As is often the case with people who are waylaid by unpredictable events, for years afterwards my mind kept returning to my encounter with the tornado. Why had I walked down a road that I almost never took, just before it was struck by a phenomenon that was without historical precedent? To think of it in terms of chance and coincidence seemed only to impoverish the experience: it was like trying to understand a poem by counting the words. I found myself reaching instead for the opposite end of the spectrum of meaning – for the extraordinary, the inexplicable, the confounding. Yet these too did not do justice to my memory of the event.

Novelists inevitably mine their own experience when they write. Unusual events being necessarily limited in number, it is but natural that these should be excavated over and again in the hope of discovering a yet undiscovered vein.

No less than any other writer have I dug into my own past while writing fiction. By rights then, my encounter with the tornado should have been a mother lode, a gift to be mined to the last little nugget.

It is certainly true that storms, floods and unusual weather events do recur in my books, and this may well be a legacy of the tornado. Yet, oddly enough, no tornado has ever figured in my novels. Nor is this due to any lack of effort on my part. Indeed, the reason I still possess those cuttings from the Times of India is that I have returned to them often over the years, hoping to put them to use in a novel, but only to meet with failure at every attempt.

On the face of it there is no reason why such an event should be difficult to translate into fiction; after all, many novels are filled with strange happenings. Why then did I fail, despite my best efforts, to send a character down a road that is imminently to be struck by a tornado?

In reflecting on this, I find myself asking, what would I make of such a scene were I to come across it in a novel written by someone else? I suspect that my response would be one of incredulity; I would be inclined to think that the scene was a contrivance of last resort. Surely, only a writer whose imaginative resources were utterly depleted would fall back on a situation of such extreme improbability?

Unlikely though it may seem today, the nineteenth century was indeed a time when it was assumed, in both fiction and geology, that Nature was moderate and orderly: this was a distinctive mark of a new and “modern” worldview. Bankim goes out of his way to berate his contemporary, the poet Michael Madhusudan Datta, for his immoderate portrayals of Nature: “Mr Datta…wants repose. The winds rage their loudest when there is no necessity for the lightest puff. Clouds gather and pour down a deluge, when they need do nothing of the kind; and the sea grows terrible in its wrath, when everybody feels inclined to resent its interference.”

The victory of gradualist views in science was similarly won by characterising catastrophism as un-modern. In geology, the triumph of gradualist thinking was so complete that Alfred Wegener’s theory of continental drift, which posited upheavals of sudden and unimaginable violence, was for decades discounted and derided.

It is worth recalling that these habits of mind held sway until late in the twentieth century, especially among the general public. “As of the mid-1960s,” writes the historian John L Brooke, “a gradualist model of earth history and evolution…reigned supreme.” Even as late as 1985, the editorial page of the New York Times was inveighing against the asteroidal theory of dinosaur extinction: “Astronomers should leave to astrologers the task of seeking the causes of events in the stars.” As for professional palaeontologists, Elizabeth Kolbert notes, they reviled both the theory and its originators, Luis and Walter Alvarez: “‘The Cretaceous extinctions were gradual and the catastrophe theory is wrong,’…[a] paleontologist stated. But ‘simplistic theories will continue to come along to seduce a few scientists and enliven the covers of popular magazines’.”

In other words, gradualism became “a set of blinders” that eventually had to be put aside in favour of a view that recognises the “twin requirements of uniqueness to mark moments of time as distinctive, and lawfulness to establish a basis of intelligibility”.

Distinctive moments are no less important to modern novels than they are to any other forms of narrative, whether geological or historical. Ironically, this is nowhere more apparent than in Rajmohan’s Wife and Madame Bovary, in both of which chance and happenstance are crucial to the narrative. In Flaubert’s novel, for instance, the narrative pivots at a moment when Monsieur Bovary has an accidental encounter with his wife’s soon-to-be lover at the opera, just after an impassioned scene during which she has imagined that the lead singer “was looking at her…She longed to run to his arms, to take refuge in his strength, as in the incarnation of love itself, and to say to him, to cry out, ‘Take me away! carry me with you!’”

It could not, of course, be otherwise: if novels were not built upon a scaffolding of exceptional moments, writers would be faced with the Borgesian task of reproducing the world in its entirety. But the modern novel, unlike geology, has never been forced to confront the centrality of the improbable: the concealment of its scaffolding of events continues to be essential to its functioning. It is this that makes a certain kind of narrative a recognisably modern novel.

Here, then, is the irony of the “realist” novel: the very gestures with which it conjures up reality are actually a concealment of the real.

What this means in practice is that the calculus of probability that is deployed within the imaginary world of a novel is not the same as that which obtains outside it; this is why it is commonly said, “If this were in a novel, no one would believe it.” Within the pages of a novel an event that is only slightly improbable in real life – say, an unexpected encounter with a long-lost childhood friend – may seem wildly unlikely: the writer will have to work hard to make it appear persuasive.

If that is true of a small fluke of chance, consider how much harder a writer would have to work to set up a scene that is wildly improbable even in real life? For example, a scene in which a character is walking down a road at the precise moment when it is hit by an unheard-of weather phenomenon?

To introduce such happenings into a novel is in fact to court eviction from the mansion in which serious fiction has long been in residence; it is to risk banishment to the humbler dwellings that surround the manor house – those generic outhouses that were once known by names such as “the Gothic”, “the romance”, or “the melodrama”, and have now come to be called “fantasy”, “horror”, and “science fiction”.

Excerpted with permission from The Great Derangement: Climate Change and the Unthinkable, Amitav Ghosh, Allen Lane.