Religion & Archaeology


Reference: Horrell, P. E. (2007) Religion & Archaeology

A short literature review of the current state of the 'study of religion' through the medium of ancient and prehistoric physical remains.


A problem statement

In his book Archaeology, Ritual, Religion (2004) Timothy Insoll excellently summarises archaeological failing to incorporate and analyse religion as:

Although potentially a vast topic, the archaeology of religion is in fact substantially neglected as regards the provision of a convenient and accessible introductory text, and this volume aims to redress this. Nevertheless, pretensions are not entertained here that what is provided is the definitive statement on the relationship between archaeology and religion, and that subsequently the archaeology of religion will be adequately theorised; it will not be. For in stating this it should be accordingly noted that archaeological approaches to religion have been remarkably naïve and it has frequently been thought of as a relatively simple area of investigation. It is not, as it is comprised of the residue associated almost wholly with people's beliefs, both individual and collective, and thus it is in fact remarkably complex (Insoll 2004:01).

Indeed, it is the lack of provision for religion in the basic texts for archaeology that archaeologists have no basis from which to investigate religion. For example, in Archaeology: Theories, Methods, and Practice (Renfrew and Bahn 1996) the archaeology of religion is only briefly covered in a small section that does not provide an adequate method for its study. In Kevin Greene’s Archaeology: An introduction (2002) religion may be considered to be included within a case study assessment (Greene 2002:53-9), but no methodology is given for the study of religion more generally through archaeology (further discussion of core archaeological texts that fail to engage religion, see Insoll 2002:1-5). When religion is more specifically discussed in archaeological works it has a tendency to be very general in content  (see for example Renfrew 1994) and focused on a singular (historic, that is, with supporting literature) religion, such as Christianity, Islam, or Buddhism (i.e. Frend 1996, Schopen 1997, Insoll 2001, 2004:1).

Archaeology in many respects has not adapted to contemporary approaches held by other disciplines, thus, its methodology for examining phenomena, such as religion, is out of date and needs urgent re-examination. E. O. James’ Prehistoric Religion: A Study in Prehistoric Archaeology (1957) is a good example of an archaeological review of evidence that may in some manner be considered religious, that is so archaeologically dominant in both style and content – that religion is not really examined per se. Mellaart (see Mellaart 1963, 1965, 1975) makes a lot of conclusions on what evidence may be religious, for example, his conclusions on what buildings are shrines in Çatal Hüyök. However, he does very little to explore and reconstruct the religion of Çatal Hüyök. Indeed, for the area that is so under defined and has no firm methodology religion, and more specifically ritual (ritualism, ritualistic) appears as a “favourite catch-all category for ‘odd’ or otherwise not understood behaviour” (Insoll 2002:01)

It is therefore clear from this brief introduction to the chapter that religion is an area that has been neglected by both archaeologists and archaeological theory. Furthermore, even the more contemporaneous cognitive archaeological works still fail to provide a definition for religion or a working methodology for its study. This chapter will now continue to demonstrate the definitions and approaches to religion utilised by archaeologists and attempt to provide context to where these ideas originate from. I will begin with a brief introduction to the etymology of the word religion, and will then continue with an exploration of the archaeological treatment of religion with specific examples from the archaeology of the Near East.

The Historical Approach

Origin of the word religion

To explore definitions and methods studying religion it is essential to briefly demonstrate where the word ‘religion’ originates. The origin of the English word Religion can be traced back first to the Middle English word religioun and ultimately to the Latin word ‘religio’, and this is believed to be an adaptation of the word religare or relegere (Webster’s Dictionary 2001). The actual meaning of religion has changed significantly in the two millennia that we can trace its existence, and if its meaning continues to change this will affect any study concerning religion. Even in the late Republic era of Roman history there were orations and debates upon the words used to define religion. Cicero argued that it originated from relegere, which is a verb meaning to re-read, and thus he thought this meant that religion was a tradition that was passed on from one generation to the next. Lactanius asserted that the word religio came from religare, which means to restrain, tie back or bind fast. Gavin Flood considers this to equate religion with the concept of a tradition that binds people to each other and to the gods of the Roman state (Flood 1999:44).

