Is there Religion in Japan? (Round 2)

The current debate prompts existential angst for Amaterasu, the Shinto Goddess of the sun.

The ‘religion’ in Japan debate prompts existential angst for Amaterasu, the Shinto Goddess.

A number of years ago I wrote a blog post about a lively debate between Timothy Fitzgerald and Ian Reader concerning whether it was appropriate to speak of ‘religion’ in Japan and whether the concept had any coherent significance prior to the arrival of the Western colonial powers and their ideological baggage. From my perspective a clear winner emerged from these exchanges (*spoiler* it did- see my previous post for details) but I’ve just become aware that, while working on my PhD, I seem to have missed a more competitive second round that has been taking place over the past few years, due in large part to the work of Jason Ānanda Josephson.

Round 1

But before getting into the new material, a quick recap of the earlier bout is in order: In cultural and social anthropology there has been a widespread ‘postmodern’ trend since the 1970s, the so-called ‘Cultural Turn’, which is associated with attempts to ‘deconstruct’ the unidentified assumptions behind existing academic theories and categories. Deconstruction is often presented as if it is something incredibly complex and theoretically deep but in my experience it often means criticising theoretical concepts, in an unnecessarily verbose manner, while emphasising the (often unacknowledged) subjective aspects of research/writing. It is within this context that Fitzgerald’s earlier attempts, to ‘deconstruct’ the concept of ‘religion’ and its appropriateness for Japan, should be understood.

Primarily, Fitzgerald charged that the term ‘religion’ (which today is usually translated in Japanese as shūkyō 宗教) made no sense when applied to Japan, as the distinction between ‘religion’ and ‘non-religion’ did not actually exist throughout Japanese history and could only really be traced to the modern era and the various impositions made by the Western powers during the forced ‘opening of Japan’ in 1853. He further contended that this artificial and coerced category had been reinforced not only due to the actions of various Japanese and Western policy makers, who had employed the distinction, but also by later generations of Western & Japanese academics, who continued to employ the term uncritically, as if it referred to a natural kind. Contrary to this view, Fitzgerald advocated that the history of the term ‘religion’ meant that it carried with it unavoidable Christian connotations, implying things like the existence of centralised doctrines, a professional priestly hierarchy, the exclusivity of ‘religious’ traditions and a combative approach with unsanctioned ‘superstitions’, all of which, he felt, made it an unsuitable term to apply to non-Western, non-Christian cultures, like Japan.

Ian Reader countered that the historical usage of the term was not equivalent to how it was employed in modern scholarship and that distinctions between ‘religion’ and other types of teachings had long been a feature of Japanese philosophical traditions, well before any interaction with the West. He emphasised, for instance, that even back in the Heian court in the early 9th Century there were clear distinctions and debates concerning the concepts of the Imperial law (ōbō 王法), the Buddhist law (buppō 仏法), and where the relevant boundaries lay. He also maintained that, contrary to Fitzgerald’s claims, he did not regard ‘religion’ as some static, universal ‘natural kind’ category but rather as a useful conceptual frame for making cross cultural comparisons concerning a varied and changeable feature of human societies. From my perspective (and that of most other later commentators I came across), Fitzgerald’s arguments were heartily dismembered by Reader’s critiques, which, aside from pointing out logical problems, also criticised Fitzgerald’s language proficiency, knowledge of Japanese history and disparaged his arguments for prioritising rhetoric and ideology over thorough scholarship. Regardless of who ‘won’ the exchanges, the discussions involved are well worth reading in full, if you want more details, or if you just want to see how cantankerous academic disputes can become (for a longer summary please see my original blog post or this detailed rebuttal from Reader).

Round 2

Fast forward to 2011, and following up on his earlier article on the Meiji educational reformer, Inoue Enryō (aka Dr. Monster), up pops an article by Jason Ānanda Josephson in Religion Compass which, at first glance, seems to be simply a rehashing of the same old arguments raised by Fitzgerald. However, first impressions can be deceiving and, indeed, this would be an unfair characterisation of the article. For one, the linguistic and cultural expertise of Josephson is certainly not in question but, more crucially, the piece also clearly demonstrates that he has a) engaged in serious historical archival research to develop his arguments and b) has a very strong grasp of Japanese history (or at least of the Meiji period and its various complexities). Josephson’s arguments hence can not be dismissed as some mere repetition of the assertions of Fitzgerald, and, indeed, despite acknowledging the inspiration of Fitzgerald’s arguments he also explicitly acknowledges the previous “fair criticisms of [his] lack of [Japanese] sources” (2011: 592).

