have characteristic masks, mimics, and finks.  A fragile vase might not break when struck because its fragility is masked by protective packaging: at some nearby worlds, the vase is both struck and protected, and at those worlds it does not break.  A sugar pill can mimic a real analgesic via the placebo effect: at some nearby worlds, a placebo is ingested by a pained individual whose pain is subsequently relieved.  An electron’s velocity, which it is naturally disposed to retain, inevitably changes when it is measured – a case of finking.[3]  Such possibilities are not evidence against the presence or absence of the disposition in question; instead, they are exceptions to the subjunctive conditional.  The vase really is fragile, despite the falsity of the subjunctive conditional.  The sugar pill is not really an analgesic, despite the pain relief.  The electron really is disposed to follow its inertial path, despite Heisenberg’s uncertainty principle.


Let’s stipulate that finks, masks, and mimics be collectively referred to as disrupters.  Since it is possible to possess a disposition that is susceptible to finks and masks, and to lack a disposition that is mimicked, the simple subjunctive conditional analysis fails.  In my view, the most attractive response to the constellation of disrupters is Choi’s (2008) anti-disrupter SCA:


(AD-SCA) object o is disposed to activity A in condition C if and only if o would A if C were the case and there were no disrupters present.


1.2. Woulda, Coulda, Shoulda


The subjunctive conditional analysis, even in the sophisticated form that allows for disrupters, expresses a very strong notion of what it takes to be a disposition.  Consider, for example, a loaded die that has a .5 probability of showing ace and a .1 probability of showing each of the other results.  Surely, one might argue, the die is disposed to show ace, even though there are plenty of close possible worlds at which it shows two, three, four, five, or six.


In light of such cases, it’s possible to provide weak and comparative analyses of dispositions.  For instance, we can analyze a weak disposition as follows:


(W-AD-SCA) o is weakly disposed to A in C if and only if o could A if C were the case and there were no disrupters present.


In the standard semantics, this means that o As at some undisrupted nearby C-worlds.  Obviously, the same object can be both weakly disposed to A in C and weakly disposed to not-A in C.  This makes weak dispositions less informative than we might like.  W-AD-SCA can be supplemented with the comparative analysis:


(C-AD-SCA) o is more disposed to A than to A* in C if and only if o is significantly more likely to A than to A* if C were the case and no disrupters were present.


In the standard semantics, this means that there are more undisrupted C-worlds where o As than undisrupted C-words where o A*s.


Which, if any, of these notions is appropriate to an analysis of virtues?  Elsewhere (Alfano 2013), I have argued for an intuitive distinction between high-fidelity and low-fidelity virtues.  High-fidelity virtues, such as honesty, chastity, and loyalty, require near-perfect manifestation in undisrupted conditions.  For these, AD-SCA seems most appropriate.  Someone only counts as chaste if he (almost) never cheats on his partner when cheating is a temptation.  Low-fidelity virtues, such as generosity, tact, and tenacity, are not so demanding.  For them, some combination of the W-AD-SCA and C-AD-SCA seems appropriate.  Someone might count as generous if he was more disposed to give than not to give when doing so was appropriate; someone might count as tenacious if she were more disposed to persist than not to persist in the face of adversity.


If this is on the right track, the analysis of virtue dispositions adds one additional step before Lewis’s two.  First, determine whether the virtue in question is high-fidelity or low-fidelity.  For instance, it’s reasonable to suppose that helpfulness is a low-fidelity virtue whereas loyalty is a high-fidelity virtue.  Second, identify the stimulus conditions and characteristic manifestations.  The most overt manifestation of helpfulness is of course helping behavior, but more subtle manifestations presumably include noticing opportunities to help, having an occurrent desire to help, deliberating in characteristic ways, and forming the intention to help.  The most overt manifestation of loyalty is refusal to betray, but again there are more subtle manifestations.  The stimulus condition for helpfulness is a normative state of affairs: that there is adequate reason to help.  For loyalty, too, the stimulus condition appears to be a normative state of affairs: that there is temptation to betray whomever or whatever one is loyal to.  Finally, the stimulus conditions and characteristic manifestations are slotted into the relevant schema.  To be helpful, then, someone must be weakly disposed to help (among other things) when there is adequate reason to do so, whereas to be loyal is to be strongly disposed not to betray (among other things) when there is a temptation to do so.



