Moral Perception

Some people consider bullfighting to be a traditional ‘sport’, involving two parties (across different styles): the toreros (i.e. the performers) and the bulls. During the performance, the toreros attack the bull, encouraging it to charge and then attempting to wound it in various places with daggers and barbed sticks until it is weakened enough for the final thrust. This bloody sport typically ends with the killing of the bull (League Against Cruel Sports, 2026). Despite its long history dating back centuries (Claramunt López, 1989), public opinion consistently opposes it. In recent times, polls in Spain (one of the countries with the longest tradition), Portugal and France suggest not only that people demand an end to bullfighting (around 77%), but also that they are aware of the pain and suffering inflicted on the animals (between 80% and 70% of respondents in a sample of 7,500 in Spain, Portugal and France, according to CAS International) (CAS International, 2024).

However, how can this not be seen as the pinnacle of a traditional sport and a representation of the exhilarating experience of fighting a wild animal one-to-one? Conversely, how can someone not see it as the culmination of the senseless abuse and killing of an innocent animal? The ethical and political arguments are clear. The evidence that the animal can feel pain is irrefutable, and therefore we must provide welfare. For further ethical argumentation towards not hurting non-human sensible animals, see the entry for ‘Animal Liberation Now’ for example.

Some would argue that these reasons are all we need to change people’s minds, and they might be right. However, there is something that comes before rational argumentation which is sometimes overlooked: The perception of the moral action itself, and the faculty that comes with it: moral perception. In this text, I am not going to discuss the reasons to condemn and ban bullfighting further, but I am going to use it as an example to analyse how we can understand these two competing views on the morality of bullfighting under Audi’s Moral Perception framework. To do so, I will first introduce Audi’s work and then explain some of the concepts that structure his framework for understanding moral perception. I will talk about: (1) what moral perception is; (2) how it is grounded; (3a) who has moral perception; and (3b) under what circumstances. Finally, I will advocate the importance of knowledge and training in recognising the moral deed from Audi’s perspective.

When we consider the perception of a moral act, such as bullfighting, two things usually come to mind: (1) an ordinary sensory perception of how the performer attacks the animal, the blood, the cuts and the death; and (2) phenomenal elements connected to this, namely, a first-person sensation of unease, the wrongness of spilling blood and the feeling that this should not be happening, or for others, a sense of magnificence.

Robert Audi, an American philosopher whose major work has centred on ethics, theory of action and epistemology, presents in his book Moral Perception (Audi, 2013) a theory of moral perception. He suggests that moral sensations, such as the perception of wrongness, do not arise in two stages (first observing the non-moral facts and then reasoning the moral conclusion), but rather arise alongside actual perception. They are an additional aspect of how an experience is perceived by the subject (assuming that moral objectivity is possible), similar to how a scene may appear threatening or how a shape or colour is a distinct feature of an object.

Across seven chapters, he sets out his understanding of general perception (Chapter 1) and moral perception (Chapter 2), the latter of which has a phenomenological component. He compares these with other types of perception and judgements, such as aesthetic ones (Chapter 5). He discusses the reliability of moral perception (Chapter 3), the sources of moral judgement (perception, Chapter 2, as well as intuition and emotions, Chapter 6), and cases of moral disagreement (Chapter 4). In the final chapter, he argues for the fundamental role of moral perception in normative ethics.

Audi argues that general perception involves observing and recognising an object in an experience. This recognition is ultimately a phenomenological experience (Audi, 2013, p. 13), in which the first-person sensation dictates the entire perception process (i.e. what we sense through our sensory systems and the emotions, intuitions and background knowledge associated with it). These phenomenally represented properties can be understood as propositions that hold truth values and are perceptually believed based on experience. For example, when we see part of a bull, we know that even if we see a 2D bull, it is not 2D. Thus, when we perceive something, there is sensory information (“I see one side of a thing”), doxastic content (“I think I am seeing a real bull”) and implicit content (“I know a bull has four legs”, or “I am afraid”).

But does moral perception follow the same pattern? Do we also recognise the wrongness or goodness of a moral act? Audi thinks so. He argues that the moral component lies in the perception itself, meaning that we recognise instances of morality (Audi, 2013, p. 31). Therefore, when we witness the performer wounding the bull, we are observing non-moral actions supervening on moral properties. These properties are linked to the act. In this case, seeing a barbed stick get into the back of the bull (non-moral perception), the physical behavior of the animal, the facial expression of the performer, all of that is what our senses pick up and what triggers (meaning is there) and anchors (meaning is fixed) the moral property (Audi, 2013, p. 55).

But which moral property, the wrongness of killing an innocent animal or the grandiosity of winning against a wild animal, is the real moral perception that we perceive or misperceive?