Josephson’s article begins by highlighting the old problems that beset scholarship which portrayed Japan as having three separate and distinct religions- Buddhism, Shintoism and Confucianism- and introducing some of the new problems he sees with those positing that these diverse traditions actually represent a single ‘unified’ or ‘common’ Japanese religion (as suggested by Reader & Tanabe, 1998) . After noting these issues need to be resolved; “Was there a native Japanese concept of religion or was its appearance an imposition? Are there clearly defined separate Japanese religions? If so, what is their proper taxonomy?” he dramatically declares that “[a]t stake in these issues is not only the discipline of scholarship focused on Japanese religions, but more fundamentally the universality of the term ‘religion’” (2011: 590). While there is definitely hyperbole at play, it is certainly true that the issue of whether religion is a useful category to apply cross culturally is a persistent theoretical zombie which re-emerges frequently (and with grim predictability) not only in the context of scholarship on Japan, but also, at recent conferences on the cognitive science of religion.  Josephson provides an excellent historical summary of the (academic) debate about whether ‘religion’ is a relevant category for Japan and, despite the intrusion of some inevitable postmodern jargon (e.g. intersecting genealogies), his account is both fair and extremely readable. His summary is also admirable in that it is bookended with an explicit staking out of his position, which, it just so happens, is in direct opposition to that advocated by researchers such as Ian Reader & Michael Pye:

If Reader’s thesis were correct, it would represent a radical reversal in the main argument in religious studies, because it would show that ‘religion’ was not fundamentally a Euro-American Christian category, but instead a natural kind. Nonetheless, I believe that Reader is fundamentally wrong, as a quick look at Japanese sources from the mid-19th century will show (2011: 592).

He then goes on to detail the initial difficulties that Japanese translators had with deciding on an appropriate term to translate the English term ‘religion’, which featured prominently in the various treaties signed with America, and highlights the subsequent debates in which “intellectuals and policymakers proposed over half a dozen possible translations for ‘religion’… [and also] debated which indigenous traditions and practices fit into the category” (2011: 593). Josephson’s account admirably presents the Japanese involved in these debates not merely as passive vessels, acceding to all of the demands of the West, but rather as a heterogenous group of agents who eventually managed to translate pressure from Western Christians “into a concept of religion that carved out a private space for belief in Christianity and certain forms of Buddhism, but also embedded Shinto in the very structure of the state and exiled various ‘superstitions’ beyond the sphere of tolerance” (2011: 594). Here, Josephson references that while Buddhism did receive official recognition as a religion (becoming constrained in the process), Confucianism did not and instead became a “scholastic concept”, and Shinto became bifurcated into a non-religious State/National ideology (i.e. ‘State Shinto’ or what Josephson terms the Shinto Secular) and a group of smaller Shinto ‘religious’ sects (including new religious groups, some of which resisted the label).

Based on the initial lack of consensus of appropriate translations for the term ‘religion’ and the clearly politically motivated decisions in the defining process, Josephson concludes that it is a foolish to talk of continuity with  pre-existing native terms or categories of ‘religion’ in Japan, as this period was clearly marked by “radical changes on many levels” (2011: 594). Instead, he advocates that scholars working on Japanese traditions (and presumably all other non-Western traditions) should seek to escape “the confines of the Western category of religion” and instead employ “indigenous categories”, while reflecting on the radical “shifts that occurred in the early Meiji invention of Japanese ‘religions’” (2011: 594). There is much to admire in Josephson’s arguments and these are fleshed out in much greater depth in a (well received) book he released the following year, The Invention of Japanese Religions (2012). His core point, however, remains largely unchanged from the article and, while I would strongly recommend his book to anthropologists and those interested in Japanese history, for those without such a deep interest, the article discussed above offers a nice executive summary of his main thesis.

Academic Debates – Is Religion like cricket or politics?