2. The Psychology of Dispositions


In this section, I argue that both high-fidelity and low-fidelity virtues, as traditionally conceived, suffer from an indeterminacy problem.  There’s often no fact of the matter about whether someone would exhibit the characteristic manifestations of a high-fidelity virtue in undisrupted stimulus conditions.  There’s often no fact of the matter about whether someone is more disposed than not to exhibit the characteristic manifestations of a low-fidelity virtue in undisrupted stimulus conditions.  This is because both bodily and social factors partially determine which possible worlds become actual.


Someone might be strongly disposed to tell the truth if others think of her as an honest person, but strongly disposed to deceive if others think of her as a dishonest person.  For such a person, it would be incorrect to say that she would tell the truth in undisrupted stimulus conditions, but also incorrect to say that she would lie in undisrupted stimulus conditions.  However, once the asocial and social context are specified as well, it generally will be possible to assert true subjunctive conditionals.


Similarly, someone might be significantly more disposed to persist than desist in the face of adversity if she is in a good mood, but significantly more disposed to desist than persist if she is in a bad mood.  For such a person, it would be incorrect to say that she is weakly disposed to persist in undisrupted stimulus conditions, but also incorrect to say that she is weakly disposed to desist in undisrupted stimulus conditions.  However, as before, once the asocial and social context are specified, it generally will be possible to assert true subjunctive conditionals.


2.1. Asocial Situational Influences


Asocial situational influences are asocial features of the immediate environment that partially determine which subjunctive conditionals are true about how someone will act in a given virtue’s stimulus conditions.  They come in two main varieties: ambient sensibilia, and mood modulators.


For instance, the volume of ambient sound influences both helping behavior and physical aggressiveness.  In a loud environment (80+ db), people tend not to help one another, but in a quiet environment (79- db), they do (Matthews & Cannon 1975).  A loud environment tends to lead to aggressive behavior, but a quiet environment does not (Donnerstein & Wilson 1976).  Or consider the olfactory modality.  People tend to engage in helping behavior when exposed to pleasant smells, such as coffee or freshly baked bread (Baron 1997).  Or consider the visual modality: subtle shifts in lighting partially determine whether people cheat.  They cheat more in an almost-imperceptibly darker room, and act more selfishly when wearing shaded glasses rather than clear ones (Zhong, Bohns, & Gino 2010).


These are just a few of literally hundreds of relevant studies.  Together, they suggest that there are stable, though weak, connections between seemingly morally irrelevant sensibilia and the manifestation (or not) of virtue.  The connections don’t all run in the same direction, and they interact.  Loud environments don’t uniformly dispose towards morally bad behavior, nor do pleasant smells invariably dispose towards morally good behavior.  Whether a given sensory input will tend to produce good, bad, or neutral results depends on what kind of behavior is normatively called for and which other sensibilia are present, among many other things.


I should also note that these connections are not crudely behavioristic.  Sensory stimuli influence behavior in large part by modifying the agent’s cognitive and motivational set.  Loud noises, for instance, result in attentional focusing (Cohen 1978), while pleasant smells produce openness to new experiences (Baron & Thomley 1994).  These internal states mediate the connection between the asocial environment and the agent’s behavior.


The other main asocial influence is the set of affect modulators, very broadly construed to include mood elevators, mood depressors, emotion inducers, and arousal modifiers.  Embarrassed people are more disposed to help than unabashed people (Apsler 1975).  People who feel guilty are more disposed to help – regardless of whom they feel guilty towards (Regan 1971).  People who’ve been put in a good mood are more disposed to help and to try new things in general (Isen 1987).  Those who feel disgusted make harsher moral judgments, even if the disgust is ostensibly directed at something other than the object of moral judgment (Schnall, Haidth, Clore, & Jordan 2008).  Sexually aroused men are more disposed to engage in disgusting, illegal, or immoral activities (Ariely 2008).


