John Hick and Candomblé: The concept of religion and the experience of evil

Abstract

The article takes as point of departure the influential book An Interpretation of Religion by John Hick both to understand an Afro-Brazilian religion called Candomblé and to suggest a way to define this human activity as a whole. After a description of Hick’s proposal, his ideas will be applied to Candomblé to see what kind of religion it is. From the problems raised in this classification, the article puts forward two formal refinements in Hick’s proposal, which can increase its internal coherence as well. In addition, in view of a salient feature of Candomblé, it is argued that the concept of religion should include the experience of evil as an essential element. The resulting definition is justified by its explanatory power, amplitude and fruitfulness.

Introduction

This article has a twofold goal. On the one hand, it intends to offer aids in conceptually understanding Candomblé, one of the most influential Brazilian religious traditions.¹ In so doing, I join a laudable recent effort of including the consideration of more diversified case studies in the philosophy of religion debate. The second goal of this article is to help expand and enrich this sub-area of philosophy. Specifically, a concern in the philosophy of religion in which this article hopes to help is the very understanding of the concept of religion itself.

One of the most influential works in contemporary philosophy of religion that has dealt with this concept is An Interpretation of Religion – Human Responses to the Transcendent (1989/2004) by John Hick, which will be taken as the starting point for achieving both aims stated above. As will be shown in the first section, Hick’s proposal amply deserves such a special consideration. One of its qualities is that, apart from having tested his concept against very important modern religions, it provides insightful interpretative keys for understanding Candomblé as this type of phenomenon as well. However, this article intends to improve on Hick’s definition. After a critical analysis of his method of definition, I put forward a two-level concept of religion, which takes into consideration a notion of evil as part of the first level, and the family resemblance idea as the approach to the second level.

Given this bibliographical main reference and the targets aimed at, this text is divided into three parts. First, we will have a look at Hick’s proposal regarding the concept of religion. Then, in the second part, we will see to what extent (if at all) Candomblé can be included in that concept. In the last section we will elaborate on the notion of religion with the help of Hick and Candomblé in a sense apparently not explored by him, and which (I hope) may contribute to a better understanding of that idea.

Hick and the concept of religion

Hick’s aim in An Interpretation of Religion is to comprehend the diversity of religions with a concept that may be both unifying and also do justice to the nuances and variety of this complex reality. Sceptics about such conceptualisation may criticise this kind of project, saying that, in addition to being very difficult to implement, one may challenge its very importance. These are good questions, mainly the first one, since the enormous diversity of phenomena recognisable under the label ‘religion’ is a clear fact. The importance of having a definition for religion seems less questionable. After all, the question ‘what are we talking about with the word “religion”?’ seems to be syntactically well-formed and to make sense. In addition, a clear and applicable concept of something is certainly a contribution to the intelligibility of it. I am not claiming that having a concept of x is a necessary condition for knowing x, but that it can be helpful in some contexts. For instance, it may be useful for anthropologists and historians of religion to better understand what they are studying, and where to start from.² Of course, their empirical research can serve to criticise the proposed definition by showing it does not apply to all prototypical cases we have in mind when we use that word or to new religious phenomena they may come to discover. In any case, it seems sensible to say that if philosophy can still collaborate with the empirical sciences of religion, one of its duties would be to fashion and delineate concepts.

As his book’s subtitle itself says, Hick’s idea is that religion is the set of human responses to the transcendent. This way he intends to include in the same concept of religion both naturalistic approaches – that see it as a human activity only – and the position of those who take part in it, who claim that there is an experience of something special or sacred, and that is not reducible to a more common origin. Hick’s concept accepts that religions are human constructions, but also admits that there may be a non-human reality with which humans try to communicate.

Another important point in Hick’s proposal is methodological. His concept of religion does not intend to be the expression of an immutable essence, but rather of family resemblances among different members, which is an adaptation of the famous suggestion of Wittgenstein on how to understand the notion of game in the Philosophical Investigations. By taking the notion of religion as a set of related qualities, without requiring that all concrete religions display all of them, Hick claimed to be able to accommodate more examples while keeping his concept still clear and operational.

