Category Archives: Books

Buddhism 2.0? Stephen Batchelor’s vision of Secular Buddhism

At the time I was writing ‘The Buddha’s Middle Way’ (published this year), in 2017, I was able to refer to Stephen Batchelor’s 2015 book ‘After Buddhism’ as its nearest forerunner, but I had not yet read Batchelor’s collection of essays ‘Secular Buddhism’, which was only published in 2017. Reading that further book, more recently, I don’t think it would have made a huge difference to any of the (largely positive) things I said about Batchelor’s work. However, its essay ‘A Secular Buddhism’ does seem to be the nearest thing to a manifesto Batchelor has produced, and I guess that it’s one of the most influential statements to date of what ‘secular Buddhism’ means. For that reason, if no other, I feel it’s worth engaging with and responding to.

The reason I’ve been increasingly interested in Batchelor’s work, and have also been very glad to meet him personally, is not because I identify myself with the label ‘secular Buddhist’ (though I have been through an earlier phase when I used the term). However, I continue to be interested in what lies behind it – namely a fruitful process of deep critical thinking about Buddhist tradition, parallel to the similar process that is going on in relation to many other traditions.

Batchelor’s other metaphor for what he means by ‘Secular Buddhism’, ‘Buddhism 2.0’ appeals to me a little more. Despite its IT connotations, and the danger that these connotations will be taken too literally or reductively, it does convey the key idea of changing the paradigm that creates key assumptions about Buddhist practice. To do ‘Buddhism 2.0’ you don’t have to change your “hardware” (i.e. your body and brain) nor your specific app (e.g. ethical or meditation practice), but you do need to change the “operating system” that gives a wider formatting to your practice. You need to do so because the old operating system is no longer well-adapted to a new set of conditions. That’s a metaphor that could just as easily be applied to the reform of other traditions: e.g. Christianity 2.0, Science 2.0, or Liberalism 2.0.

The switch from one operating system to another may also seem like a sudden one, because it involves a disengagement from one set of connected assumptions in order to re-engage with another. Batchelor captures this very well when he writes about a “gestalt switch” from a metaphysical interpretation of Buddhist doctrines to a pragmatic one. In the work of Robert Kegan on levels of adult psychological development, that switch can be seen as the one between level 4 (thinking based on paradigmatic rational rules) to level 5 (flexibly moving between paradigms with a practical justification).

When this switch occurs, how does the new ‘operating system’ differ from the old one? Here, the key point where I agree with Batchelor wholeheartedly is that the new operating system is pragmatic whilst the old is dogmatic. The question, however, is what exactly “pragmatic” means. The implications of pragmatism need to be simultaneously pursued both in theory and in practise if they are to provide a strong enough alternative to the dogmas people are used to relying on. The more long-term and universal one tries to make one’s pragmatism, the further one will need to look beyond one’s personal experience and one’s immediate audience, and thus the wider and more fruitful one’s pragmatism is likely to become. Batchelor’s strength is that of communicating new possible approaches to people who are interested in Buddhism in a way that can inspire them and is consistent with practical experience. However, his limitations lie more in the area of how much he critically examines the perspectives he offers, to make his pragmatism more universal.

This takes me to my concerns about the framing of ‘Buddhism 2.0’ as ‘Secular Buddhism’. In his essay ‘Secular Buddhism’, Batchelor starts off by trying to clarify exactly what he means by that term. Let’s start with ‘Secular’. Batchelor says that there are three senses of secularity he is using: (1) opposition to religion, (2) being concerned with this world, and (3) involving a transfer of authority from ‘church’ to ‘state’. If that is what ‘secular’ means, then identifying my approach with any of them would worry me, because they all seem to rely on a (thoroughly false, in my view) dichotomy between ‘secularity’ and ‘religion’. If we state that we are concerned with this world, that seems to imply that we are still avoiding concern with another one, and if we are transferring authority away from religious institutions, this also suggests that we are rejecting religious institutions as holders of power. Surely that is continuing with the old paradigm, but just flipping our priorities within that paradigm, rather than adopting a new one? To adopt a new pragmatic paradigm, surely we need to apply the same sorts of autonomous criteria to critique both ‘religious’ and ‘non-religious’ spheres, recognising that religion is a complex system embedded in human societies?