As with sensibilia, affect modulators are connected in weak but stable ways to the manifestation (or not) of virtue.  Fair moods don’t necessarily make us fair, nor do foul moods make us foul.  The valence of the effect depends on what is normatively called for in the particular circumstances.  And as with sensibilia, these connections are not crudely behavioristic.  Internal states, such as perceptual sensitivities, desires, and expectations, mediate the effect on behavior.


It’s important to point out, furthermore, that while asocial influences tend to have fairly predictable effects on behavioral dispositions, they by no means explain action all by themselves.  Indeed, any particular factor will typically account for between 4% and 16% of the variance in behavior (Funder & Ozer 1983).  This is one reason among several that I am not proposing a purely bodily account of personality and behavior.  As Walter Mischel and Yuichi Shoda (1995) have emphasized, behavioral coherence is to be sought not in one factor alone, but in the interaction between personal, social, and asocial factors.


2.2. Social Influences


In addition to the asocial influences canvassed above, there are a variety of social influences on the manifestation (or not) of virtue.  Somewhat roughly, there are two types of such influences: expectation confirmation, and outgroup bias.  In cases of expectation confirmation, what happens is the agent reads others’ expectations off of explicit or implicit social cues, and then acts in accordance with the expectations so read.  Of course, people often mistake or misinterpret others’ expectations, so what they end up doing isn’t necessarily what others expect, but what they think others expect.  In cases of outgroup bias, the agent displays a prejudice against someone who is perceived as not part of the ingroup.  Since everyone belongs to myriad social groups, whether someone is perceived as in or out depends on which group identities are salient at the time, which leads to systematic misperception of social distance.  In this section, I have room to discuss only social expectations in detail.


Much of the groundbreaking social psychology of the second half of the 20th century investigated the power of expectation confirmation.  The most dramatic demonstration was of course Stanley Milgram’s (1974) studies in obedience, in which roughly two thirds of participants were induced to put what they thought was 450 volts through another participant (actually an actor who was in on the experiment) three times in a row.  While there were many important features of this study, the key upshot was that the participants were willing to do what they should easily have recognized was deeply immoral based on the say-so of a purported authority figure.  They performed exactly as expected.  Blass (1999) shows in a meta-analysis that Milgram’s results were no fluke: they have been replicated all around with the world with populations of diverse age, gender, and education level.


Another striking paradigm for investigating the power of social expectations is the Asch (1951, 1955, 1956) paradigm, in which participants publicly answer questions after hearing the unanimous and obviously wrong responses of at least three confederates.  Roughly half the participants give what they themselves recognize to be the wrong answer in such circumstances because they detect an expectation to say what others have said.


Yet another example of the power of social expectations is the large literature on bystander apathy that grew in response to the brutal rape and murder of Kitty Genovese in New York City (Darley & Latané 1968; Latané & Darley 1968; Latané & Darley 1970; Latané & Nida 1981).  It turns out that, the more bystanders are present in an emergency situation, the lower the chances that even one of them will intervene.  What seems to happen in such cases is that people scan others immediate reactions to help themselves determine what to do.  When they see no one else reacting, they decide not to intervene either; thus everyone interprets everyone else’s moment of decision as a decision not to intervene.  Then, since they think that intervention is not expected, it doesn’t occur.