Early Christians used the word religio or religion to mean pagan religions and not Christianity. However, by the seventeenth century, it had come to mean communities of people who had devoted their lives to the service of God (Flood 1999:44). Our understanding of the word religion is detached from the Latin word ‘religio’ and instead has risen out of the Enlightenment period as an abstraction in understanding and criticising Christianity. That is, the Latin is inclusive of all forms of religion, pantheist and monotheist, whereas by the Enlightenment, the term religion had a more single-minded, monotheistic interpretation. The result of this is that certain understandings of the term religion are useless when not commenting on monotheistic religions, especially Christianity. In many non-industrialized communities, there is no word equivalent to the English ‘religion’. It is a Western concept, which is never completely satisfactory when applied to non-Western societies because of its Western origin. Anthropology has moved the primary study of religions away from the monotheistic, but this has been criticised as causing a comparative effect when studying non-Western religions, that is, comparing religions to Christianity. This can be observed by the concentration of many scholars on religious text when trying to understand other religions because the Bible is so central to Christianity.

There is also the argument that religion as a concept has no value beyond our western culture and is too incoherent to be applied to non-western or ancient (or prehistoric) religions:


Past (or non-western) forms of what we would identify as ‘religion’ operated in terms of mechanisms and self-understandings that we no longer associate with religion as such. … The concept as we understand it (and hence, tend to define it) is a by-product of the special historical and political circumstances of Western modernity (Asad 1993:28-48 and Arnal 2000:30-1).


What this means to anyone attempting to study prehistoric religion is that the values assigned to religion in modernity and post-modernity may not be the ones that were essential in the past. And, therefore, any definition derived from contemporary religions, will be inadequate in the study of ancient or prehistoric religions due to a misunderstanding of the focus of religion. Consequently, any definition derived for the study of prehistoric religions may differ from that used to study more contemporary religious expressions, such as New Religious Movements (NRM), because of the evidence used in studying the religious phenomenon.

It would be impossible, within the confines of this thesis, to explore all of the theories, ideas, and definitions that are available for studying religions. However, an attempt has been made to cover those definitions that are most influential to what has shaped our understanding of religion from an archaeological viewpoint.

In the modern era, drawing from roots in the Enlightenment, there is a tendency to ‘deconstruct’ any given highly complex process or phenomenon, in order to study it in the form of its basic component parts. Many scholars have deconstructed religion, claiming religion to be, for example, explainable as the belief in the supernatural, to be a purely social phenomenon, to be a by-product of economics, or a mental illness. In many ways, these component parts are viewed differently by different disciplines. In this section we will consider the definitions and methodological approaches offered by the disciplines of anthropology, sociology, psychology, and phenomenology, reviewing their relevance for the study of prehistoric and ancient religions. In gathering this information I make one major omission, that being, the theories held in the discipline of theology. This is because this discipline is only applicable from what is called an ‘insider’ perspective, that is, one who is a believer in the religion being studied. Clearly, this method is not practical when studying prehistoric, or any other long-dead, religions.


Religion as the belief in the supernatural

Many of the theories developed by early anthropologists characterised religion purely by reference to a realm of the supernatural, and considered humankind’s interaction with the unknown to form the basis of religion. Towards the end of the nineteenth century anthropologists such as Sir Edward Tylor and Sir James Frazer became interested in the beliefs and practices of these so-called primitive societies (Gellner 1999:10-11). These ideas began in the nineteenth century and were products of the European enlightenment, and they saw in themselves and their society a situation where reason, logic and science had triumphed over the ‘primitive’ mind. Despite the age and constant criticism to the supernatural definition of religion it is still an important concept that is still employed in archaeological works, for example, there is a section devoted ‘Identifying the Supernatural Powers’ in Renfrew and Bahn (1996:392).

