Scholars of religion generally agree that belief is a Western, Christian, and even Protestant construction that obscures more than illuminates the lives of those who do not use the term themselves. Rodney Needham, for instance, argued that while some hypothesized psychological states, such as imagination, are indeed based in panhuman capacities, belief is not. Instead, our idea of belief is merely the product of the word “belief” and the conventions that grew around it through its lexical and ideational history in Judaism and Christianity. Malcolm Ruel further argued that the notion of belief as an emotional commitment to a set of values has been secularized and now extends to belief in human rights, democracy, and rationality. In a more recent treatment, anthropologists recognize some value in the term, but recommend thinking “against” rather than “with” belief in ethnographic writing.

The arguments against unreflectively using the term belief in cross-cultural work are convincing and have been discussed and illustrated by numerous anthropologists. My own experiences of the problems with Western ideas of belief took place in Tokyo, Japan in 2018. There, one of the most common responses to being asked whether one believes in the existence of kami was “I don’t know, I’ve never really thought about it,” even by those who routinely visit shrines and temples. Religious practice requires no deep emotional commitment to the existence of the kami, though many would, if pushed, assent to their existence.

But to move from cautioning scholars that many of the Western assumptions about the nature of religious beliefs do not hold across cultures to jettisoning the possibility that there is a real psychological phenomenon (or phenomena) with some significant overlap with our folk notion of belief is unjustified. It would be to throw out a number of babies with the bathwater of the term belief.

This view comes from my disciplinary position within the cognitive and evolutionary anthropology of religion, where the majority of scholars have little problem using the term belief as a scientific object of analysis. This is not because they are unaware of the arguments of Needham, Ruel, and others (though many are), but because the term belief in the cognitive sciences did not inherit the problematic Western assumptions discussed above.

While only rarely systemized or precisely defined, something very much like belief has been an important element of cognitive science since its beginnings in the mid-twentieth century. With early computers demonstrating the capacity of material substances to take in inputs, process information, and produce outputs, and with subsequent developments in neuroscience, comparative psychology, and philosophy, researchers in the cognitive sciences routinely found themselves needing to understand what information a system or organism is assuming to be true as it acts in the world. Reading this literature, one can infer that the most widely held concept of belief in the cognitive sciences is treating some information as true in the generation of further thought and behavior.

In the cognitive sciences, then, belief is a general, and perhaps even necessary, feature of cognitive systems and, consequently, pertains to a much broader array of phenomena than the entities and processes traditionally thought of as religion. For some, such a minimal notion of belief may appear uninteresting and irrelevant for investigations of religion and secularism. However, the cognitive science of religion reveals that understanding how this more vanilla version of belief works has relevance for many of the phenomena studied by scholars of religion.

One example of this stems from the cognitive scientific insight that, just as there are distinct substances that share the label “jade” (jadeite and nephrite), there are distinct psychological types of belief. Or, in other words, there are distinct ways in which minds treat information as true. While the distinction has been made in subtly different ways, a key finding of the cognitive sciences is that some beliefs are typically produced unconsciously and operate quickly and automatically as we negotiate our worlds (e.g., people have minds, unsupported objects fall) while others are produced via slower, conscious deliberation and may not always have a strong impact on further thought and behavior (e.g., Paris is the capital of France, objects can spin in multiple directions at the same time).

This insight is relevant to the study of religion because 1) reflective beliefs often diverge from intuitive ones; 2) some intuitive beliefs may be very difficult to eradicate, regardless of reflective beliefs; and 3) some reflective beliefs are very difficult to hold intuitively. For example, Justin Barrett has shown that while reflectively believing God is omnipotent and omnipresent, Christians intuitively believe God is limited in space and time. Similarly, Deborah Kelemen and colleagues have shown that while reflectively believing in a mechanistic universe, Ivy League science professors intuitively believe natural processes exist for a purpose. Further, Shaun Nichols and colleagues have shown that monastic Tibetans, while reflectively believing in the lack of an enduring self, nevertheless have an intuitive belief in its existence and show significant levels of fear at the prospect of its annihilation. Nothing in this work necessitates that we view one type of belief as more “real” than another. Minds treat ideas and information differently for different purposes. This work on two different types of belief can help make sense of various field observations, including Tanya Luhrmann’s observations of Vineyard Christians having firm reflective beliefs in the presence of Jesus, but finding it very difficult to turn these into intuitive beliefs.

This example of intuitive versus reflective beliefs may strike some scholars of religion as a cold, outdated cognitivism that mistakenly treats religious beliefs as literal claims about what is true about the world, rather than as symbolic social commitments. Even if the Western construction of a deep, exclusive, emotional belief in some transcendent order isn’t a natural, cross-cultural phenomenon, there are many people strongly committed to the reality of a transcendental order and cognitive accounts of reflections and intuitions may seem to have little relevance in understanding their commitments.

While the distinction between intuitive and reflective beliefs alone may indeed have limited relevance to understanding such committed beliefs, other emerging accounts of belief in the cognitive sciences have significant implications. Philosopher of psychology Neil Van Leeuwen, for instance, has presented evidence for a partitioning of beliefs that directly bears on the matter of symbolic social commitment and, further, raises important questions for how we think about the influence of the content of a religious belief system on the sometimes extreme actions of its followers. Van Leeuwen outlines the different causal origins and behavioral effects of what he calls “factual beliefs” and “religious credences.” The opposition isn’t an antireligious argument about religious beliefs being false. Rather, factual beliefs 1) guide behavior in all practical setting in which their content is relevant; 2) cognitively govern other attitudes and inferences; and 3) are involuntarily prone to being extinguished if in conflict with perception or seen to lead to a contradiction. Religious credences, on the other hand, are none of these things. Instead, religious credences 1) have a normative orientation, such that believers view actions to be virtuous when guided by them; 2) are susceptible to free elaboration, such that believers can generate further ideas about the content of credences without clear support from induction or deduction; and 3) are vulnerable to special authority, such that credences are prone to extinction if seen as in contradiction to the dictates of the same special authorities from which they themselves came. For Van Leeuwen, religious credences are more psychologically similar to pretend play than to the web of factual beliefs our cognitive systems hold together to consistently navigate our worlds.

Van Leeuwen’s work has attracted criticism from those who argue that, for the “true believers,” religious credences are indeed treated as factual beliefs, thus explaining their willingness to fight and die for them. But I find his argument convincing, not least because other work in the field is demonstrating that psychological mechanisms having to do with group identity and cooperation, rather than ontological conviction, are responsible for the willingness of some individuals to die for social groups, including religious traditions. Further, while credences can be held with strong emotional fervor, there is nothing in the account that says they have to be. And in fact, in a great many social contexts (e.g., much of Japan), people do not hold their religious credences as central components of group identity.

In sum, the term belief should indeed be used with caution in ethnographic writing, as its Western connotations can mislead us. Irrespective of its ethnographic utility, however, the term belief has proven both rich and useful in better understanding the minds and actions of human beings.