Reading off others’ expectations and acting accordingly doesn’t always lead to bad outcomes, though.  Recent work on the phenomenon of social proof shows that the normative valence of acting in accordance with expectations depends on what’s expected.  For instance, guests at a hotel are 40% more likely to conserve water by not asking for their towels to be washed if they read a message that says, “75% of the guests who stayed in this room participated in our resource savings program by using their towels more than once” than one that says, “You can show respect for nature and help save the environment by reusing towels during your stay” (Goldstein, Cialdini, & Griskevicius 2008)


Psychologists and behavioral economists have also investigated the effect of extremely subtle, thoroughly embodied, social distance cues on moral behavior.  In a string of fascinating studies, it’s been shown that people are much more willing to share financial resources (Burnham 2003; Burnham & Hare 2007; Rigdon, Ishii, Watabe, & Kitayama 2009), less inclined to steal (Bateson, Nettle, & Roberts 2006), and less disposed to litter (Ernest-Jones, Nettle, & Bateson 2011) when they are “watched” by a representation of a face.  The face can be anything from a picture of the beneficiary of their behavior to a cartoon robot’s head to three black dots arranged to look like eyes and a nose.  In debriefings, participants never voiced suspicions about the of the face, and yet their behavior was clearly influenced by it.


The last example of the power of social expectations involves the use of labeling to induce label-congruent thoughts, feelings, and behavior.  As I document and explain in more detail elsewhere (Alfano 2013), people tend to act in accordance with the traits that are attributed to them, whether the traits are minor virtues such as tidiness (Miller, Brickman, & Bolen 1975) and ecology-mindedness (Cornelissen et al. 2006, 2007), major virtues such as charity (Jensen & Moore 1977), cooperativeness (Grusec, Kuczynski, Simutis & Rushton 1978), and generosity (Grusec & Redler 1980), or vices such as cutthroat competitiveness (Grusec, Kuczynski, Simutis & Rushton 1978).  On this view, when people act in accordance with a virtue, they often do so not because they have the trait in question, but because they think they do or because they know that other people think they do.  I call such simulations of moral character factitious virtues.


These empirical considerations weigh heavily in favor of Merritt’s (2009) argument that the very social structures that sometimes enable us to act virtuously also sometimes lead us to act viciously.  We are deeply social animals, and it would be both against our nature and in many cases counter-productive to try to deny this.


3. Revising the Metaphysics of Virtue


In the previous section, I argued that both social and asocial influences shape how people are disposed to think, feel, and act.  What we notice, what we think, what we care about, and what we do depends in part on bodily and social features of our situations.  This is not to deny that people also bring their own distinctive personalities to the table, but it suggest that both high-fidelity and low-fidelity virtues, as traditionally conceived, are not thick on the ground.


To see why, let’s walk through the three-step analysis of a traditional cardinal virtue: honesty.  For current purposes, I’ll assume that it’s uncontroversial that honesty is high-fidelity.  Next, we specify the stimulus conditions and characteristic manifestations.  I don’t have space to do justice to the required nuances here, but it wouldn’t be too far off to say that the stimulus conditions C are temptations to lie, cheat, or steal, and that the characteristic manifestations A are behavioral (not lying, cheating, or stealing), cognitive (noticing the temptation without feeling too much of its pull), and affective (disapprobation of inappropriate behavior, desire to extricate oneself from the tempting situation if possible, perhaps even prospective shame at the thought that one might end up acting badly).  Finally, we slot these specifications into the schema for high-fidelity virtue:


(AD-SCA-honesty) The agent is disposed to activity A in condition C if and only if she would A if C were the case and there were no disrupters present.


In other words, at all nearby undisrupted worlds where she is tempted, she thinks, feels, and acts appropriately.  It’s only reasonable to assume that at some of these undisrupted temptation-worlds, however, the lighting will not be bright; at others, she might not feel watched; at still others, she will have been labeled dishonest, or at least not have been labeled honest.  What she would do depends in part on these factors, and so the subjunctive conditional is false, which in turn means that she is not honest.


It could be objected that these factors are disrupters, and so should be ruled out by fiat.  This objection is unmotivated, however.  We can see how protective packaging masks an object’s fragility.  Does it make sense to say that being in slightly dim conditions would mask someone’s honesty?  What good is honesty if it gives out so easily?  Does it make sense to say that not being watched would fink someone’s honesty?  What good is honesty if honest people need constant monitoring?  Ruling out all of the asocial and social influences described in the previous section as disrupters is not only ad hoc; it destroys the evaluative valence of honesty, making it no longer a virtue.