There are, indeed, other possible uses of the term ‘secularism’, that I think may get a bit closer to what I take Batchelor to mean. Charles Taylor, in his big book ‘A Secular Age’ offers another alternative sense of secularity, as a transition “from a society in which it was virtually impossible not to believe in God, to one in which faith, even for the staunchest believer, is one human possibility among others”. The phenomenon of secular thinking has given us options that were not there before: to follow the church, to think oppositely to the church, and potentially, also, to break the dichotomous paradigm on which the popular view of religion as absolute belief depends. That’s a type of secularism I would personally find it much easier to sign up to: both individually and socio-politically, I think it’s very important to maintain those options. But secularists would need to be a great deal clearer about this as the prime reason for secularism than they tend to be for me to sign on the dotted line. Batchelor’s account of it does not suggest that it’s this kind of definition that he has in mind.

The other side of Batchelor’s account of ‘Secular Buddhism’ is the ‘Buddhist’ part. Here it may be helpful to quote him at some length.

On what grounds would such a Buddhism 2.0 be able to claim that it is “Buddhism” rather than something else altogether? Clearly, it would need to be founded on canonical source texts, be able to offer a convincing interpretation of key practices, doctrines, and ethical precepts, and to provide a sufficiently rich and integrated theoretical model of the dharma to serve as the basis for a flourishing human existence. (p.80)

Firstly, here, I wonder why he is so concerned as to whether ‘Secular Buddhism’ could possibly be mistaken for “something else altogether”. It’s a worry I’ve also encountered amongst many other Buddhists when discussing these issues, to such an extent that when I asked one Buddhist scholar why he was so narrowly focused on the Buddhist tradition, he answered “because I’m a Buddhist” – as though that answered the question! It should hardly be necessary to point out to Buddhists that essentialism is not part of their brief. Surely, if the practices work, it matters not in the least whether they are thought of as ‘Buddhism’? What matters practically, in relation to the Buddhist tradition, is whether it is a source of inspiration and practical support for spiritual progress, not what we call it. Its practical function as a tradition does depend on continuity, but not on essentiality, and helpful continuity can be maintained without essential identity.

Charitably assuming, then, that what Batchelor wants here more deeply is an effective practical relationship to Buddhist tradition rather than ultimate grounds for claiming that it is not ‘something else’, we can then find three criteria for it in the quotation above, only two of which are obviously pragmatic. A convincing interpretation of all the elements of the tradition is undoubtedly needed to maintain helpful continuity with it, and “a sufficiently rich and integrated model of it” to be practically helpful is also a pragmatic criterion. Why ever, though, does it need to be “founded on canonical source texts”?
There seems to be a contradiction between this criterion and Batchelor’s approach on the very next page:

The more I am seduced by the force of my own arguments, the more I am tempted to imagine that my secular version of Buddhism is what the Buddha originally taught, which the traditional schools have either lost sight of or distorted. This would be a mistake, or it is impossible to read the historical Buddha’s mind in order to know what he “really” meant or intended. (p.81)

Why does secular Buddhism need to be “founded on canonical source texts” if these very source texts have such a highly debatable relationship to the Buddha? Even if the scholarly lines of transmission were clearer than they are, would there not also be a basic issue of responsibility here? If we have responsibility for our practice, surely we cannot take any “canonical source text” as a complete account of it? Once we take responsibility for an idea, its source becomes irrelevant.

However, responsibility needs to be exercised in interpreting Batchelor as well as in interpreting scriptures. I’m fairly sure, having met him and discussed some of these issues, that he would deny that he intended “canonical source texts” to be an absolute source of authority. Nevertheless, I think there is a major issue about what one implies when one argues about such texts. In my own book ‘The Buddha’s Middle Way’, I have tried to make it extremely clear that I am using source texts from the Pali Canon as a source of inspiration rather than as a source of ‘truth’. Even then I found that some early readers of my draft book did not understand this approach, assuming that any reference to the texts was effectively an appeal to authority, and any changed interpretation must be based on a rival historical or textual claim of some kind. Appealing to authority is such an engrained habit in every sphere of religion, that one has to make a supreme effort even to open the Overton Window to other possibilities. People will still read in what they are used to even when you try to make it very explicit. In the absence of an extremely explicit statement about how one is using texts, then, I think it is very difficult to avoid the presumption, in readers if not in the writer, that one is offering a new version of what the Buddha “really” meant. For that reason I would like Batchelor to be very much more explicit on this point.