Even so, Hick attempts to make the concept more precise by distinguishing basic types of religion and by focusing on some relevant examples. Part of the strategy he adopts to delimit the concept is to distinguish between two broader types of religion, according to a notion borrowed from Karl Jaspers in Vom Ursprung und Ziel der Geschichte (1949), namely, that of the ‘Axial Age’. According to Jaspers, this period was a time in human history (between 800 and 200 BC) when most of the main ideas behind current world religions arose. In this period of about 600 years, we had the appearance and the influence of Lao Tzu and Confucius in China, Gautama and Mahavira in India, Zoroaster in Persia, great Hebrew Prophets (Amos, Oseas, Jeremiah, Isaiah, Ezechiel) in Israel, Pythagoras, Socrates, Plato, and Aristotle in Greece. From the contributions of those great Axial Age sages, we can distinguish between post-axial and pre-axial religions, which give us the broadest classification of religion in Hick’s interpretation.

Those that are post-axial are called by Hick ‘great world religions’, and include Judaism, Christianity, Islam, Hinduism, and Buddhism. According to him, thanks to those seminal thinkers, scattered across Eurasia, we saw the emergence of individual self-consciousness in post-axial religion, instead of the more communal spirit of pre-axial religions. In addition, those influential characters of the Axial Age conveyed universal messages, instead of merely local ones, and emphasised a process of ‘inner’ transformation, instead of an emphasis on ‘external’ cults.

Hick calls these great world religions, which emerged through the influence of the Axial Age ideas, religions of salvation or liberation. In salvation religions (mainly Judaism, Christianity, and Islam) as well as in liberation religions (Hinduism and Buddhism), mundane reality is taken as defective in view of a more perfect transcendent state. Salvation or liberation consists in the overcoming of worldly defects.

As a result, the general scheme of the doctrine and practice of post-axial religions would be something like:

1. 1) The recognition of human weakness and the deficient character of this world;

2. 2) The proclamation of an infinitely better alternative in a transcendent dimension;

3. 3) A way of salvation/liberation (or how to go from 1) to 2)).

According to Hick, the different ways of salvation or liberation proposed by the great world religions are variations of the same theme of human transformation (see Hick 2004, ch. 3). By this, he means the individual change from a self-centred focus to a new centre in the unity of reality and transcendent value, which Hick shows in detail in each of these post-axial religions.

Pre-axial religions should be described according to other values and ideas. Their main concern is to keep the cosmic equilibrium by means of rituals and sacrifices, so as to avoid chaos and to ensure the conditions of life. It means this world is not denied, but should be preserved, enjoyed, and enriched. The world is full of gods, ancestors, spirits, and the like, who are responsible for world order and for what happens in human life and the cosmos in general. These beings need to be respected and appeased so that threats to world equilibrium may be countered.

So, although Hick’s reflexion was focused on post-axial religions, which may be justified by their demographic importance today, the distinction he suggested, based on Jaspers’s proposal, aims also to reach ancient and contemporary religious manifestations so different from the five great world religions.

Is Candomblé pre-axial?

Hick’s considerable efforts have been widely recognised. The merits of his book justified its translation into many languages, including Portuguese. In the preface written to the Brazilian edition (Portugal 2018), the question was raised of considering the Afro-Brazilian traditions in the task taken up by Hick of understanding the religious phenomenon in a philosophical way. The project with which this article wishes to collaborate is an opportunity of revisiting that idea, attempted before in another paper (Portugal and Carvalho 2009). So, let us start with a brief description of Candomblé, based upon several empirical investigative writings and anthropological interpretations,³ apart from my direct observation.

In Candomblé, existence is a gift by Olorum (Olodumare), creator of the world, the universal deity or the supreme being.⁴ No matter how we describe Olorum, the fact is that there is no liturgy or religious activity related to him in Candomblé, and references to him are very rare.⁵ Cult and other religious works are devoted to the orixás (Exu, Oxalá, Ogun, Oxun, Oxóssi, Iemanjá etc.), who are intermediary deities, each one with a special quality, both moral and physical. Every human being is thought to be affiliated to an orixá, from whom one is meant to expect protection, existential model, and guidance, and to whom one has obligations.