Furthermore, social and asocial influences are nearly ubiquitous.  Indeed, it’s difficult even to think of them as influences because they are so common.  Should we say that very bright lighting is the default condition, and that lower levels of light are all situational influences?  To do so would be to count half of each day as disrupted.  Or should we say that twilight is the default, and that both very bright and very dark conditions are situational influences?  It’s hard to know what would even count in favor of one of these proposals.


Even more to the point, what one might want to rule out as a disrupter in one case is likely to contribute to the manifestation of virtue in other cases.  Should we say that being watched, or feeling that you are watched, is a situational influence?  Then it might be that all of the reprehensible behavior displayed in the Milgram experiments was committed by people whose virtue was finked or masked, but it would also mean that all of the praiseworthy behavior someone exhibits while “watched” by a representation of a face is not virtue but mimicked virtue.  If we try to rule out all of these factors, leaving just the agent in her naked virtue or vice, we may find that she disappears too.  Strip away the body and the society, and you leave not the kernel of authentic character, but something that’s not even recognizably human.


Instead of filtering out as much as possible, I want to propose that we should try to include as much as possible, by expanding the unit of analysis, the bearer of virtue.  Instead of thinking of virtue as a property of an individual agent, we should construe it as a triadic relation among a person, a social milieu, and an asocial environment.


There are two ways of fitting the milieu and the environment into the subjunctive conditional analysis.  They could incorporated into the stimulus conditions:


(AD-SCA*) Person p is disposed to activity A in condition C, social milieu S, and asocial environment E if and only if p would A if C were the case, p were in S and E, and there were no disrupters present.


Or they could be fused with the agent:


(AD-SCA†) Person p in social milieu S and asocial environment E is disposed to activity A in condition C if and only if p-in-S-and­-E would A if C were the case and there were no disrupters present.


According to AD-SCA*, the person is still the sole bearer of the disposition; it’s just a more limited disposition.  According to AD-SCA†, in contrast, the bearer of the disposition is now a complex, extended object: the person, the milieu, and the environment.  What I want to suggest is that, given the sorts of creatures we are – embodied, socially embedded, with cognition and motivation extended beyond the boundaries of our own skin (Clark 2008, Clark & Chalmers 1998) – AD-SCA* is more attractive.


Virtue would inhere, on this view, in the interstices between the person and her world.  To be honest, for example, would be to have certain basic personality dispositions, but also to be considered honest by one’s peers (and to know it), to consider oneself honest, to be watched or at least watchable, and to be in whatever bodily states promote the characteristic manifestations of honesty.  Someone could become honest, on this view, in standard ways, such as habituation and reflection on reasons.  But someone could also become honest in non-standard ways, such as having the expectations of others signaled or the lights turned down.  This makes it both easier and harder to be virtuous: deficiencies in personality can be made up for through social and bodily support, but strength of personality can also be undermined by lack of social and bodily support.


One of the salutary upshots of this way of thinking about virtue is that it helps to make sense of the enormous diversity named by any given trait term.  Different people are more or less honest, and on many dimensions.  By making explicit reference to the social milieu and the asocial environment, this framework suggests ways in which partial virtue could be differently instantiated.  Two people might both count, at a very coarse-grained level of description as mostly honest, but one could do so because of personal and social strengths but despite asocial weaknesses, while the other does so because of social and asocial strengths but despite some personal weaknesses.


Furthermore, the framework allows for the plausible idea that there is a kind of asymmetry amongst the bearers of virtue.  Someone’s personality can only be so weak before we are no longer inclined to call him virtuous, even if that weakness is counteracted by great social and asocial strengths.  We can also make sense of the feeling that someone is extremely virtuous if he displays characteristic manifestations despite weaknesses or pitfalls in the social and asocial environment.  I’m not inclined to go too far in this direction, but the framework does allow for such asymmetries.