Despite agreeing with most of what Josephson says concerning the need for a better appreciation of how concepts of ‘religion’ morph throughout history and are almost always firmly interconnected with politics, I still fundamentally disagree with his main conclusion that it is misleading to discuss ‘religion’ in Japan prior to the Meiji. This is true, if you define ‘religion’ as being about Christian like, centralised, exclusive, government recognised institutions with designated doctrines which elevate beliefs above practices, but what isn’t clear from either Josephson or Fitzgerald’s arguments is why modern scholars have to employ such a restrictive definition. The fact that the concept of religion has involved such Judeo-Christian biases in the past does not mean that modern usage, especially the modern academic usage, must replicate such mistakes. Ian Reader makes this point in his earlier exchange with Fitzgerald and again in his recent (positive) review of Josephson’s book (2013:306,312):

There are not that many people in religious studies nowadays who would characterize what they study as stable and unchanging over time, or who see the field as intellectually unchanging. While it is widely acknowledged that academic studies of religion were founded in nineteenth-century, Euro-American Protestant concepts and models associated with piety, doctrine, and belief, things have moved on greatly. While Josephson criticizes the field for holding on to an unchanging universal, reified category of religion, he himself appears to overlook changes… The concept of “religion,” as discussed and studied by Japanese academics, has more recently broadened far beyond the narrow parameters analyzed by Josephson to incorporate the various “superstitions” that Meiji lawmakers sought to eradicate.

As Reader points out, much ink has already been spilt identifying and responding to such biases, and proposing more inclusive definitions for ‘religion’. Moreover, Harvey Whitehouse (my supervisor) and many others, have argued that it is possible to both acknowledge the artificiality of any unitary concept of ‘religion’ and still seek to explore the diverse phenomena labelled as religion by “fractionat[ing] religion into numerous different traits, each of which must be explained on it own account” (2008: 35).

Another argument relevant to these debates appeared in a recent article by the cognitive anthropologist Pascal Boyer (who has written several books ‘explaining religion‘) who noted that “some anthropologists are tempted to think that people’s categories more or less define their world, so that people who have no concept of x have no x” (2013: 169). Boyer contends that this can be misleading, as while “in some cases having a concept is necessary to create a reality. People who have no concept of “cricket”… certainly have no games of cricket” it is also the case that “whether people have a notion of demography or economy or not, they all have demography and they all engage in economic transactions” (2013: 169). Ironically, Boyer elsewhere argued against the general ‘usefulness’ of the religion category, for reasons similar to Josephson, and in the same article also emphasises that “the notion of “religion” as a special domain is ideological. It is the creation of the large, corporation-like established religious guilds” and thus “is an impediment in more serious discussions of the social dynamics or cognitive processes involved” (2013: 171). Again, I disagree with this conclusion, while being in full agreement with almost all of Boyer’s analysis…

To explain why, let me take as a point of comparison the concept of ‘politics’, it is certainly the case that this is not a category of human activity that has always been distinguished as distinct from other spheres, such as religion, warfare, hunting, family life, etc. in all human societies (and even in the modern era I doubt many would claim that politics does not overlap with many other supposedly non-political aspects of life). It is also the case that what we would refer to as ‘politics’ looks very different at different times and in different locations and there may well be cases where applying the term would obscure more than it reveals. However, this does not make the term semantically incoherent, nor make it useless to apply to societies that lack the concept ‘politics’. Indeed, there is both a clear family resemblance and a line of continuity stretching between the recruitment of supporters and formation of alliances in early human hunter gatherers and the politicians and leaders of the modern era, still debating and scheming in grand international alliances. The point here is that, despite the widespread variation, the term remains a useful analytical frame when discussing human activity cross culturally, and I would contend that ‘religion’ serves the same purpose. If you limit religion to referring to formal institutionalised religion, then you will indeed only find it in certain eras and places and it has a distinct monotheistic intellectual history, but if you broaden the definition to refer to, for instance, thoughts and behaviours concerning invisible/superhuman agents, it becomes difficult to locate any society for which we have records that ‘religion’ does not exist.

Another line of argument that somewhat undercuts Josephson’s argument about religion being ‘invented’ in Japan during the late Tokugawa/early Meiji period is that the Japanese authorities had, many centuries previously, already encountered a developed religious tradition that included many institutionalised and formalised aspects, such as detailed cosmologies, complex doctrines, a professional priesthood and competing sects with various canonical texts and which received various levels of political support. And here, of course, I am referring to Buddhism, imported to Japan over centuries from kingdoms located in present day China and Korea. Native Shinto, or if you prefer ‘folk’, traditions and practices were fundamentally transformed by their interaction with Buddhism and there were evidently political machinations involved in the recognition and promotion of various competing sects from both Buddhist and native Japanese streams. Josephson does address the existence of these kinds of pre-Modern distinctions, devoting two chapters to discuss how Christianity was interpreted as a heretical Buddhist sect by the Tokugawa authorities, but rather than counting this as evidence of pre-existing religious traditions he instead interprets it as reinforcing the distinctive nature of the developments in the late Tokugawa/early Meiji period. He suggests that the Japanese response in this earlier period is actually indicative of the fact that they lacked a universal category of religion and consequently chose to appropriate/subsume Christianity within the existing Buddhist framework.