In this religion’s metaphysics, reality is made up not only by facts, states, and events explainable in material and physical ways, but also (in fact, mainly) composed of spirits, on whom everything depends. The world is mostly spiritual, that is, the ‘physical’ dimension is secondary, since what really matters are the intentions that motivate what happens. In consequence, evil (such as suffering, disease, misfortune) has mainly an intentional cause, due to spiritual or human purposeful action.

Although doctrinal preaching and moral debate are not absent, one may say that Candomblé is predominantly a ritualistic and mystical religion. Nago religion (as it is also called⁶) is a complex system of rituals, myths, and interpersonal interactions aimed at keeping and renewing axé, the energy that is behind everything. Axé is not just physical energy, since older members of the community may be privileged bearers of this force, given that they have had more years of life (which is what axé principally represents) than the younger ones. The vital energy of axé is constantly threatened by malicious intent, illness, and death, and this is why Candomblé is so concerned with its strengthening. It involves activities such as trances and incorporations of the orixás, animal sacrifices, initiation processes, tabus, offerings, chants, dances, and participation in a community according to traditional roles, following seniority and oracular prescriptions.

Axé may be renewed in different ways of making deals with the orixás through offerings, ascetic interdictions, or purification rites. Nonetheless, the main rite through which axé is preserved and reinforced is sacrifice, particularly the slaughter of an animal whose blood has a special ritual value because of its vital force. These rites are collective enterprises, and there is a strong submission to rules and hierarchies laid down by the orixás. Candomblé communities are true ‘spiritual families’, providing reception and protection, but demanding obedience to their precepts and respect for seniority. However, each person has his or her orixá, to whom one is linked by personal characteristics, and to whom one owes individual duties.

So, to answer the question whether Candomblé should be understood as a pre-axial religion according to Hick’s classification, we might first consider the concept of the Axial Age itself. The very characterisation of this period and even the time span and the geographic loci it comprises, is a matter of controversy (see Mullins et al. 2018). In other words, Jaspers’s periodisation and localisation of this phenomenon has not been met with general consensus. In addition, historically speaking, it is questionable whether some of those founding characters (Lao Tzu and Zoroaster, in particular) really existed and if there was a genuine connection between their ideas. However, even if poorly founded in history, the distinction can still play an instrumental role in a philosophical definition.

But more importantly for us is that, although Candomblé seems to display mainly features of pre-axial religions, some post-axial qualities are clearly present as well. Let us recall that, for Hick, while pre-axial religions are concerned with preserving the world, post-axial ones aim at salvation or liberation from it, through a path that takes us out of a world understood as mostly unsatisfactory. While pre-axial religions turn principally to animal sacrifice and ritual activities, post-axial ones are interested in the inner transformation of the individual. Whilst pre-axial religions are mostly communitarian, with little or no appeal to individual consciousness, post-axial religions are turned mainly to the unique subjectivity of each human being.

Now, despite being very concerned with the proper performance of its rituals, Candomblé is also turned to physical and spiritual transformation of the individual in a path that involves imitation of the model of each orixá, and thus taking part in its power. Although it may seem that the (external?) worship dimension prevails, inner transformation is also valued in Candomblé. Altruism is praised, and egotism – manifested in disobedience or disrespect, for instance – is reproached. Despite the fact that hierarchy is rigid, and that there is a submission of the individual to the collective, those features of Candomblé communities do not remove individuality but actually presuppose it. In this Afro-Brazilian religion, in an invisible way, the dead live close to the living. Each individual soul is immortal, and may or may not reincarnate, apart from staying alive in the memories of descendants, relatives, and friends. In addition, side by side with the individual dimension of afterlife, the fate of the dead and of the living are tied in a communitarian eschatology, which is both personal and collective.

As a result, the best answer to the question posed in this section seems to be that Candomblé displays characteristics that belong both to pre-axial and to post-axial religions, following Hick’s classification. In the next section, I will try to draw out two suggestions to the philosophy of religion from this brief discussion about Hick and Candomblé.