Before concluding, I want to point to two normative upshots of this view.  The first is that each of us is to some extent causally and even constitutively responsible for the character of others, and in several ways.  By signaling our expectations, we tend to induce expectation-confirming responses.  By interacting with others, we alter their moods.  When we help to construct the material and bodily environment, we also construct others’ character.  For someone who is on the verge between virtue and vice – that is, for most people – such influences can make all the difference.  (Hagop Sarkissian [2010] makes a similar point.)  The comforting myth of individual responsibility notwithstanding, each of us truly is our brothers’ and sisters’ keeper.


This is a heavy responsibility to bear, but it pales in comparison to the responsibility borne by those with the power to set the default expectations that govern a society and to shape the material conditions of people’s lives.  On the view I am proposing, politicians, corporate leaders, reports, and architects, among many others, quite literally have the power to make people virtuous – or prevent them from being or becoming virtuous.  If this is right, we need to hold such people more accountable, and to stop pretending that it’s possible to separate political and economic power from morality.



While any proposed interpretation of the vast empirical evidence on character is inevitably tentative, speculative, and corrigible, I think we now have good reason to pursue the idea that character traits are not dispositional properties of individual agents, but dispositional relations among agents and their environments.  Many of the key elements of the environment, on this view, are grounded in the fact that the human animal is embodied, socially embedded, and extended outside the limits of its own skin.  Perhaps it should come as no surprise, then, that virtue, too, is embodied, embedded, and extended.









Adams, R. M. (2006). A Theory of Virtue: Excellence in Being for the Good. Oxford UP.

Alfano, M. (2013). Character as Moral Fiction. Cambridge UP.

Annas, J. (2011). Intelligent Virtue. Oxford University Press.

Apsler, R. (1975).  Effects of embarrassment on behavior toward others.  Journal of Personality and Social Psychology, 32, 145-153.

Ariely, D. (2008). Predictably Irrational.  New York: Harper Collins.

Asch, S. (1951). Effects of group pressures upon the modification and distortion of judgment. In Guetzkow (ed.) Groups, Leadership, and Men. Pittsburgh: Carnegie Press

Asch, S. (1955). Opinions and social pressure. Scientific American, 31-35.

Asch, S. (1956). Studies of independence and conformity: A minority of one against a unanimous majority. Psychological Monographs, 70:9.

Baron, R. (1997).  The sweet smell of … helping:  Effects of pleasant ambient fragrance on prosocial behavior in shopping malls.  Personality and Social Psychology Bulletin, 23, 498-503.

Baron, R. A., & Thomley, J. (1994). A whiff of reality: Positive affect as a potential mediator of the effects of pleasant fragrances on task performance and helping.  Environment and Behavior, 26, 766-784.

Bateson, M., Nettle, D., & Roberts, G. (2006). Cues of being watched enhance cooperation in a real-world setting.  Biology Letters, 12, 412-414.

Blass, T. (1999).  The Milgram paradigm after 35 years:  Some things we now know about obedience to authority.  Journal of Applied Social Psychology, 29:5, 955-978.

Boles, W. & Haywood, S. (1978).  The effects of urban noise and sidewalk density upon pedestrian cooperation and tempo.  Journal of Social Psychology, 104, 29-35.

Burnham, T. (2003).  Engineering altruism: A theoretical and experimental investigation of anonymity and gift giving.  Journal of Economic Behavior and Organization, 50, 133-144.

Burnham, T. & Hare, B. (2007).  Engineering human cooperation.  Human Nature, 18.2, 88-108.

Carlsmith, J. & Gross, A. (1968).  Some effects of guilt on compliance.  Journal of Personality and Social Psychology, 53, 1178-1191.

Choi, S. (2008). Dispositional properties and counterfactual conditionals. Mind, 117, 795-841.

Choi, S. & Fara, M. (Spring 2012). Dispositions. The Stanford Encyclopedia of Philosophy (Spring 2012 Edition), Edward N. Zalta (ed.), URL = <http://plato.stanford.edu/archives/spr2012/entries/dispositions/>.