I think Batchelor’s concern with the dating and origin of texts also leaves him open to this interpretation. Sometimes the fact that one text is earlier or later than another is an informative part of the total story it offers and its significance. For example, knowing that ‘The Tempest’ is a late play of Shakespeare’s does make a difference to our appreciation of it, as we can understand the echoes of Shakespeare himself in the character of Prospero. In a similar way, awareness that the ‘Chapter of Eights’ in the Sutta Nipata is probably a very early text may help us to interpret it contextually. However, the dating of religious texts is very often part of an authority game in which the earlier text is taken to ‘win’ as the reward for a convincing (but unavoidably fallible) scholarly argument. In such cases, it has nothing much to do with the meaning of the text. Claims about dating may become hostages to fortune, and the practical meaning of the text very quickly becomes submerged in scholarly competitiveness. I very much feel that serious pragmatism demands agnosticism about claims that are heavily associated with the authority of texts, at least as the default option.

Overall, then, I think Batchelor’s account of secular Buddhism, though it has the great virtue of engaging many people in a pragmatic critique of Buddhist tradition, leaves a great deal that is of importance still unresolved or unnecessarily open to unhelpful interpretations. One reason for this is that it tries to use the concept of ‘secularism’ for a purpose for which it is ill-equipped. The other is that it remains unclear in practice how committed Batchelor is to a pragmatic interpretation, rather than one that continues (at least implicitly) to rely on tradition through the historical appeal to canonical sources.

Of course, I also think that one underlying reason for these limitations in Batchelor’s account is his neglect of the Middle Way, which really only gets passing mentions here and there, and is never really highlighted as important, despite all Batchelor’s discussion of the character and teachings of the Buddha. As Batchelor was kind enough to write an appreciative foreword for ‘The Buddha’s Middle Way’, I hope this may change in future. The key point missing here so far, though, is that, although Batchelor recognises that metaphysics is a problem, he doesn’t show any recognition that negative metaphysics is just as much of a problem as positive. As negative metaphysics in reaction to positive is such a feature of many interpretations of secularism, this is obviously a crucial reason for being cautious about identifying oneself with it.

The MWS Podcast 149: David Robson on the intelligence trap

My guest today is David Robson, David is an award-winning science writer and editor, who specialises in writing in-depth articles probing the extremes of the human mind, body and behaviour. He was a features editor at New Scientist for five years and is currently a senior journalist at BBC Future. He regularly features on the BBC World Service discussing scientific issues, and his writing has also appeared in Guardian, the Atlantic and the Washington Post. His first book ‘The Intelligence Trap: Why Smart People Make Stupid Mistakes and How to Make Wiser Decisions was published earlier this year and this will be the topic of our discussion today

Stream podcast from the Itunes Library

Click here to view other podcasts

The MWS Podcast 148: Sally Kohn on the Opposite of Hate

Our guest today is Sally Kohn, who is arguably one of the leading progressive voices in America. A frequent guest on CNN, MSNBC and Fox News. Sally is a popular keynote speaker including most recently with the Forgiveness Project , talking about political division, hate, otherizing, diversity and identity — and how we can solve the deep problems of our past and present. Her first book ‘The Opposite of Hate’ came out last year and will be the topic of our discussion today.

MWS Podcast 148: Sally Kohn as audio only:

Download audio: MWS_Podcast_148_Sally_Kohn

Stream podcast from Itunes Library

Click here to view other podcasts

The Age of Wonder

Over-specialisation, with its attendant false dualities between different areas of thought, is in my view a major issue of our time. It even affects our widespread current inability to face up to the climate crisis, as the authority of specialised experts is such a matter of conflict wherever scientists have to communicate unpalatable findings to the wider population. Surely, I often reflect, if we had a slightly more adequate education system in which both critical thinking and general psychological awareness featured more strongly, it would be a good deal harder for opportunistic denialists to gain such a popular following, and the politics of facing up to difficulty would be at least a bit easier? There are potential future ways of healing the rift – if we have a future, that is.

From time to time, though, I find contemplating the current and likely future situation too much to bear, and take refuge in the past. The past, particularly, before all this specialisation – of the gentleman scholar and the polymathic clergyman. In no way do I want to idealise the past, but it does show us in some ways, not only that things can be done differently, but that they have been. The difference is nearly always that in the past, the things that were better only applied to a very small minority, whereas now our expectations have risen to such an extent that we expect everyone to share in any improvement. When I look (with a temptation of nostalgia) at a less specialised past I have to keep reminding myself of this. Nevertheless, the intellectual integration that some groups managed in the past can still potentially be developed by a wider population in the future.