Candomblé, the concept of religion and the experience of evil

A formal suggestion

The description of this Afro-Brazilian religion presented above raised some problems as to whether it would be a pre-axial religion only. As we saw, although Candomblé shows some features of pre-axial religions, it also bears characteristics of the post-axial. This is something Hick might admit, since his distinction did not mean to be strict and exclusive but allowed some room for overlapping and interactions as the parenthetical ‘but not solely’ condition in the following quotation seems to state: ‘The second widely accepted large-scale interpretive concept is the distinction between pre-axial religion, centrally (but not solely) concerned with the preservation of cosmic and social order, and post-axial religion, centrally (but not solely) concerned with the quest for salvation or liberation’ (Hick 2004, 22). As a result, a religion that one would take as clearly post-axial could exhibit pre-axial traits in some of its manifestations as well, and vice versa.

Given the problems of classifying specific religions as either pre-axial or post-axial, a first suggestion aiming to improve Hick’s concept of religion is pragmatic, that is, concerned with what we can do with the definition we get. Instead of serving to classify a religion in its totality only, the distinction could also be helpful to understand specific manifestations in each religion or even a certain behaviour of a given religion’s follower. So, the ideas philosophers of religion suggest for understanding this phenomenon from a conceptual point of view could be useful for clarification of what happens in each religion or to each participant of them, in addition to being an attempt at comprehending what makes a certain activity to be religious in general.

In fact, if we take seriously Hick’s own methodology of speaking in terms of family resemblance instead of a univocal concept that should fit entirely to each religion, the features that stem from the distinction between post-axial and pre-axial should also be considered in this way. In consequence, the fact that Candomblé possesses points which belong to both these main types is not a counter-example to Hick’s approach, despite being a problem in a more rigid typification. The clear distinction between pre-axial and post-axial religions we get from his book does not really cohere with his own method of conceptual construction.

Instead of defining religion by means of a single characteristic, the family resemblance approach defines it by means of a set of characteristics, which do not need to be present in all its instances to the same degree. This set of qualities is the definition of the type through which we give meaning to a certain entity, and through which it can be understood. In the end, the set must exclude some individuals to help us understand what we are talking about with the word ‘religion’, because if it includes too many qualities and exemplifications the definition will not be able to provide any intelligibility. So, as it is traditionally said, the task of definition lies between the risk of including what should not be included, and of excluding what should be included. The family resemblance approach does not seem to be exempt of this risk, that is, if its definition excludes too many instances it runs the risk of being parochial and ethnocentric, and if it includes too many, it leaves us in the dark about its meaning and will not contribute to our comprehension of the subject.

The solution Hick proposes to this traditional challenge regarding definition is dubious and paradoxical. According to him, ‘given this family-resemblance understanding of the concept, different scholars and communities of scholarship are free to focus their attention upon the features that specially interest them’ (Hick 2004, 5). Yet, it is doubtful whether this approach really permits this kind of move, since the main reason it was proposed in the first place was the admission that a certain object or state of affairs would be better understood in view of many different qualities equally important for its meaning and comprehension, although not all of them are displayed by all exemplifications. Even so, Hick’s choice is to concentrate his concept of religion on the idea that lies behind the subtitle of his book, that is, religion as a human response to the transcendent. It is a feature that could allow us to exclude things like Marxism or a Darwinian-inspired atheism as religion. Since they deny a spiritual dimension to which human beings could be in relation, these movements should not be classified as religious, despite featuring some other characteristics we may find in the religion family (the formation of a community around some ideas about the meaning of human existence, and a way to deal with evil, for example). However, he does not exclude them from the ‘religious family’ but take them as ‘distant cousins’ (see Hick 2004, 5), which has the puzzling result of including among religions some movements that criticise, reject, and call for the abolition of religion.