Clark, A. (2008). Supersizing the mind: Embodiment, Action, and Cognitive Extension. Oxford UP.

Clark, A. & Chalmers, D. (1998). The extended mind. Analysis, 58, 7-19.

Cornelissen, G., Dewitte, S. & Warlop, L. (2007). Whatever people say I am that’s what I am: Social labeling as a social marketing tool.  International Journal of Research in Marketing, 24:4, 278-288.

Cornelissen, G., Dewitte, S., Warlop, L., Liegeois, A., Yzerbyt, V., Corneille, O. (2006). Free bumper stickers for a better future: The long term effect of the labeling technique.  Advances in Consumer Research, 33, 284-285.

Darley, J. & Latané, B. (1968).  Bystander intervention in emergencies:  Diffusion of responsibility.  Journal of Personality and Social Psychology, 8, 377-383.

Donnerstein, E. & Wilson, D.  (1976).  Effects of noise and perceived control on ongoing and subsequent aggressive behavior.  Journal of Personality and Social Psychology, 34, 774-781.

Doris, J. (1998).  Persons, situations, and virtue ethics.  Nous, 32:4, 504-540.

Doris, J. (2002).  Lack of Character: Personality and Moral Behavior.  Cambridge:  Cambridge University Press.

Ernest-Jones, M., Nettle, D., & Bateson, M. (2011). Effects of eye images on everyday cooperative behavior: A field experiment. Evolution and Human Behavior, 32:3, 172-178.

Flanagan, O. (1991). Varieties of Moral Personality: Ethics and Psychological Realism. Harvard UP.

Flanagan, O. (2009). Moral science? Still metaphysical after all these years, in Narvaez & Lapsley (eds.), Moral Personality, Identity and Character: An Interdisciplinary Future. Cambridge University Press.

Funder, D. & Ozer, D. (1983).  Behavior as a function of the situation.  Journal of Personality and Social Psychology, 44, 107-112.

Grusec, J., Kuczynski, L., Rushton, J., & Simutis, Z. (1978). Modeling, direct instruction, and attributions: Effects on altruism. Developmental Psychology, 14, 51-57.

Grusec, J. & Redler, E. (1980). Attribution, reinforcement, and altruism: A developmental analysis, 16:5, 525-534.

Hacking, I. (1999). The Social Construction of What? Harvard UP.

Harman, G. (1999).  Moral philosophy meets social psychology: Virtue ethics and the fundamental attribution error.  Proceedings of the Aristotelian Society, New Series 119, 316-331.

Harman, G. (2000).  The nonexistence of character traits.  Proceedings of the Aristotelian Society, 100: 223-226.

Harman, G. (2003). No character or personality.  Business Ethics Quarterly, 13:1, 87-94.

Isen, A. (1987).  Positive affect, cognitive processes, and social behavior.  In L. Berkowitz (ed.) Advances in Experimental Social Psychology, volume 20, 203-254.  San Diego:  Academic Press.

Isen, A., Clark, M., & Schwartz, M. (1976). Duration of the effect of good mood on helping: “Footprints on the sands of time.” Journal of Personality and Social Psychology, 34, 385-393.

Isen, A., Shalker, T., Clark, M., & Karp, L. (1978).  Affect, accessibility of material in memory, and behavior:  A cognitive loop.  Journal of Personality and Social Psychology, 36, 1-12.

Jensen, A. & Moore, S. (1977).  The effect of attribute statements on cooperativeness and competitiveness in school-age boys.  Child Development, 48, 305-307.

Johnston, M. (1992). How to speak of the colors. Philosophical Studies, 68, 221-263.

Latané, B., & Darley, J. (1968).  Group inhibition of bystander intervention in emergencies.  Journal of Personality and Social Psychology, 10, 215-221.

Latané, B., & Darley, J. (1970).  The Unresponsive Bystander:  Why Doesn’t He Help?  New York:  Appleton-Century-Crofts.