My appreciation of a pre-specialist past has been boosted recently by reading a book by Richard Holmes (best known as a biographer of Romantic English poets), called ‘The Age of Wonder’, published in 2008. This book offers an account of a crucial phase in the development of science (focusing especially on Britain), from around 1770 to around 1830, but also highlights its constant connections with the Romantic movement. It may come as news to many that during this period all scientists, as well as other thinkers, were still known as ‘philosophers’, as they were in ancient Greece. Such great figures as Joseph Banks (botanist and anthropologist), Humphrey Davy (chemist) and William Herschel (astronomer) described themselves as ‘natural philosophers’. Herschel was also a musician and Davy a poet. Davy was a close friend of Coleridge, Southey and other poets: the arts and the sciences actually talked to each other and maintained an integrated view of the world!

The theme of wonder, that gives the title to the book, is an important one. Poets and scientists were united by wonder, and wonder is an open emotion implying a strong continuing access to experiential right hemisphere perspectives. Though of course none of these figures were free of dogmatic assumptions, their sense of wonder, and their social sharing of it, could provide a check to the tendency of a specialised theoretical perspective to assume that its view of the world provides a total explanation. Not only did this make these great early scientists apparently open to new forms of discovery, but it also made them much more flexible in their use of their expertise than one would expect from a modern academic. Sir Humphrey Davy not only pioneered our understanding of the constituent gases of the atmosphere, the carbon cycle, and the isolation of substances like chlorine and iodine through electrolysis, but he also developed a safety lamp that greatly reduced the likelihood of coal miners dying in gas explosions. Today that would be considered the work of a technologist, not a scientist, but in Davy’s day there were no such distinctions.

There was wonder, too, in William Herschel’s astronomical achievements. Not only did he discover Uranus, but he was also the first to recognise the vast distances between the Earth and the stars, together with the existence of galaxies. He did this partly by developing larger and better telescopes, but also (on his own account) by just learning to look at the stars. Observation became not just a matter of passively watching, but rather a matter of actively altering your method of observation so that the visual data coming through the telescope made sense. There was wondering receptivity as well as theoretical expectation there, in a form that can add to our confidence that Herschel was not merely observing what he wanted to observe. We can never entirely rid ourselves of confirmation bias, but we can definitely make a difference.

In the hands of these inspirational early scientists, science often seems like a Middle Way practice – one in which we are constantly adjusting our view of the world in the light of new experiences and perspectives, rather than seeking only to confirm or deny. Of course, there are also many modern scientists who take this approach. There are also far more educated people in the population able to understand what they do than there were in 1830. Nevertheless, it seems to me that something important has been lost, not always in individual attitudes, but primarily in the educational and academic system that frequently isolates those individuals from each others’ perspectives. Academics can usually only compete effectively in their specialised niches by hardening their assumptions to fit the social expectations of that discipline. The relentless managerialist emphasis on ‘efficiency’ of research in modern academic life also removes much of the leisure enjoyed by those in the Age of Wonder, and so further closes the doors of wider exploration.

There are also many signs of a thirst for a return for that lost interdisciplinarity today. Recently I was intrigued to learn of the launch of the London Interdisciplinary School, due to open its doors in 2020 to an entirely interdisciplinary cohort of students. They were looking for founding faculty, so I applied, but I’ve since heard that they had 612 applications for 6 places. I thus don’t expect to be successful, but this tells you something about how much pent-up longing there is to get back, in some sense, to the age of philosophers worthy of the name (ones who are not just experts in Dismissiveness Studies).

The Age of Wonder is after all still going on, and will continue as long as we are capable of wonder.  We’ll be capable of wonder as long as we can reach out beyond our own little niches of assumption – whatever they are. As long as we’re capable of wonder, we’ll also be capable of creative discovery.

The MWS Podcast 147: Robert M Ellis on the Buddha’s Middle Way

We are joined today by the philosopher and founder of the Middle Way Society. Robert has been a regular guest on the podcast and is the author of a range of books on Middle Way Philosophy, both within and beyond Buddhism, including The Christian Middle Way (Christian Alternative 2018). He has a PhD in Philosophy and a Cambridge BA in Oriental Studies and Theology. He has taught in many different contexts, and was formerly a member of the Triratna Buddhist Order. He’s here to talk to us about his latest book The Buddha’s Middle Way (Equinox 2019).


MWS Podcast 147: Robert M Ellis as audio only:

Download audio: MWS_Podcast_147_Robert_M_Ellis

Stream podcast from Itunes Library

Click here to view other podcasts