So, given what we can draw from Candomblé regarding the distinction between pre-axial and post-axial religion, and the reasoning developed about Hick’s family resemblance approach in view of it, I suggest an amendment to his method of definition. The goal is to combine the advantages of a more flexible set of characteristics needed to accommodate more religious phenomena with an idea able to do the semantic and epistemological job we expect a definition to do. This amendment of Hick’s method divides the definition in two levels, the level of contingent or non-essential characteristics (individually speaking), which may appear in some instances but not in others, and the level of essential qualities in the sense of appearing in all human manifestations one could call ‘religion’. That might strike us as an unusual combination of two approaches that normally are opposed, but this does not mean they are contradictory. Actually, the strategy of putting them in two different although related levels is a way to prevent this putative incoherence. In consequence, Hick’s notion of human responses to the transcendent – with a major addition and a small qualification, as we will see shortly – could be taken individually as an essential element of the definition of religion. On the other hand, a set of several qualities such as showing the attributes normally featured either in pre-axial or post-axial religions, having rites and mythological narratives, supposing a doctrine and a behaviour which is accepted and cultivated by a given community, and so on, would be part of the definition as well. The whole set would be essential to the meaning and comprehension of ‘religion’, but each of its members do not need to be displayed in each exemplification of it.

In fact, this way of accommodating Hick’s pre-axial and post-axial distinction makes even clearer one of his points. Hick avers that there is no progressive sense in the idea that a new type of religiosity emerged in the axial age.⁷ One may not say that post-axial religions are better than pre-axial ones unless a precise criterion for assessment is specified. It should be noticed that Hick provides a criteriology in the last part of his book, using as criterion of evaluation the ethical consideration of overcoming egotism, and placing the centre of everything in the transcendent ‘Real’, as he calls the common reference in all world religions' worship.⁸ This stems from his preference for the post-axial religions, which we saw as incoherent in the face of the family resemblance approach he adopts. So, my suggestion has the additional quality of being more in keeping with the rejection of the idea that there would be progress in post-axial religions as regards the pre-axial ones.

The literature on Hick is immense, even if we focus only on An Interpretation of Religion, which is a measure of its importance. However, a survey of it shows that little attention has been given to Hick’s method for defining religion. By far, most criticism is directed to his pluralist hypothesis.⁹ A second target has been Hick’s epistemology and the eschatological verification of religious language.¹⁰ In the volume edited by Harold Hewitt (1991), most articles about An Interpretation of Religion are focused on Hick’s pluralist hypothesis as well. Some also deal with his thesis that the world is religiously ambiguous, and that world religions have a common moral ground. The questions I explore here were seen as less controversial. This is the position assumed by Chester Gillis, for instance, according to whom Hick improves (in Gillis’s view) his previous definitions of religion by accommodating archaic religions, new religious movements, and world views like Marxism in the family resemblance approach. On the other hand, Gillis argues that Hick ‘prefers to focus on those religious expressions that acknowledge transcendent reality, and that is his legitimate prerogative’ (Gillis 1991, 30). As I said above, I find this in conflict with the family resemblance approach, and the definition suggested by the book’s subtitle rather than a ‘legitimate prerogative’.

A deeper analysis of Hick’s definition of religion I found in Byrne (1991). He also sees an incoherence in Hick’s proposal, but of a different sort than the one I pointed out. For Byrne, the conflict lies between the family resemblance approach adopted by Hick, and the search for generalisations about religion that Hick includes in his project (Byrne 1991, 126). The family resemblance definition of religion presupposes a lack of essential unity among the members of this class, which is a clear way of respecting the varieties of the religious phenomenon. Hick wants to take this fact into consideration, but he also postulates an ethical essence uniting all post-axial religions. Indeed, the very idea of a common origin behind the great world religions of today is explained by Hick based on the ethical core in this phenomenon (Byrne 1991, 127). In the end, the incoherence in Hick’s definition is resolved by means of the distinction between apparent superficial diversity and deep essential ethical unity. Yet, Byrne argues, despite being useful to postulate a common essence to all religions, Hick’s move faces some important objections. First, it is normative, that is, it imposes to the religious phenomenon a set of values it should abide to, and according to which they are going to be judged. This is problematic especially because the core of religions has to be identified with a common pattern of behaviour. Second, accounts of essence work by abstraction and generality, which ends up disregarding the specific elements of each religion, and not leaving any real likeness among them. A third objection is that ‘the philosophy that lies behind the theory is not derived from a neutral phenomenological survey of religions but comes instead from metaphysical postulates about the ultimate validity of all major faiths’ (Byrne 1991, 128).

