The future of writing

A well-read academic colleague of mine once remarked that books that become important in the intellectual sphere are generally “well written.” That observation could apply to novels, fiction, non-fiction, popular science, academic books or any other genre. The judgement about what makes writing “good” presumably includes the quality of its arguments, messages, narrative content, but also the phraseology, use of grammar, choice of words, metaphors and idioms.

Most reviewers and critics are aware that content and style elide (“the medium is the message”). But let’s subscribe to the autonomy of “style” for a moment. I think of Strunk’s Elements of Style and Ernest Gower’s The Complete Plain Words, and their focus on clarity, simplicity, and of course appropriate use of grammar, metaphor and idiom. Style concerns how you deliver your message as opposed to what you want to say– the message stripped of style.

Now we can deploy LLMs to tidy up our prose, correct spelling and grammar, and turn dull wordy texts into writing that is colourful yet succinct. So, here’s a future scenario of writing practice that may well take hold as LLMs come into common use.

Style and substance

We might think of asking an LLM to correct and enhance a poorly written draft before publishing it as a blog, article, book or advertising copy. But might writing practice reach a stage where we need to pay less attention to the quality of the author’s output? Can we leave it to an LLM further along from the authoring task in the work flow to adapt, refine, or customise the material in line with a reader’s style preference or receptivity?

Here are the bones of a short essay about “luck.” I present it as a series of points. A reader would need to exert some effort to receive it as a coherent and engaging narrative. You could say it is heavier on substance than on style.

Imagine two contrasting worlds: One where random events mostly have positive outcomes; Another where misfortunes are expected and frequent; In the lucky world: It only rains when you have an umbrella; You run out of petrol next to a petrol station; Dropped objects land safely (e.g. vase on a cushion, toast jam side up); In the unlucky world: Picnics are routinely rained on; Illness strikes just before holidays; Keys are dropped over stormwater drains; These worlds reflect psychological outlook more than statistical likelihood; The word “happy” derives from “hap,” meaning good luck or fortune; Related words like “happenstance” and “happen” carry this sense of serendipity; A world skewed toward favourable outcomes is perceived as “happy.”; Conversely, the world of bad outcomes is “unhappy.” Earth is arguably a lucky planet; Example: Jupiter’s gravity shields Earth from many space hazards; Life exists due to a fine balance of chemical, geological, and astronomical conditions; The film Luck (2022, dir. Peggy Holmes) explores dual worlds of good and bad fortune; In the good-luck world (as depicted in the film): Platforms always appear under pedestrians; Food lands perfectly when thrown; Litter always ends up in the correct bin; In the bad-luck world: Drinks spill immediately; Doors jam; Darts miss their target; The architectures of luck are telling: Pre-modern cities were shaped by chance before regulation; Disasters prompted formal controls; Good luck for some can entail bad luck for others; The duality of luck is also explored in: Animation and cartoons; Games of chance and play scenarios.

I prompted ChatGPT with some instructions about style: “Please convert this draft into a well-written short popular magazine article.” It did so without further input from me, even generating a title.

The Architecture of Luck: How We Imagine Fortunate and Unfortunate Worlds

Imagine two parallel worlds. In one, chance always leans in your favour. Rain falls only when you’ve remembered your umbrella. You run out of petrol just as a garage appears. Dropped objects land safely—toast lands jam-side up, a vase falls onto a cushion. In the other world, things go awry as if by design. Picnics are routinely spoiled by rain. Illness arrives just before the holidays. Keys fall through the gaps of a storm drain.

These contrasting scenarios reflect more than just imaginative extremes—they illustrate psychological outlooks. The concept of “happiness” itself is rooted in luck. The word “happy” comes from “hap,” an old word for fortune or chance. Related terms like “happenstance” and “perhaps” suggest that to be happy is, in some sense, to be lucky.

Perceiving the world as more or less fortunate shapes how we feel in it. A world where good things happen by chance can seem charmed, while one tilted toward mishap feels anxious or even hostile.

Even Earth, on a cosmic scale, might be considered a “lucky” planet. Life exists here thanks to a delicate interplay of atmospheric, geological, and astronomical conditions. The gravitational pull of Jupiter, for instance, helps deflect many space-borne threats. Fortunate accidents have shaped the conditions for life.

The 2022 animated film Luck (dir. Peggy Holmes) explores this idea through two imagined realms: one of perpetual good luck, the other of relentless bad. In the good-luck world, platforms appear just in time to catch you. Food lands perfectly. Litter always finds its way into the right bin. In the bad-luck world, doors jam, drinks spill instantly, and darts miss the board.