The idea of religion as a belief in the supernatural has itself developed and has contemporary advocates within the SoR, for example, Bryan Wilson who defines religion as "any set of symbolic forms and acts that relate human beings to ultimate conditions of existence, cosmic questions, and universal concerns” (Wilson 1990:11). Wilson's theory holds that religion is a mode of interaction, whereby the relationship between humankind, its world and the supernatural universe (cosmos) can be explained. The 'symbolic forms' relate to the iconography of a given religion and the 'acts' that relate humans to the superhuman are the rituals of religion. Indeed, it is this idea symbolic representation of the supernatural that is advocated in Renfrew and Bahns (1996:392). This may explain why the only area of prehistoric religion to have been written about extensively is that of the Upper Palaeolithic where there is an abundance of symbolism in the form of cave art (see for example Maringer 1956 189-200, and Dickson 1996:123-158). From the archaeology of the Near east this concept that the images and idea of the supernatural was a basis for the understanding of religion can be observed in works such as Pollock’s Ancient Mesopotamia (1999:173-195) where the brief discussion of religion is done so under the chapter heading ‘Ideology and Images of Power’.

Wilson’s definition of religion is based upon the definition of the anthropologist Edward Tylor,[1] who defined religion as "belief in supernatural beings" (Tylor 1871:42) this is a much narrower definition, and as Michael York rightly points out it would exclude contemporary religions such as Buddhism (York 2001a: 2-3).[2] Thus, although the supernatural does play a large part in most religions, it is not a vital element of all religions. Therefore, it is important to consider both the iconography and their representation of the supernatural this is too simplistic a definition to use for the analysis of Prehistoric religion, especially as this relies upon the survival of art and symbols in the archaeological record.


Another important concept introduced by Tylor was the idea of the belief in the soul and animism. This idea is again one that is still very popular in archaeology and is frequently employed in the explanation of burial customs (see for example Maisels 1993:90). Tylor argued that the earliest and purest form of religion was what he termed as ‘animism’; this is the belief that all things are in some way in possession of a soul (Tylor 1866). Tylor thought that the belief in the ‘soul’ originated through combining observation of the living compared to the dead with dreams where one could reputedly meet the soul of a dead friend. The soul was seen as the essence of the individual that was confined within the body during life and yet existed independently. He considered that the next step was then to extend this theory to animals and inanimate objects, for there was no apparent reason why they could not also have an indwelling soul. The belief that the souls of the dead continued to exist after the death of the physical body led primitive humans, so Tylor thought, to believe in disembodied souls that were untethered by physical restraints. This translated into a belief in gods, goddesses, demons, and other spirits. Humans modeled these gods not only in their own images but also organised them as a reflection of their own society.[3] Pantheons of gods ruled by a king eventually led to monotheism as a supernatural autocracy (Tylor 1887a: 400-78).

Tylor’s idea can be considered to be a core element of archaeological understanding of religion, especially when combined with the ideas of James Frazer. Frazer defined religion as ‘The propitiation or conciliation of powers superior to man which are believed to direct and control the course of nature and of human life’ (Frazer 1932:222).  For Frazer religion is not only the belief in the supernatural but humankind's attempt at ‘propitiation or conciliation of these supernatural powers. This idea can be observed by the much-cited (both supported and critiqued in Pollock 1999:188-9) work by Thorkild Jacobson The Treasures of Darkness, who states:

Basic to all religion – and so also to ancient Mesopotamian religion – is, we believe, a unique experience of confrontation with power not of this world… It is the positive human response to this experience in thought (myth and theology) and action (cult and worship) that constitutes religion (Jacobson 1976:1).


Indeed, Jacobson’s book is based primarily on textual evidence that is projected back towards prehistory, nevertheless, it is one of the rare examples where a clear definition of religion is provided.