The Academic Zombie of ‘Religion’ Debates

I agree with Josephson’s assessment that the nature of the response to Christianity differed dramatically between these periods  and that in the latter period new official definitions of ‘religion’ were formulated (although see this article by Hans Martin Krämer for a different view on the motivations for the change) but I disagree with Josephson’s rather extreme conclusion that “there were no Japanese religions before the mid-nineteenth century” (2012: 256). As Michael Pye responded in his review, while Josephson “relentlessly probes the artificiality of every constructed concept… one also begins to wonder whether… ordinary people ever went to visit shrines and temples or carried out ritual activities of any kind, as they had done in the Edo Period and still do in their millions”. To see just how much religious activity (and debate) there was in pre-Modern Japan a good start would be two recent (thorough) reviews of Japanese language works on religions in pre-Modern Japan produced by Brain O. Ruppert available here and here. Although these reviews also make it clear that there is a lot of debate in such research relating to constructed histories, definitions and boundaries, it would seem entirely unhelpful to try and discuss these topics without invoking the category ‘religion’ and its Japanese equivalents. 

To summarise then, I remain unconvinced by the arguments that there were no ‘religions’ in Japan before interactions with Western nations in the late Tokugawa and early Meiji period. I do not however dispute, in any sense, that the definition of what officially constituted a religion changed dramatically and that there was a significant amount of politics and ideology involved in the changes and the designating of translation terms. Josephson’s research, is a boon to the field and, at least from my perspective, he seems to be engaged in a good kind of deconstruction, in that he does propose alternatives to what he attacks and his arguments are firmly wedded to in depth historical analysis and not just theoretical rhetoric. I would recommend his book and articles to anyone interested in Japanese history, or religion more generally, but I would also caution against accepting some of his more extreme conclusions. More generally though I sincerely doubt that this issue will ever truly be resolved, debating definitions is the bread and butter of academia and popular analytical terms are inevitably those most prone to deconstruction. So no matter how many shotgun blasts either side fires into the shambling horror that is the academic definitions of religion it seems quite clear to me that this beast will continue to get up and terrorise each new generation of scholars. But as long as each time we get new well-researched work, like The Invention of Religion in Japan, it seems that the endless battle might just produce something worthwhile.

Original Article: Josephson, J. Ā. (2011), The Invention of Japanese Religions. Religion Compass, 5: 589–597. doi: 10.1111/j.1749-8171.2011.00307.
Original Book: Josephson, J. Ā. (2012), The Invention of Japanese Religions. University of Chicago Press. ISBN-13: 978-0226412344.

 Postscript: Concerning the Increased Accessibility of Research(ers) 

The interesting thing about research in the modern era is that debates and discussions are no longer restricted solely to academic journals and books, as such, after coming across Josephson’s article and book, I was quickly able to locate a large number of responses/reviews, along with the usual academic reviews. Some might argue that looking at such extra information is distracting from focusing on the specific content and arguments contained in the original work, but personally, I always like to know more about the author’s I’m reading and other’s perspectives on their research. Indeed, a ‘pre-modern’ habit I’ve long held, is to always read any preface/acknowledgment section, especially in an academic text, as I’ve found these often ignored sections can reveal quite a lot about an author’s theoretical commitments & ‘lineage’, sources of information and potential influence from funding sources.

Anyway, from my limited research some notable finds included a detailed (and freely accessible) review by Jolyon Thomas, responses from both Ian Reader (in Monumenta Nipponica- a traditional academic journal) and Michael Pye (freely available on, a few in-depth interviews with Josephson (including this one in which Josephson summarises the main points of his book, his theoretical interests and even the relevance of his family background), Josephson’s personal blog (mainly covering preparations for his next book) and a short discussion of his argument on the Genealogy of Religion (a blog that I often frequent).

On the one hand, all of this is great because even just 10 years ago many of these resources would have been unavailable, or at least much harder to track down, and this new ease of access means that, if you make a little effort, you can easily dig a fair bit deeper and find out more about a specific scholar’s research, wider body of work and theoretical perspective. But on the other hand, if you think about all this increased visibility and access too much, I think it also risks making commentators, especially early career academics, wary and could result in public commentary being reduced to banal niceties (indeed, I wouldn’t be surprised if, at some point in the future, Josephson himself, came across this blog entry). I’m certainly not immune to such pressures and while I do actively try to ignore such influences, self-perception is not the most reliable guide. And it is often easy to unconsciously self-censor so I’d be interested to hear any feedback on whether people, who get this far, feel that I am being open and honest about my views or hedging my comments. And if you did get this far, thanks for reading… I’ll try to be shorter in the future!