Unlike moral relativists, who believe that the same actions can carry different and relative moral considerations, Audi proposes a ground-based phenomenological reliability for true moral perception (Audi, 2013, p. 63). He posits that moral experience reliably generates true beliefs when properly formed, conceptually adequate and grounded in the correct natural facts. Similar to how ordinary vision counts as knowledge-producing not because it is trustworthy, but because the physical processes involved (e.g. how light hits the eye and the brain forms a belief) are reliable, this kind of moral experience, when well-formed, reliably tracks moral truth.

This conceptualisation of moral perception does not necessarily imply the formation of moral beliefs or judgements (Audi, 2013, p. 45). Furthermore, Audi suggests that it can precede concept formation (Audi, 2013, p. 46). A child may have a moral perception of hurting the bull, accompanied by a sense of unease, and later, as an adult, develop a moral conceptualisation of such an act based on this perception.

Within this perception, intuitions and emotions play a fundamental role in interacting with the moral deed. Similar to perception, intuitions are direct, non-inferential responses to something perceived (Audi, 2013, p. 89). Contrary to some authors who propose that intuitions are fast gut reactions (Haidt, 2001), Audi suggests that what characterises an intuition is not how quickly it arises, but the absence of real conceptual premises (Audi, 2013, p. 93). These intuitions can manifest as beliefs (‘I believe the bull is feeling fear’) or as mere ‘seemings’ (‘It seems the bull is angry’). Both can do justificatory work. In contrast, Audi argues that emotions are richer and more complex than intuitions. They have a cognitive part (believing that the bull is in pain), an affective part (feeling sad about that possible sensation), and most importantly a motivational part (the desire to avoid it) (Audi, 2013, p. 125). Both emotions and intuitions are only derivatively reliable. The beliefs on which they are based are reliable when properly grounded in relevant natural facts, as with perception (Audi, 2013, p. 137). Thus, a confused or ungrounded emotion or intuition carries no evidential weight.

But what happens when a person’s perceptions, intuitions and emotions conflict? And among different people? Audi suggests that, in the case of conflicting intuitions or emotions, the first port of call should be moral perception, the most direct and knowledgeable tool for moral decision-making. If needed, imagination can also be used. Audi believes that this capacity serves as a bridge between conflicting conceptualisations of the moral deed and infinite possible occurrences that transcend this same perception, and that it might help construct different pathways (Audi, 2013, p. 157).

On the other hand, how does Audi explain conflicting perceptions, intuitions and emotions between someone who sees the wrongness of killing an innocent animal and someone who sees the grandiosity of winning against a wild animal? Who is really perceiving the right moral component, and who is mistaken? Throughout chapter four, Audi argues that most disagreements we encounter in everyday life are not about the specific content of the moral act, but are indefinite (i.e. rejecting something without pinning down exactly why this cognitive proposition is in dispute) or illocutionary (i.e. rejecting something on the basis of the speech act itself) (Audi, 2013, pp. 71-75).

If the disagreement is specific to the moral act and the disputants are epistemic peers (i.e. idealised rational and equally informed and reflective), who is right? Audi suggests that this is rarer than it seems (Audi, 2013, p. 81) because most real disagreements involve asymmetries in background knowledge, sensibility, or degrees of moral education. As he says, ‘it is a matter of moral education’ (Audi, 2013, p. 75).

However, this leads us to focus on a less direct source of moral knowledge than perception, namely the ongoing process of moral education and sensibility. With this in mind, moral perception begins to resemble a faculty for seeing the world and a way of viewing moral deeds. Consequently, (1) some people will be more perceptive than others, (2) some of this perception can be honed, and (3) neither disputant can ever be fully certain that they are witnessing the correct moral act and not a misperception of the moral instantiation.

Audi does not explain how this moral perception can reveal the true moral trait to someone without proper moral knowledge. Therefore, he cannot solve the problem of how to guarantee that the moral education and training received is correct. Morality may lie in the actual object, but if our moral perception ultimately relies on the training and moral education we have received, we may lack the tools to determine who is right.

Can we accept Audi’s conceptualisation without falling into relativism? How can we be sure that the moral education we are receiving will enable us to perceive moral institutions correctly? Audi is not the first to place the give training and sensibility some resposibility. Aristotle, for example, is well known for believing that good actions (and thus good moral perception) come from practical reasoning when one is properly educated in knowing what to do, when to do it and under what circumstances (Aristotle, 2000).

In conclusion, when we perceive something, we are not just seeing it; we are also bringing to it some prior cognitive artefacts that play a fundamental role in how we use the faculty of moral perception. One person’s moral perception may be another person’s mere observation. Thus, our normative standards play a fundamental role in how we perceive things. Two people can have the same visual input and reach genuinely different moral conclusions, simply because prior sensibilities influence what our integrated experience brings about.