The idea of animism was refined by the later anthropologist R. Marett who did not believe that ‘animism’ was the first and purest form of religion; for he believed that he could identify an even earlier stage in the evolution of religion. He believed that the first step was a belief in what he called ‘mana[4]’, an impersonal force or power that was present in natural forces such as wind and thunder, as well as in certain people like the chief or wise woman. The word is Melanesian, and Marett’s theory developed from the study of the Melanesian culture. Mana was something to be both in awe of and coveted by those that were not in possession of it. All rituals were either attempts to ward off mana or gain it (Marett 1914:31). However, the concept of mana may be seen as a facet of animistic religions, in that it is part of the greater magico-religious universe of a culture/society. Marett’s theories have been employed by archaeologists studying Upper Palaeolithic religion, for example, the ‘Sorcerer’ (a humanoid with antlers and other body parts of animals) has been compared to 18th Century depictions of Siberian shamans (Bahn 1997165-8).


[1] Despite never actually reading for a degree at university, Tylor was given first a readership and then a chair in Anthropology at Oxford University.

[2] It may be argues that Buddhism is in fact a philosophy and not a religion at all, or that it is the exception that is inherent in all rules. The is considered more fully in the methodology chapter of this study where the areas of interaction for religion are examined in detail.

[3] Ideas similar to this begin with Greek philosophers in the 5th century BCE and resurfaced in the renaissance and the enlightenment periods and were expanded upon and developed. The first person who was to investigate the possibility of religion being a social phenomenon and social construct was Xenophanes (Thrower 1999:93). Xenophanes lived c. 570-475 BCE, and he asserted that it was men and women who made the gods in their own image, shaping, clothing and imagining them to speak like themselves. He believed that this was a logical reaction and that if “oxen and horses and lions could draw and paint, they would delineate the gods in their own image” (Kirk and Raven 1957:frag. 14-15).

[4] The force that is called mana has been identified in other cultures under different titles, for example, Waka among American Indians, Ngai among the African Masai, and Baracka among the Arabs.


Evolutionary development of religion

The idea of cultural evolution was, and is, an important concept in archaeology; and began with the introduction of Darwinian theory into the archaeological framework. There are many different classifications of evolutional development, and the specifics usually depend on the archaeological specialisation, however, these classifications are usually based on the technology level as at their most basic are broken down into Ages, such as Palaeolithic (old Stone Age), Mesolithic (Middle Stone Age), Neolithic (New Stone Age), Chalcholithic (Copper Age), etc. This idea was first developed by the pioneer archaeologist Christian Jürgensen Thomsen who introduced the three-Age system (Insoll 2004:44). Early anthropologists (especially Edward Tylor and James Frazer) argued that religion could be understood as an evolutionary step in the human psyche and that these steps could be tied into the technological development of cultures. Tylor asserted that all humans had the same mental capacity but the crucial difference was their place on the evolutionary ladder as he saw it. Following this line of reasoning Tylor’s argued that newer cultures were more advanced and had a fuller and more accurate understanding of how the world actually worked. The first of his assumptions was criticised almost immediately, especially by Lucien Levy-Bruhl, a French anthropologist. The second, however, was a lasting theory that only became generally disputed in the latter half of the twentieth century (Thrower 1999:127), nevertheless, it is still popular in certain scholarly niches.

The major flaw in Tylor’s second idea was that dark ages are common occurrences through history and are a time where the newer cultures become less advanced than the preceding culture. For example, although late medieval society was more technologically advanced than Ancient Greek or Ancient Egyptian societies, they did not always have a more accurate understanding of the nature of the Universe. Cultural evolution is now considered to be more like a nebulous bush with many branches that are able to co-exist but sometimes interact.

Tylor, like other post-enlightenment thinkers, believed that religion was an early way of rationalising both the natural and supernatural worlds. He believed that humans were now entering a new phase where rational science would dominate. His general views can be summed up in the quote, “Mechanical astronomy gradually superseded the animistic astronomy of the lower races” (Tylor 1887b: 229). The main criticisms of the theories proposed by Tylor were summarised by Sir Edward Evans-Pritchard, who asserted that Tylor’s theories could never be proved by hard evidence as they concerned the possible thoughts of people long dead (Evans-Pritchard 1965:25-7). Nevertheless, Tylor’s ideas are worth remembering when reading any older archaeological publication as facets of his theories continue in archaeological circles.