I believe the amendments I am suggesting are not affected by Byrne’s criticisms. As to the conflict between the family resemblance method of definition, and the postulation of a unifying essence behind the varieties of the religious phenomenon, I propose a way of conciliating them, which aims at eliminating the problem. By dividing the definition into two levels, my proposal combines the essentialist and the family resemblance methods without falling into incoherence. In addition, since my definition does not resort to a common ethical core in all religions, it is not clearly normative as Hick’s idea. Although I do not see anything wrong in distinguishing good and bad religions according to a given moral criterion, my aim is not providing a basis for judging instances of the phenomenon but, rather, collaborating for their understanding. As to the other two of Byrne’s criticisms of Hick, I postulate that the phenomenology of religion presupposes a general concept that philosophical analysis should clarify. I do not see the search for an abstract general concept as an issue. It is a fallible enterprise of course, but this is not a reason to refrain from attempting.

The final result of this methodological proposal will be shown just before the conclusion of the chapter, but to do it we need first to include some elements in the content of Hick’s definition.

A content suggestion

Candomblé was born in Brazil. Rites and myths, which were practised and told in separated tribes each one linked to a single particular orixá in Africa, became part of the worship of many orixás by the same community. Apart from this syncretism internal to the African matrix, Candomblé also combined some elements of Christianity, linking the orixás to Catholic saints and incorporating some of its moral values to the African traditional ethical patterns.¹¹ Its syncretic character, by the way, may be an important historical cause of the presence of both pre-axial and post-axial elements in their religiosity.

Yet, what is in the root of its origin, and that must be faced by its adherents even today to some degree, is what we could call a strong and vivid experience of evil. Candomblé emerged in the context of the European trade of African slaves starting in the late sixteenth century, which lasted for almost 400 years. Intense physical pain, disease, early death, rape, humiliation, murder – as consequences of treating human beings as tradeable objects – were a constant part of this context in which and to which Candomblé began as a response in religious terms. Nowadays, although slavery is no longer legal, these communities still face aggression from other religious groups – they are the main target of religious violence in Brazil today – as well as prejudice and racism from many religious and non-religious people in Brazilian society. Candomblé is a religion of resistance to the experience of evil.

However, even though it is a strong and vivid experience, evil is not taken as the only and defining reality. The renovation of axé through sacrifices and obligations, and the very presence of the orixás, are part of the day-to-day experience of Candomblé religious practice. Orixás and other spirits are perceived as both transcendent, since they are not objects of sensorial perception, and immanent, given that they become incorporated in some of its participants in the rites, and their activity is experienced as part of the world. The positive experience of axé through the interaction with the orixás comes first and gives them the power and meaning to deal with the negative experience of evil. So, deliverance from evil through the religious way is made possible by a relationship with an invisible power, which is not only strongly felt but also provides the phenomenological and religious background in which evil is dealt with.

The use of the term ‘experience’ is important for the thesis I am sketching here. A way to clarify my point is to contrast the concept of evil as a human experience with evil as an objective reality. From a metaphysical point of view, evil would better be understood as depending on a positive reference in order to exist. In other words, as the traditional privatio boni theory of evil defends, evil is a lack of something, a defect, not something in itself. There must be something positive first so that it may be manifested. Finitude, corruption, insufficiency, and annihilation are well understood as privations of something that is or should be in the first place. Finitude is a quality of something that has either emerged or will end at some point. Corruption is the deterioration of something that was meant to be in a certain way but lost its property or some degree of it. Analogous analyses seem adequate to insufficiency and annihilation so that they can also be shown to suppose something to be characterised as a denial of something else. On the other hand, physical evil can be described in (modern) scientific terms so that the evil part can be explained away. So, pain can be analysed as an activation of certain neuronal fibre resulting in a particular nervous stimulus. Human death occurs when the heart or the brain (or both, depending on the criterium prevailing) stops functioning. The destruction of Lisbon due to an earthquake in the eighteenth century happened because of tectonic activity in the area close to where this city was (and still is) situated and to the type of construction it had at the time. It is not only that there is nothing empirically testable in claiming that pain, death, and a city destruction are evil things. We do not need to include the quality of being evil to explain them in those terms, and even to create means to prevent them. In addition, moral evil may also be explained away in an evolutionary approach to morality or in a positivistic perspective of law, for example. So, lying would be the intentional utterance of a falsity, which is harmful to the adaptation of a human group, since humans need communication for their success as a species; it should be reproached because it is non-adaptative, not because it is evil. Murder, rape, or disrespect may be analysed as facts that may be phrased in empirical propositions that are typified in normative propositions in a penal code, which prescribes punishments for those acts so described. As a result, even what is normally called ‘moral evil’ can be accounted, explained, and dealt with so that we do not need to qualify the corresponding acts as such. Still, even if we do not take it as a metaphysical reality in itself nor as scientifically objective, evil is an essential part of human experience.¹² This means that, when we feel or perceive something like annihilation, pain, or rape, we experience them in a negative way, instead of only neutrally describable by metaphysics, science, or law.