These imagined architectures of luck offer insight into how we shape and understand environments. Historically, cities were once shaped by chance—organic, unregulated, unpredictable. Over time, disasters gave rise to formal planning, building codes, and risk management: ways to reduce bad luck and institutionalise good.

But good fortune is rarely universal. What is lucky for one person may be unlucky for another. This duality plays out in everything from cartoons to games of chance, where luck is both plot device and philosophical puzzle.

Ultimately, the worlds we imagine—lucky or unlucky—are not just about outcomes. They tell us something deeper about what it means to live with uncertainty, and how we cope with the unpredictable forces that shape our lives.

To set up a contrast, I then asked the LLM platform to rewrite my draft points as the introduction to an academic article suitable for a readership versed in sociology, psychology and philosophy.

Introduction: Architectures of Luck

Imagine two contrasting experiential worlds. In the first, contingency consistently favours the subject: rain falls only when one carries an umbrella; petrol runs out adjacent to a filling station; a falling vase lands softly on a cushion; toast touches down jam-side up. In the second, misfortune prevails: picnics coincide with sudden downpours; illness strikes just prior to a long-anticipated holiday; keys fall irretrievably through a storm drain. These scenarios are not statistical representations of probability, but psychological and cultural configurations of luck—schemas through which individuals perceive and interpret everyday events.

The concept of happiness, etymologically linked to “hap” (meaning fortune or chance), foregrounds the constitutive role of luck in subjective well-being. Terms such as “happenstance” and “perhaps” further underscore the centrality of contingency in human experience. To inhabit a “happy” world, in this linguistic and philosophical sense, is to live in a world that appears inclined toward serendipity and positive resolution. Conversely, an “unhappy” world is one where events tend to conspire against one’s intentions.

This paper explores the cultural, spatial, and psychological imaginaries of luck and misfortune. It draws on examples ranging from animated narratives—such as Luck (2022, dir. Peggy Holmes), which literalises dual domains of good and bad fortune—to urban design and social regulation. The built environment itself reflects evolving relations to contingency: pre-modern urban forms often emerged organically and irregularly, shaped as much by chance as by design; modern planning systems arose in part to mitigate risk, redistributing exposure to misfortune through regulatory infrastructures.

Yet, the distribution of good and bad luck is rarely neutral. It intersects with questions of inequality, systemic bias, and power. What constitutes good fortune for some may entail its opposite for others. This duality, embedded in narratives, spatial arrangements, and institutional practices, calls for closer examination. Through a cross-disciplinary lens, this article investigates how luck functions not only as a personal affective orientation but also as a socially and materially mediated phenomenon.

I like the use of the plural “architectures” in the title as a typical signal of academic pluralism. The rewording of the text adopts a questioning, less certain, format. Then there’s the use of academic terms such as “etymology,” contingency,” “imaginaries,” and “mitigate.” The rewrite expanded my original point “Good luck for some can entail bad luck for others” to “the distribution of good and bad luck is rarely neutral. It intersects with questions of inequality, systemic bias, and power.”

There are limits to this exercise of style adjustment. LLMs are trained so as to make no distinction between style and content. Were I to suggest that it style my piece as if written by Karl Marx, I would have to instruct it “Please re-write in the style of Marx’s Communist Manifesto, though do not change the content to reflect Marxist thought or philosophy.”

I provided that prompt and the platform obliged with a rewritten version “in the rhetorical and stylistic mode of The Communist Manifesto—aphoristic, declarative, and rhythmically insistent—while preserving [my] original conceptual content.”

The rewrite started “A spectre haunts the everyday—the spectre of luck. All the powers of chance and necessity, of happiness and misfortune, have entered into an endless struggle for the shape of our experience.” The style rewrite satisfies the prompt, but the emphasis on “struggle” does alter the message of the piece.

I would be happy with such translations if I knew that my author-sanctioned canonic original was preserved. Nor would I be concerned if someone had an LLM rewrite my prose to suit their reading style, i.e. making it more accessible to them.

But, I don’t know of any experienced writer who would be content to distribute their work as anodyne points ready to be styled by an AI over which they have no control. Writers are also more likely to want control over how their prose and style might be corrected or enhanced by an AI prior to publication. That said, attitudes to writing and reading, and their attendant crafts, will no doubt change under the influence of LLMs.

Note: I don’t usually compose by first laying out what I want to say in point form, though some writers do. My example of the points about “luck” were reverse engineered by ChatGPT from my blog post Lucky places.

References

• Gowers, E., S. Greenbaum and J. Whitcut. (1988), The Complete Plain Words, Boston: D.R. Godine.
• Strunk, W., E. B. White and M. Kalman. (2005), The Elements of Style, New York: Penguin Press.

Note

• Featured image generated by ChatGPT: “create an image of a bunch of house keys on top of a stormwater drain grating.”