Monday, June 12, 2017


Carlin A. Barton, Daniel Boyarin, Imagine No Religion. How Modern Abstractions Hide Ancient Realities. New York: Fordham University Press, 2016. Pp. 325. ISBN 9780823271207. $35.00 (pb).

Reviewed by Anders Klostergaard Petersen, University of Aarhus (

Version at BMCR home site


Barton's and Boyarin's monograph is related to Brent Nongbri's Before Religion: A History of a Modern Concept (New Haven, Yale UP 2013), a volume that has justifiably attracted considerable attention in disciplines devoted to the study of the ancient Graeco-Roman world. Barton and Boyarin refer in important places to Nongbri who has also provided them with their introductory, orally-transmitted, Edwin Judge contention that one should avoid the term religion in translations of ancient texts. Barton and Boyarin follow this injunction with relentless effort, arguing that any rendering of Greek and Latin terms (notably thrēskeia and religio, but also the related terms deisidaimonia and superstitio) by the word 'religion' is an anachronistic distortion. The argument is cogently pursued in Tertullian and Josephus (and some additional authors) held emblematically to illustrate the problem at stake.

Although it is obvious from syntax, style and expertise which parts of the book originate with which author, Barton and Boyarin have chosen to voice their argument by means of a joint 'I'. The book is elegantly written with admirably few typos. Apart from a brief but important introduction to the modern concept of religion (1-9), the book is divided into two parallel structured sections, each comprising two parts: a mapping of the pertinent word and a subsequent thick description case study. Section one is focused on religio and provides an in-depth study of Tertullian's use of the word. The second section concentrates on thrēskeia, which subsequently is dealt with in a comprehensive study of its use in Josephus. Notes are presented as end notes comprising almost 75 pages. The book contains a useful index of ancient texts and a fairly comprehensive general index.

There is much to praise. The first part dealing with religio makes it patently clear that the term does not reflect the modern third-order concept religion, but was 'especially evoked by highly charged boundaries and limitations and the fear of transgressing "taboos," such as those surrounding the oath, the treaty, the domus or fanum. The Romans approached such boundaries as we might approach an electric fence. But it is important to notice that the gods were not inhibiting or disciplining forces in any of these instances of religio' (22). With Cicero, however, the term is brought to bear on matters pertaining to gods (cf. De natura deorum II iii: religione id est cultu deorum multo superiores, and De inventione II liii 156). So how may we conceive of the Ciceronian turn as the authors dub it?

Barton and Boyarin argue that Cicero's philosophical oeuvre in general and De natura deorum in particular paved the way for what eventually gave rise to the distinct Christian coinage of the term religio among Christ-adhering authors like Tertullian, Minucius Felix, Lactantius and Ambrose (46). In his philosophical work, religio came to designate a realm associated with the sacred inasmuch as Cicero asserted that, by the right of a limited group within the Republic to define ʽthe sacred', religio came to be made into a particular and separate sphere of life (51). With Cicero laying the foundation for the subsequent use of religio among Christ-adhering authors, Barton and Boyarin proceed to Tertullian. They extract two trajectories of key importance. First, Tertullian in those works characterised by an apologetic or negotiatory mood applies religio to designate a "distinctly nebulous" realm coordinate to or subordinate to the Roman imperial government. This constitutes a scheme by which Tertullian rhetorically pursues the reconciliatory strategy of serving two masters. Second, there are works defined by an unremitting segregationist or separatist mood in which Tertullian zealously insists on serving one master only, that is, the divine king at the cost of the Roman emperor (58). In his works there is a constant wavering between these two paths with the former being particularly dominant in his apologetic oeuvre. The authors accentuate the distinct nature of Tertullian's uncompromising stance by comparing him to the IRA prisoners in the Long Kesh prison "who in their Blanket Protest refused all clothing and comforts" by pursuing self-sacralisation as a form of via negativa (70). This is excellently seen and one of many brilliant comparisons made to later historical figures and events.

With regard to Tertullian's utopian separatist stance, the authors note how there could be "… no distinct cultic sphere, no possible separation of anything like our "religion" and "politics" and "economics" any more than there could be for the Amish, the Tibetans, the Navaho, the authors of the Mishna, the umma of Muhammad, or the inhabitants of More's Utopia" (100f.). If we grant this, how then may Tertullian's use of religio be seen as a precursor of the modern usage of religion?

This is most obviously grasped in the context of Tertullian's negotiatory mood, because it was "while pursuing this trajectory of thought that Tertullian whittled Latin religio (under the influence of Cicero's definition of religio as the cult of the gods) to fit one of the niches occupied by thrēskeia as it appeared in the Greek Jewish and Christian apologetic writers" (111). With reference to Apol XXIV 9 (sed nos soli arcemur a religionis proprietate — "But we alone are warded off from a distinct religio") they argue that here religio begins to take on the contours of constituting a distinct, separate, and privileged dimension of life "that could exist within and still be distinguished from the government of the Empire" (ibid.). 1

At this point we enter the Greek world of thrēskeia, which is understood as inhibitions or prohibitions claimed to evolve into designating excessive fearfulness and thence to denote frenzy. Eusebeia conversely comes to designate appropriate and measured strictness in observance (131). At the same time, however, thrēskeia also retains its basic meaning of practice, that is, "our practices" contrary to those of others. In the Judean-Greek context thrēskeia is used with both meanings. Frequently it denotes the superstitio of the others, thus constituting the rhetorically created castigation of the cult of the others, while eusebeia or therapeia contrariwise are never used in this way. Scrutinising the use of thrēskeia in Philo and 4 Macc., the authors demonstrate how the two uses also appear in these texts.