  1. Hi Chris, I’m afraid I’ve forgotten most of what I read 25 or so years ago on Buddhism and how it came to Japan, and that is, or rather was, the extent of any knowledge I might have had concerning something that might, or might not, be considered “religion” in Japan.

    I did, however, read to the very end 🙂

    – Mark


  2. “whether people have a notion of demography or economy or not, they all have demography and they all engage in economic transactions”

    — I agree with Boyer (and your) criticism.

    I hear tinges of the strong Whorf-Sapir ghost: Language determines thought. But as you and I know, Chris, they may not have a word or a phrase, but they got a way to get the same ideas across.

    Yet, like you to, I hear and agree with the cautions and criticisms Josephson and others: the term often is a trojan horse with all sorts of stuff snuck in with it — even in “academic” circles.

    So, what is the answer:
    (1) avoid generalizations about “religion” when possible
    (2) be aware of the trojan horse
    (3) pay attention to language and don’t be tricked by it
    (4) sure, use “religion” if you are convinced you and your speaking partner are doing the above — but often one of you aren’t.

    Postscript: Yes, this post was way too long, with too many links to things needed to understand it. I read it all, but was constant lost in the verbosity. Which appeared a bit ironic in relation to this criticism you made:

    Deconstruction is often presented as if it is something incredibly complex and theoretically deep but in my experience it often means criticising theoretical concepts, in an unnecessarily verbose manner, while emphasising the (often unacknowledged) subjective aspects of research/writing.

    Liked by 1 person

  3. lol, fair comments Sabio!

    If I write about something I’ve researched I often elaborate/reference ad nauseam, this post has actually been edited down quite a bit, something of an occupational hazard I guess! But to try and present a (partial) defence, my issue with postmodern verbosity isn’t in regards to length but rather in terms of clarity. I don’t have an issue with long pieces or ‘big words’ when used appropriately, but when they are used primarily to make simple ideas sound more complex, which is often the case in works of deconstruction, I strongly object. I might be guilty of using academic babble at times but if so its certainly an unintentional failing rather than a conscious strategy.

    I do however consciously chose to link out a fair bit because I prefer for people to have the option to check for themselves whether I am presenting something fairly or not and/or not have the hassle of hunting out an article/post I’ve already found. I don’t typically expect people to read all the links I provide, although you may well be right, that I’m not appropriately gauging levels of background knowledge- this is a bit ‘inside baseball’ after all!

    Oh and to be clear I agree with your four suggestions. I would also add:

    5) Be consistent. If you argue ‘religion’ is an unsuitable category, the same reasoning should be applied to other similar concepts, even fashionable ones like patriarchy, for instance.


  4. Yeah, I understand the verbosity thing and the links. I like the links — but that meant it took me 4 days to read this post! 🙂

    Concerning #5, I like it but I may instead say: “Religion” is a flimsy, easily and often abused category, usually only best for inhouse back slapping. So if you use it otherwise, be careful and accept criticism humbly!


    Liked by 1 person

  5. Certainly only a partial defence 😉

    I do actually sympathise, and enjoy, when people like Josephson do the good kind of deconstruction- highlighting the artificiality and historical construction involved with ‘religion’ in Japan- but I think that ultimately you can perform the same exercise on every single categorising concept, especially those that would be used comparatively.

    For example:
    Does the concept of a ‘mother’ really mean the same thing in ancient Egypt that it now means in modern Europe? Are there not culturally specific connotations to the modern term which have been heavily influenced by European traditional family structures and the legal legacies of normative government paternalism, along with more recent attempts to legislatively and socially instill gender equality that make it impossible to compare with ancient Egyptian ‘mothers’. And in fact there are over a dozen translations for the concept of mother in ancient Egyptian, including definitions which do not include concepts of biological parentage… and so on.

    The individual points may be valid and the contextual nuances are definitely worth considering but I still think its not only useful but entirely meaningful to recognise this but still employ broad comparative categories, like ‘mother’, ‘Buddhism’ and ‘religion’. I know I’m not saying anything new here, but I am arguing against the notion that ‘religion’ is much more flimsy than the other comparative analytical concepts we regularly use, typically without objection.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s