Let us elaborate a little more on this notion of the ‘human experience of evil’. The content of the conceptual background with which we experience a given phenomenon this way may vary, and there may be specific situations or events that are taken as evil by some, although not by all. However, whatever the cultural or historical value with which particular occurrences may be judged, we cannot help seeing at least some of them as evil. When we have perceptual or emotional contact with these events, these mental states are tied to an evaluation which takes them in a negative manner that we call evil. If so, at least the evaluative component would explain the relative diversity of specific judgements of evil. Nevertheless, this specific relativity does not prevent us from claiming that evil is a universal human experience, whatever the value content it may have in the background. Basic, universal emotions expressing negative reactions such as fear, aversion, anger, and sadness also play an important part in the phenomenology of this human experience. They are not mere internal feelings, since they have physical, perceivable components like alteration of heartbeat, tears, and the like, which show how deep this experience can be.¹³ The idea is that we perceive some states of affairs or some deeds as evil, as a quality that defines them in the context they happen and given our ontological constitution. This experience is so rooted in our way of being that it elicits emotional manifestations in particularly strong situations. So, the human experience of evil is universal both because we cannot help evaluating some facts and events as evil, and because we experience them as emotions.

Since evil is a universal human experience, we may say that if you do not react to or perceive any deeds or events at all as evil, then there is probably a problem with you. Even leaving generous room for cultural and historical relativity, it is justified to claim that not considering as evil the kinds of happenings suffered by the African slaves who founded Candomblé (the case study we are starting from here) is a symptom of cognitive disfunction. The immorality of not perceiving these events as evil has an epistemological component as well. If you do not take them as evil, it means you do not know how to process the experience you have had correctly. In other words, the experience of evil is not only a central component of human metaphysical constitution, but also an important element of our epistemic capability.

In view of the above considerations, my second suggestion to Hick’s concept of religion is that it is not only a human response to the transcendent. In fact, to avoid being too restrictive, due to failing to embrace some religious traditions that do not make a sharp distinction between natural and supernatural, we should put it as a matter of degree. As we saw it, Candomblé participants do not perceive the orixás and axé as wholly transcendent, but also as very immanent, and this is the case with some other religions too. We then had better say they are human responses to an experience of being (i.e., of reality in general) that is taken as transcendent to some degree. Besides, in addition to being a response to that kind of positive experience of being, I suggest that religion is also a response to the human experience of evil, for the reasons I stated above. So, the first level of the definition is that religion is a response both to an experience of being – perceived as in some degree of discontinuity between common and uncommon (see King (2005)) – and to an experience of evil in view of the properly religious qualities described in the second level.