The most important work for the evolutionary theory of religion was James Frazer’s The Golden Bough: a study in magic and religion (1911-15), in which Frazer argues that religion is an intermediary stage developing from simple superstition and magic,[1] ultimately superseded by science. This is often called the rationalistic view of religion as it views religion as something empirical and as a development in the evolution of the human intellect.[2] However, this does not help to define what a religion is as it only identifies religion as belonging to a lower level of cognition in the human psyche. What it does, though, is raise the important question of when do we see the change from superstition and magic, which can be said not to be religion, to what we can call religion (if any such step exists)? From Frazer’s point of view, magic is the first stage and religion is the second in the chain of development of the human intellect. This understanding considers religion as the perceptions of the humans who adhere to it and how they interpret these perceptions. Science is then considered the true perception of reality, seeing things as they truly are, whereas religion is at best an imperfect and faulty perception of reality. It is flawed, as beliefs are generated to explain what cannot be explained within the context of the social and intellectual development (society) that creates it (Frazer 1911-15). This view of religion can be considered flawed and biased, as it was formulated in a time where there was a great shift in Western society towards science from the prevailing monotheistic religions.

Frazer was critical of archaeology and considered it to be a lesser area of the study compared to anthropology and ethnography (Ucko 2001:273, Insoll 2004:44). However, Frazer’s work has been very influential on archaeological works and his ideas on the scientific nature of magical practices and their use of logical observations are still useful. Although Frazer believed that magic evolved into religion, which itself evolved into a science, he also argued that magic was more directly related to science than it was to religion. This was because magic did not involve worship but the manipulation of impersonal, un-animated but occult forces. “Religion … stands in fundamental antagonism to magic as well as to science, which holds that the course of nature is determined, not by passion or caprice or personal beings, but by the operation of immutable laws acting mechanically” (Frazer 1922:51). Frazer argued that magic was essentially sympathetic and that it obeyed what has become known as the ‘law of sympathy’. This simply states that any action, object or person that looks like something is in some spiritual way connected to that thing. Likewise, any piece of a thing, no matter how small, is connected to the thing that it once was part of and remembers being what it once was. A good example for the rule is the so-called voodoo dolls, which are made to look like the intended victim and some part of the victim, such as a strand of hair is usually used in their construction. This ‘like for like’ philosophy can often be abstract (such as the copulation of a couple in order to make land fertile) or obvious (such as, removing a brick from a wall and destroying it, in order to bring down the wall). Frazer believed that this made magic more like a science than a religion, as it had fixed rules and laws that had to be observed. Much of this was related to the principle of ‘cause and effect’. The main criticism of this theory was that magic was often intermixed with religion and religious beliefs and that they could not be viewed as wholly independent of each other. Eric Sharpe identified this point and argued that what we were observing was a ‘magico-religious’ phenomenon that was neither magic or religion in their own right, as in many circumstances they were inseparable (Sharpe 1975:72).[3]

Andrew Lang criticised Frazer’s evolutionary model that argued polytheism was replaced by more advanced monotheism. Lang pointed out that amongst what are considered to be the most primitive and ancient societies on the planet, namely the aboriginal Australian, there is belief in a supreme and high god (Lang 1908:111). Wilhelm Schmidt expanded on this theory by arguing for what he called the ‘Aboriginal revelation’ as a true form of religion that became degenerated and warped over time. Most people have disregarded this theory (Schmidt 1931:Vol I). Rudolf Otto also criticised Frazer’s approach. Otto believed that religion began as human observation of the character of being and the universe, rather than as an attempt to rationalise and understand it (Verkamp 1995:52).


Figure I: Evolutionary model of Frazer's 'Primitive Mind'



The evolutionary model (evolutionism), whereby the primitive savage would evolve into civilised man,[4] was severely shaken with the atrocities of the Great War. The model that was put forward by Christian anthropologists was that monotheism was the final point of the evolution of religion. A progression of this type was suggested as early as the Nineteenth Century by the Count de Saint-Simon (Capps 1995:61-2) and this was in contrast to the evolutionary model proposed by Tylor and Frazer (Verkamp 1995:73-90). However, it is worth mentioning, although in brief, as there are areas within archaeology where this ideas is still persistent, such as Biblical Archaeology.