Latané, B., & Nida, S. (1981).  Ten years of research on group size and helping.  Psychological Bulletin, 89, 308-324.

Lewis, D. (1997). Finkish dispositions. The Philosophical Quarterly, 47, 143-158.

MacIntyre, A. (1984).  After Virtue: A Study in Moral Theory.  Notre Dame: University of Notre Dame Press.

Martin, C. (1994). Dispositions and conditionals. The Philosophical Quarterly, 44, 1-8.

Matthews, K. E., & Cannon, L. K. (1975).  Environmental noise level as a determinant of helping behavior.  Journal of Personality and Social Psychology, 32, 571-577.

McKitrick, J. (2003). A case for extrinsic dispositions. Australasian Journal of Philosophy, 81, 155-174.

Merritt, M. (2000). Virtue ethics and situationist personality psychology. Ethical Theory and Moral Practice, 3:4, 365-383.

Merritt, M. (2009). Aristotelian virtue and the interpersonal aspect of ethical character. Journal of Moral Philosophy, 6:1, 23-49.

Milgram, S. (1974).  Obedience to Authority.  New York: Harper Collins.

Miller, R., Brickman, P., & Bolen, D. (1975). Attribution versus persuasion as a means for modifying behavior.  Journal of Personality and Social Psychology, 31:3, 430-441.

Mischel, W. (1968).  Personality and Assessment.  New York: Wiley.

Mischel, W. & Shoda, Y. (1995). A cognitive-affective system theory of personality: Reconceptualizing the invariances in personality and the role of situations.  Psychological Review, 102:2, 246-268.

Regan, J. (1971).  Guilt, perceived injustice, and altruistic behavior.  Journal of Personality and Social Psychology, 18, 124-132.

Rigdon, M., Ishii, K., Watabe, M., & Kitayama, S. (2009). Minimal social cues in the dictator game. Journal of Economic Psychology, 30.3, 358-367.

Russell, D. (2009). Practical Intelligence and the Virtues.  Oxford UP.

Sarkissian, H. (2010). Minor tweaks, major payoffs: The problems and promise of situationism in moral philosophy. Philosophers’ Imprint, 10:9.

Schnall, S., Haidt, J., Clore, G., & Jordan, A. (2008). Disgust as embodied moral judgment. Personality and Social Psychology Bulletin, 34, 1096-1109.

Schwartz, S. & Gottlieb, A. (1991).  Bystander anonymity and reactions to emergencies.  Journal of Personality and Social Psychology, 39, 418-430.

Searle, J. (1995). The Construction of Social Reality. New York: Free Press.

Slote, M. (2001). Morals from Motives. Oxford UP.

Smith, A. (1977). Dispositional properties. Mind, 86, 439-445.

Snow, N. (2010). Virtue as Social Intelligence: An Empirically Grounded Theory. New York: Routledge.

Sreenivasan, G. (2008). Character and consistency: Still more errors. Mind, 117:467, 603-612.

Upton, C. (2009). Situational Traits of Character: Dispositional Foundations and Implications for Moral Psychology and Friendship. Lexington Books.

Weyant, J. (1978).  Effects of mood states, costs, and benefits on helping.  Journal of Personality and Social Psychology, 36, 1169-1176.

Zhong, C.-B., Bohns, V., & Gino, F. (2010). Good lamps are the best police: Darkness increases dishonesty and self-interested behavior. Psychological Science, 21:3, 311-314.


[1] The canonical firebrands are Doris (1998, 2002) and Harman (1999, 2000, 2003).  Flanagan arrived at the party both too early (1991) and too late (2009) to shape the course of the debate.

[2] Upton (2009) is the only book-length effort, but her work makes little use of the literature on dispositions, relying instead on her own naive intuitions.  Sreenivasan (2008) also discusses the dispositional nature of virtues without reference to the literature on dispositions.

[3] The concepts of masking, mimicking, and finking were introduced by Johnston (1992), Smith (1977), and Martin (1994), respectively.

Leave a Reply