The second level of the definition gives more precision to what we call religious. This specific element would be a set of qualities, which is essential as a whole, but whose elements are contingent individually. The extension of this set is given by the properties Hick ascribes to both pre-axial and post-axial religions, which summarise features such as what Winston King (2005) called ‘structures of religious life’ (a reference tradition, mythical narratives and elaborated symbolic language, concepts of salvation, sacred places and objects, rites and liturgies, sacred writings or oral narratives, sacred community, mystical experience at different degrees) or what William Alston (1967, 141–142) named ‘religion-making characteristics’ (belief in gods, a distinction between sacred and profane objects, ritual acts focused on sacred objects, a divine moral code, feelings aroused in rituals or other types of interaction with the sacred, prayer, a world view with responses to foundational or deep existential questions, which gives meaning and organises one’s life, a community around all these elements). These lists are very similar, but there is no room to argue for it here. As to the question of how many of these items must be present in a human activity to be considered religious, it is difficult to say, but perhaps we should remember here Aristotle’s advice that we must not expect more precision than the subject-matter admits.¹⁴

Summing up, for the proposal this article is formulating, religion should be defined in two levels as:

Level 1: a human response to an experience of reality perceived as having a special or transcendent dimension in various degrees AND a response to the human experience of evil in view of the elements of level 2

AND

Level 2: a set of qualities whose elements should be present in a human activity to characterize it as religious such as the properties, activities and institutions Hick describes in his accounts of both pre-axial and post-axial religions.

Conclusion

This article took as point of departure John Hick’s acclaimed book to discuss both Candomblé as a religious tradition and the concept of religion in general. After describing Hick’s proposal and Candomblé’s main characteristics, the article argued that this Afro-Brazilian tradition showed traits of Hick’s characterisation of both pre-axial and post-axial religions.

In addition, I suggested three amendments to Hick’s definition of religion, one pragmatic, one methodological, and another regarding content. The pragmatic suggestion was that the concept could be used to help understand individual practitioners and particular groups of a religion, apart from being useful for illuminating the religious phenomenon as a whole. The methodological proposal was to divide the definition in two levels, so that we could combine a general part, purportedly applicable to all religions, with a specific part, composed by a set with various elements. This way, the definition could benefit from the advantages of the family resemblance approach Hick embraced and be more in keeping with it as well. The content suggestion had two facets. One was to introduce a gradation to Hick’s idea of religions as human responses to the transcendent to the effect that the extent to which the positive experience of being may vary in intensity among different religions in terms of their discontinuity between the common and the special. The other was, in view of the Candomblé example (given its historical background, the violence they suffer today, and the constant need of renewing axé through their rites and practices), to consider religion also as a human response to the negative experience of evil. This double experience is expressed by means of a set of specific elements which define the religious phenomenon – although they do not need to all be present in all religions individually.

A consequence of the idea of religion as a response to the experience of evil could be that this type of human activity may be criticised as being somehow dependant on evil to survive. This might be a reason why those who face it more intensely (the economically vulnerable, the seriously ill, terminal patients, etc.) are more prone to display a more religious attitude. Nietzscheans would criticise this as an exaggerated emphasis on suffering, which is a disposition to rejecting life, as an unacceptable ‘the worse the better’ outlook. Marxists would see it as a basis for their thesis that religion is a narcotic that alleviates pain instead of resolving it. Freudians may take it as confirming their idea that religion is doomed to disappear as soon as we find alternative means to deal with evil. Of course, the defender of religion may reply that the way it deals with evil does not deny life, since the experience of evil is responded in view of a deep positive experience of being perceived in a special way; that religion is not necessarily politically alienating, but that it could be an additional motivation for the political fight; and that other ways to deal with evil (like science or technology) are not necessarily contradictory to the religious form. Anyhow, if the concept I am suggesting makes sense of these classical common criticisms and responses to them, it does not mean it endorses any of the parties. Actually, this allows us to conclude that the definition can help to understand a little better some important debates in the philosophy of religion.

In addition, the proposal advanced here could be fruitful to deal with two other issues in the philosophy of religion apart from the concept of religion. One of them is the problem of evil itself. As an experience, evil could be taken as both a reason for disbelief (as in Gellman 2013) or a reason for increasing religious belief as a way to give meaning to such a universal and puzzling phenomenon (as in Adams 2013). Another important question that could be approached from this idea is the relation between religion and other human activities, such as science, politics, art, or philosophy. These would also be – each in its own way – responses to the human experience that is evil, and this could help understand both what links them to and what separates them from religion.

So, apart from its explanatory power and more adequate scope (in terms of what it both excludes and includes), the definition above can prove fruitful to deal with some traditional problems in the philosophy of religion, and perhaps to open the field up to new horizons as well.