Figure II: Evolutionary model of theism




The current approach today is not to look at any one culture as more advanced than another in its forms of religiosity, but religion is to be evaluated in its own context.[5] Max Weber advocated this view, but his works were not seen as relevant by early anthropologists, and therefore were overlooked (Gellner 1999:16) and this may also be said for early archaeology. Indeed, this is a very inherent issue in the discipline of archaeology which is constantly comparing cultures (and, in fact grading them) in terms of their technological development. To then consider religion as something outside of this evolutionary, or ‘progressional ladder’ based, approach.

Counter arguments for the evolutionary model can be located in the works of Malinowski and Radcliffe-Brown, and these are generally considered to be functionalistic. Malinowski argued that any given society had to be studied as a whole, that the practices, beliefs, and customs had to be understood within their context and explained in terms of their function within the society that generated them (Gellner 1999:17). Scholarship is thereby forced to operate within its own provisional time frame and must recognise that the views of the society in which it is conducted will affect its outlook on all other forms of civilisation. This in general was easier for archaeologists to incorperate as it advocates the idea of keeping evidence within its own context.

Malinowski offered a functionalist explanation of both religion and science, based upon his ideas of human needs. He argued that religion gave psychological support when facing the unknown and magic assurances when facing disconcerting events or conditions (Gellner 1999:17). He listed seven needs, or basic requirements, which needed to be fulfilled by the society or culture: reproduction, metabolism (staple requirements, such as, air, food and water), safety, bodily comfort, movement, growth, and health (Malinowski 1944:91). Radcliffe-Brown was a contemporary of Malinowski and held similar views that were non-evolutionary. However, rather than focusing on the physical needs of an individual within society, Radcliffe-Brown concentrated on the needs of society as a whole. His paradigm was called ‘structural functionalism’ to differentiate it from Malinowski’s ‘functionalism’ (Gellner 1999:18-9).


[1] The first-ever university chair in anthropology was created for Sir James Frazer at Liverpool University in 1907 (Ackerman 1987:207). Frazer’s work in anthropology, The Golden Bough was constantly revised in numerous editions. The task that Frazer set himself in the Golden Bough was to explain the meaning and purpose of the ancient and ‘barbaric’ ritual of succession of the priest of Diana at Nemi. This problem caused him to digress and consider other rituals as well. What originally started out as only two volumes, in 1890, were twelve volumes by 1915. Frazer felt that to explain this one ritual he needed to explain how the primitive mind itself worked.

[2] That is, something that can be studied using methods from empirical science, such as biology, and be understood as a logical step towards rational thought.

[3] It may be a valid argument that religion and magic are totally inseparable, that some form of belief or faith that is religiously sanctioned is required for magic to exist. Nevertheless, this is not an issue central to this thesis.

[4] Man used as a contextual reference rather than Human.

[5] This may have be related to the Post-modern belief that we are not progressing along a path to perfection always walking upward, consequently, the idea as evolution as a nebulous branching bush.

Section Title

Section Subtitle

Every website has a story, and your visitors want to hear yours. This space is a great opportunity to give a full background on who you are, what your team does and what your site has to offer. Double click on the text box to start editing your content and make sure to add all the relevant details you want site visitors to know.

If you’re a business, talk about how you started and share your professional journey. Explain your core values, your commitment to customers and how you stand out from the crowd. Add a photo, gallery or video for even more engagement.

Section Title

Section Subtitle

Every website has a story, and your visitors want to hear yours. This space is a great opportunity to give a full background on who you are, what your team does and what your site has to offer. Double click on the text box to start editing your content and make sure to add all the relevant details you want site visitors to know.

If you’re a business, talk about how you started and share your professional journey. Explain your core values, your commitment to customers and how you stand out from the crowd. Add a photo, gallery or video for even more engagement.