Overview of Paper
In order to study how our minds form moral judgments, psychologists often use moral dilemmas based on the now famous philosophical thought experiments involving runaway trolleys. In one such dilemma, people are asked to imagine that a runaway trolley is about to kill five workers but that this can be prevented if one were to push a large person off a footbridge. This person will die, but by blocking the runaway trolley, the lives of the five workers will be saved.
When participants in an experiment judge that we should sacrifice the large person in order to save a greater number, they are usually said to be making a “utilitarian” judgment because they seem to echo the utilitarian idea that our moral decision should only focus on the consequences. A growing body of research has investigated the processes that lead to such “utilitarian” judgments and the factors that can influence them. A number of studies have suggested that judgments rejecting such sacrifices are based in more emotional processing than judgments endorsing them. More disturbingly, “utilitarian” judgments were found to be associated with anti-social traits such as psychopathy – both at a sub-clinical and clinical level. This result has led one major newspaper to conclude that “goodness has nothing to do” with utilitarianism, and that “utilitarians are not nice people”.
But such a tie between utilitarianism and antisocial traits is more than a little surprising when one considers the historical origins and influence of utilitarianism. Developed in 18th century Britain, utilitarianism is a philosophical theory grounded in the core idea that we should always act in the way that would impartially maximize the well-being of everyone on the planet, whether friend or stranger, near or far, human or animal. This is a simple but revolutionary idea. Utilitarians like Jeremy Bentham, John Stuart Mill, and Peter Singer have played a critical role in ‘expanding the moral circle’ and making morality more inclusive by fighting against sexism, racism, and ‘speciesism’, passionately arguing for political and sexual liberty, and making influential efforts to eradicate poverty in developing countries. In the last few years, a new movement of “Effective Altruism” has arisen, inspired by Peter Singer’s utilitarian-based advocacy of self-sacrifice in the name of global charity. How can we square such an impartial ideal of self-sacrifice and charitable giving to distant strangers and with the idea of utilitarians as antisocial, even psychopathic? In this paper, Guy Kahane and I worked in collaboration with Nadira Faber, Julian Savulescu, Molly Crockett, Brian Earp, and Lucius Caviola, to try and resolve this puzzle by suggesting a new two-dimensional model of utilitarian psychology.
According to our 2D model, utilitarian decision-making involves two psychological dissociable and independently important aspects. The first, impartial beneficence, reflects the extent to which individuals endorse the impartial promotion of everyone’s welfare without regard to distance or personal ties. The second, instrumental harm, reflects the extent to which people regard harm to innocent people as a morally acceptable means to achieving a greater good. The widespread use of ‘sacrificial’ moral dilemmas to study utilitarian psychology has yielded such puzzling and even contradictory findings because it focuses almost exclusively on the negative side of utilitarian decision-making. So, while psychopaths may be more willing to push someone off a footbridge to save five others, it would be surprising if these same psychopaths signed up to join an Effective Altruism Club or showed care for the plight of strangers in the developing world. In other words, the sacrificial dilemmas paradigm ignores or downplays the positive, impartial and altruistic core of a utilitarian approach to ethics.
Our 2D model is developed and supported on the basis of both theoretical analysis and empirical work in which we developed, refined, and validated a new approach to measuring utilitarian thinking that taps both positive and negative dimensions of proto-utilitarian tendencies in the lay population: The Oxford Utilitarianism Scale.
A large part of the contribution of this paper is theoretical, but it also has an essential empirical aspect. For this empirical component – to develop and validate a new scale of utilitarian tendencies -, we set in advance and then followed a formal scale development procedure to ensure that our measure was both reliable and valid. First, we created an initial pool of items based on the existing literature on the target construct of utilitarianism. After paring the pool down by, e.g., eliminating redundancies or unclear items, we submitted the items to be reviewed by an expert panel of academic philosophers and then modified the items in response to their feedback. Third, we recruited a large sample of participants to complete the revised pool of items and then conducted a series of exploratory factor analyses (EFA) to obtain the best factor structure with the best items. Fourth, we conducted a series of confirmatory factor analyses (CFA) to evaluate and refine the best factor structure based on pre-determined indices of model fit obtained from recent recommendations in the literature. Fifth, we recruited a new sample of participants to complete the items determined by the CFA and then we confirmed, using this new sample, that the measure had appropriate factor structure and psychometric properties. Sixth, we confirmed that the data fit this model and factor structure better than alternative models (e.g., that a multi-dimensional model obtained from the CFA accounted for the data better than a one-dimensional model, or vice versa). Seventh, we explored construct validity by testing how scores on the final scale obtained from the previous steps were connected to other established measures. Finally, based on helpful feedback from reviewers of a previous draft of this paper, we investigated external validity by administering the scale to an expert sample of graduate students and academics specializing in moral philosophy.
OXFORD UTILITARIANISM SCALE
Kahane, Everett, et al. (2018)
As part of this paper, we have developed the Oxford Utilitarianism Scale (Kahane, Everett, et al. 2018). This consists of 9 items across two sub scales. The first subscale – Impartial Beneficence (OUS-IB) – consists of 5 items that all tap endorsement of the impartial maximization of the greater good, even at the cost of personal self-sacrifice. The second subscale Instrumental Harm (OUS-IH) consists of 4 items that tap a willingness to cause harm in order to bring about the greater good.
- We develop, refine, and validate a new scale – the Oxford Utilitarianism Scale – to dissociate individual differences in the ‘negative’ (permissive attitude to instrumental harm) and ‘positive’ (impartial concern about the greater good) dimensions of utilitarian thinking as manifested in the general population.
- We show that these are two independent dimensions of proto-utilitarian tendencies in the lay population, each exhibiting a distinct psychological profile.
- Empathic concern, identification with the whole of humanity, and concern for future generations were positively associated with Impartial Beneficence but negatively associated with Instrumental Harm; and while Instrumental Harm was associated with sub-clinical psychopathy, Impartial Beneficence was associated with higher religiosity.
- Importantly, while these two dimensions were independent in the lay population they were closely associated in a sample of moral philosophers.
The use of sacrificial dilemmas to study the contrast between utilitarian and deontological judgments has dominated research in moral psychology. But while the psychological source of this major ethical dispute is of great interest, sacrificial dilemmas are a limited tool for studying proto-utilitarian tendencies in the lay population (Kahane, 2015; Kahane et al., 2015; Kahane & Shackel, 2010). In this paper we have introduced the 2D model of utilitarian decision-making, a new theoretical framework for studying such tendencies. The 2D model treats utilitarian moral decision-making not as an all-or-nothing category but as a matter of degree, and as involving two largely independent ‘positive’ and ‘negative’ dimensions. On this basis, we developed the OUS, a new scale that is both philosophically rigorous and empirically driven, and which attempts to address concerns about sacrificial dilemmas and existing scales. Our preliminary application of the scale already demonstrates how the distinction between impartial beneficence and instrumental harm can help to clarify the relationship between a utilitarian moral outlook and a range of other psychological constructs and measures while generating important new avenues for further research. Importantly, our results strongly suggest that, in the context of the lay population, utilitarian decision-making does not constitute a unitary psychological phenomenon.