Web dev at the end of the world, from Hveragerði, Iceland

Intellectual terrain

Books today are for sharing, not reading

Literature is the art of writing something that will be read twice; journalism what will be read once. Cyril Connolly, Enemies of Promise

I’ve been reading Cyril Connelly’s Enemies of Promise. It is wonderful, brilliant, and meandering; analytical where complexity requires it to analyse; spiritual where the soul needs to be fed; and optimistic just when your spirit is about to break.

It also manages to make you think about what you’re doing and where you’re coming from.

Which is humbling.

Despite the wide ground it covers — style, autobiography, grammar — it maintains a steady focus on the subject of promise, what it means to be a promising writer and how it either pans out or doesn’t.

It’s meandering in the same way that a hiker meanders. Like Connelly, the hiker has a destination and they aren’t diverging from their path, but the terrain they are covering simply doesn’t lend itself to direct routes. You can’t run a marathon or sprint without a road or a track. Uneven terrain requires a wandering path.

Modern writing, the chatter that fills websites, newspapers, and short ebooks, doesn’t account for terrain. They are mental sprints — short bursts along a paved road where everything uneven and unnatural has been removed, cut away, or flattened. The longer books might qualify as marathons, but they still only track along the ready-made roads of pre-fabricated ideology and and cookie-cutter abstract arguments.

These books all gloss over the uneven and divergent nature of their subject matter because they aren’t about what they claim to be about. They are statements, shibboleths, flags to rally behind. The purpose of writing a book today that is sceptical of the role the web plays in modern society isn’t to critique that role except in the most shallow, obvious, way possible. The criticisms laid out are there to demonstrate which side you’re on, not to engage with ‘the enemy’. These books are a box of shibboleths. A set of markers you can pull out and use in conversation to demonstrate what sort of person you are, where you stand in relation to other social groups, and what your social group’s attitude towards technology is.

The purpose of writing an essay, book, or blog post that is in favour of ‘digital’, a book that highlights the benefits and possibilities of the web, ebooks, or whatever inane buzzword dominates, isn’t to educate people about how they can benefit from technological improvements. It isn’t to guide technology into a new and glorious future. It isn’t to criticise technology when it diverges from the ideal and guide it back to its proper path.

The purpose is identity production. Each book is an identity factory for its buyers. People read and share the blog posts that show the world who they are. Display the print book on your desk or shelf (if you’re not in the ebook camp) to show what sort of person you are. Post quotes from books to establish your identity. Reuse prefabricated arguments and points in conversation to claim a group as a part of your identity. Sincerely engaging with those who disagree is the last thing you want to do because the differences between you is what defines you. Your identity hinges on your group having divergent opinions from other groups. There can be no debate, only confrontation. The only discussions allowed are those that sharpen the lines between you and them. They are skirmishes to establish the exact real-world position of the borders, not intellectual discourse, and certainly not thought.

Modern publishing is hay for cattle.

Any subject that lends itself to a laser-like focus with no deviations is highly abstract and doesn’t account for unpredictable elements such as other human beings. The only way you can avoid meandering and wandering when you’re tackling a difficult subject is by picking a single, clearly defined, problem — it has to be a tiny one as well — and write just about that problem with no deviations, no connections, no digressions. The role of social media in our lives is too big of a problem. As are both the print to ebook transition and the growing role of the web in everyday life.

To avoid meandering you need to write about something like the price variations of impressionistic paintings sold in the period 1970–1980.

Or, the latest crap belched out by tech companies.

Go wider than that and you begin to involve the vagaries of human behaviour and the subject matter becomes divergent; the deeper you delve into it, the more you analyse it, the more complicated it becomes.

Books that go deep can’t be used for identity production because they resist being broken down into encapsulated, standalone, statements. Or, if they can, that breakup causes an irreversible loss in the book’s meaning — a loss that bleeds back into misinterpretations of the book itself. It becomes read for the fragments claimed by one group or another for identity purposes and the argument made by the book as a whole is elided.


Cyril Connelly’s book is simultaneously an autobiography and a thoughtful study of literary styles and trends. Were somebody to try and tackle the same subject matter today—i.e. what does it mean to be a promising writer and how do you write a book that lasts ten years?—they would integrate the autobiographical bits into the main text as a set of cute Gladwellian anecdotes that directly support the current argument.

But Connelly builds up the two halves of his book as exactly that: two halves. They are not integrated but they are inseparable. They are not cute. They are not easy to digest. They are not immediately persuasive. Parts of the book are harrowing. They don’t support the argument of the book but they do explore it and build the context necessary to engage with the argument. They are two halves that are effective precisely because they are allowed, in each section, to explore the terrain as deeply as is necessary.

You can’t do that in a book any more, not if you want to be ‘persuasive’, appeal to your chosen demographic, and be published by somebody other than yourself. You can’t do that in a blog, not if you want your readers to stick around. You might be able to do that in a self-published ebook, but most of the freelance editors self-publisher’s hire these days would see the style as a flaw to be fixed.

It’s a writing structure and tactic with no place in modern publishing.

One of the most effective ways of intellectually engaging with complex—potentially unsolvable—problems has become little more than bad writing.


Since Hannah asked, here are a few suggestions for further reading. Not necessarily directly related to the argument here, but they build context for it.

Leo Tolstoy’s What is Art? is proof that the problems described here aren’t new (as Simon St.Laurent pointed out, although, I do think it has gotten much worse in modern publishing). There Tolstoy ranted against art becoming more and more based on borrowing and on the memory of other art previously experienced.

If you like the style and grammar parts of Connelly’s book and want to see how different the rhetoric around writing was decades ago, you can’t go wrong with Arthur Quiller-Couch’s On the Art of Writing.

My argument on the deterioration of rhetoric and deep writing is little more than a complete rip-off of Amusing Ourselves to Death by Neil Postman.

Some of my ideas here on the role of groups and identity come from papers on group polarisation and attitude polarisation. Dig into the references on the wikipedia pages to get a good overview.

My favourite book that discusses the issue of problems from a high level is E.F. Schumacher’s A Guide for the Perplexed. It contains the best definition of the difference between convergent and divergent problems that I’ve seen yet. Divergent problems as Schumacher says ‘do not yield to “straight-line” logic’. They are the rough intellectual terrain that requires meandering and exploration if you are to tackle them. His economics book Small is Beautiful is brilliant as well.

You can also find me on Mastodon and Twitter