From the terminological mapping we enter the Josephan territory. Here the overall argument of the book is brought to a head. It is not only possible but also distinctly advantageous to describe Josephus' world without recourse to the concept religion: Imagine no Religion (155, cf. 178). Interestingly we are told how Josephus by occasionally understanding thrēskeia as synonymous with eusebeia entered into an indirect polemic with the different understanding of the term in Philo and 4 Macc. (169ff.). In an ingenious but perhaps too subtle reading of Josephus, 2 the authors strive to demonstrate how Josephus deliberately and supremely oscillated between different and mutually incongruent positions: Josephan doublespeak (178-99, see 189-91 in particular). He is shown to waver elegantly in using a dual sense of deisidaimonia by occasionally ascribing it positive connotations while generally retaining it with its traditional negative undertones.

Rounding up this section of the book, the authors cast light on the development subsequent to Josephus in the apologetic literature of second to fourth century Christ-religion. Similar to the apologetic context of Philo's use of thrēskeia, some Christ-adhering apologists could also use the word in a way anticipating the future ascription of meaning to the term religion as denoting a separate and autonomous sphere of life (209).

The conclusion poignantly and succinctly repeats the main argument. To translate thrēskeia or religio by religion obscures ancient matters. Both religio and thrēskeia are marked by the risk of tipping over the edge and evolving into something problematic: Too much religio leads to superstitio. Thrēskeia and deisidaimonia, conversely, often mark inordinate fear of the gods or illegitimate frenzy (212).

I sympathise with the overall argument and the injunction to abandon the term religion in translations of (any) ancient texts. That said, however, I also have some severe queries with the argument. To be crude, one could argue that the authors are carrying coal to Newcastle. For a scholar in the study of religion it is an old truth that there can be no term 'religion' in the pre-modern world. This is the basic argument of Max Weber in his seminal Zwischenbetrachtung, and one may add the central contention of Durkheim as well.3 A century ago Weber emphasised how the invention of 'religion' presupposed the detraction of the phenomenon from the wider cultural sphere—something dated by Weber to modernity. One could draw a distinction in the ancient world between the sacred and the profane, the latter designating diminishing degrees of sacredness (pro-fanum) but never something categorically secular. I find it striking that this central contention of emerging sociology a century ago has neither been taken into consideration by Barton and Boyarin nor by Nongbri. Ultimately, I claim that their argument affirms Weber's and Durkheim's view, but from the perspective of early sociology it is pouring new wine into old wineskins.

The second and more harsh as well as fundamental point of criticism relates to the book title and the relentless effort to discard the term religion when studying the ancient world. In their introduction, Barton and Boyarin, with reference to J. Z. Smith and Timothy Fitzgerald (others could have been mentioned), emphasize how a third-order concept of religion is a modern Western construction and, therefore, illegitimate with respect to studying non-Western religions. As regards the first part of the argument I would immediately say: Obvious. As concerns the inference made from it, I could not be more in disagreement. I find this conclusion philosophically naïve and in terms of philosophy of science and the overall argument brought forward self-contradictory.

First, obviously there was no concept of male chauvinism in the ancient world, but does that imply the non-existence of the phenomenon? Definitely not. Second, how could one possibly tell that there was no concept of religion in the ancient world if one does not subscribe to a concept and an undergirding definition of the phenomenon by which one can make differentiations between ancient and modern usages of terms (cf. 4: "When we make the claim that there is "no religion" as we know it in the cultures we are describing…")? Third, given the previous point, it is conspicuous that the authors do not present a third-order concept and a concomitant definition of religion—this is notably different from Nongbri. Frankly I do not see how we could or should avoid it in order to know what we are talking about. And if one defines religion as, for instance, "semantic and cognitive networks comprising ideas, behaviours and institutions in relation to counter-intuitive superhuman agents, objects and posits,"4 I do not understand how this may be conceived as an imposition on ancient religion in our understanding of it, since applied in the context of pre-modern religion it only involves acknowledging the conflation between culture and religion.

These critical points aside, I want to close the review in a positive tenor, since the laudable nature of the book is indubitable. It is an elegantly written and thoughtful work which once and for all should prevent translators of ancient texts from using the term religion. Additionally, future scholarship on Tertullian and Josephus cannot ignore this genuinely in-depth examination of these two authorships.


1.   Unfortunately, the rendering of this passage is erroneous by the authors' double repetition of sed non soli…, which is not in the Latin text.
2.   I am not contesting Josephan doublespeak, but I question the degree of deliberateness assigned to it by Barton and Boyarin.
3.   Max Weber, "Zwischenbetrachtung: Theorie der Stufen und Richtungen religiöser Weltablehnung," in idem, Gesammelte Aufsätze zur Religionssoziologie I, Tübingen, Mohr-Siebeck 1963, 536-73.
4.   Jeppe Sinding Jensen, What Is Religion?, London, Acumen 2014, 8.


  1. Having not had a opportunity to read the book that you have reviewed; nevertheless, I have a question that may or may not have been covered in the book.
    It seems to me that the ancient Graeco-Roman world was steeped in "religio." There were gods or goddesses for the home, city, nation, ethnic group, guild, etc. One could go to any city or country without running into "religio." The fact that Barton and Boyarin want to delete "religio" from the vocabulary of θρεσκεια, ευσεβεια, and δεισιδαιμονια, let alone of superstitio and religio smacks of historical revisionism.
    The post-modern world with its phobia of any thing religious forgets that religion, i.e. theology, is behind every thing that we think, say or practice.
    Finally, if Barton and Boyarin are trying to bring attention that "religion" is being used too often when translating certain texts, then they might have a point, but ALL the texts. I don't think so.
    BTW, I have had seven years of formal Greek and five years of formal Hebrew language instruction.

  2. How about reading the book and its arguments before dismissing them?

  3. tantum religio potuit suadere malorum