An artist's rendering of spaceships over a city, casting yellow tractor beams down as the skies darken with clouds above.

A future obscured by exponential growth

A couple months into the COVID-19 pandemic, I think most of us realised how hard it is to comprehend the phenomenon of exponential growth. Mathematically, it’s trivial – a geometric progression – but more physically, the difference between linear and exponential growth is very non-trivial, as a cause-effect chain where each effect leads to multiple new cases according to a fixed growth ratio. The effect is an inability to fully anticipate future outcomes – to prepare mentally for the ‘speed’ with which an exponential series can scale up – rendered remarkable by us not having planned for it.

For example, the rice and chessboard problem is a wonderful story to tell because it’s hard for most people to see the punchline coming. To quote from Wikipedia: “If a chessboard were to have wheat placed upon each square such that one grain were placed on the first square, two on the second, four on the third, and so on (doubling the number of grains on each subsequent square), how many grains of wheat would be on the chessboard at the finish?” The answer is 18,446,744,073,709,551,615 – a 100-million-times greater than the number of stars in the Milky Way. Many people I know have become benumbed by the scale of India’s COVID-19 epidemic, which zipped from 86k active cases on May 30 to 545k on July 31, and from 1M total cases on July 17 to 7.3M on October 15. On August 1, 1965, Vikram Sarabhai delivered the convocation address at IIT Madras, which included the following quip:

Everyone here is undoubtedly familiar with the expression ‘three raised to the power of eighteen’. It is a large number: 38,74,20,489, thirty-eight crore, seventy-four lakh, twenty thousand, four hundred and eighty-nine. What it means in dynamic terms is quite dramatic. If a person spreads gossip to just three others and the same is passed on by each of them to three others, and so on in succession, in just eighteen steps almost the entire population of India would share the spicy story.

Because of its mathematical triviality and physical non-triviality, I think we have a tendency to abstract away our impression of exponential growth – to banish it out of our imagination and lock it away into mathematical equations, such that we plug in some numbers and extract the answers without being able to immediately, intuitively, visualise or comprehend the magnitude of change, the delta as it were, in any other sense-based or emotional way. And by doing so, we are constantly surprised by the delta every time we’re confronted with it. Say the COVID-19 epidemic in India had a basic reproductive number of 1.4, and that everyone was familiar with this figure. But simply knowing this value, and the fundamental structure of a geometric progression, doesn’t prepare people for the answer. They know it’s not supposed to be N after N steps, but they’re typically not prepared for the magnitude of 1.4^N either.

I recently came across a physical manifestation of this phenomenon in a different arena – technology – through a Twitter account. The oldest Homo sapiens technologies include fire, tool-making, wheels and cropping. But while the recursive application of these technologies alone may have given rise, in a millennium (i.e. 1,000 steps), to, say, a subsistence agriculture economy with some trade, that’s not what happened. Instead, two other things did (extremely broadly speaking): the technologies cut down the time required for different processes, and which subsequently came to be occupied by the application of these technologies to solve other problems. The geometric-like progression that followed exponentiated not the technologies themselves but these two principles, of sorts, rapidly opening up new methods and opportunities to extract value from our surroundings, and eventually from ourselves, to add to the globalising value chain.

To get a quick sense of the rapidity of this progress, check out @MachinePix on Twitter. Their latest tweet (as of 11 am on October 17) describes a machine that provides a “motion-compensated” gangway for workers moving between a ship and an offshore wind turbine; many others depict ingenious contraptions ranging from joyously simple to elegantly complicated – from tape-dispensers and trains windows that auto-tint to automated food-packaging and super-scoopers. There’s even a face-mask gun that seems to deliver an amount of pain suitable for anti-maskers.

But closer to the point of this discussion: taken together, @MachinePix’s tweets demonstrate the extent to which we have simplified and/or automated different processes, and the amount of time humans have collectively saved as a result. This, again, can’t be a straightforward calculation: we don’t just apply the same technologies over and over to perform the same tasks. We also apply technologies to each other to compound or even modify their effects, effectively leading to new technologies and, thus, new applications – from the level of toothbrush plus toothpaste to liquefaction plus rocket engines. The tools we develop also alter the structure of society, which in turn changes aspirations and leads to the birth of yet more technologies, but ordered along different priorities.

In the last few months, I learnt many of these features in an intimate way through Factorio, a video-game that released earlier this year. The premise is that your spaceship has crashed on an alien planet, with many of the same natural resources as Earth. You now need to work your way through a variety of technologies and industrial systems and ultimately build a rocket, and launch yourself off to Earth. The ‘engine’ at the game’s centre, the thing that drives your progress, is a recipe-based manufacturing system. You mine resources, process them into different products, combine them to make components, and combine the components to make machines. The machines automate some or all of these processes to make more sophisticated machines and robots, and so forth. To move objects, you use different kinds of inserters and conveyor belts; for fluids – from water to lubricant – there are pipes, tanks, even fluid wagons attached to trains.

A zoomed-out scene from Factorio. This is ‘Main Station’, one of five bases I operate in this scenario.

I’m still finding my way around the extent of the game; the technology tree is very high and has scores of branches. The scenario I’m currently playing goes beyond a rocket to using satellites, but doesn’t include the planet’s alien creatures, who attack your base if you antagonise them or pollute too much. I often think it would’ve been much better to allow final-year students of mechanical engineering (which I studied) to play this game instead of making them sit through hours of boring lectures on logistics, quality control, operations research, supply-chain management, etc. Factorio doesn’t set out to teach you these things but that’s what you learn – and on the way, you also discover how easy it is for things to get out of control, become too complicated, too chaotic – sometimes just too big to fail.

Sometimes, you’ve invested so much in developing one technology that you’re unable to back out, and you start to disprivilege other ambitions in favour of this one. This happened to me recently: being hell-bent on building nuclear reactors to keep up with the demand for power, I had to give up on building a satellite.

Instead of a linear or even a tree-like model of technology development, imagine a circular one: at the centre is the origin, and the circumference is where you are, the present (it’s not a single point in space-time; it’s multiple points in space at one time). Technologies emerge from the origin and branch out towards the perimeter in increasingly intricate branches. By the time they’ve reached the outer limits, to where you are, you have nuclear power, rocketry, robotic construction networks and high-grade weapons. But in this exponentially interconnected world, what do you change and where to effect a difference somewhere else? And how can you hope to be sure there won’t be any other effects?

My new favourite example of this, from the few-score @MachinePix tweets I’ve scrolled through thus far, is the rotary screen printer. It shows, among many other things, that there’s a second way in which exponential growth disrupts our ability to predict its outcomes. Could a fantasy writer working all those millennia ago have predicted this device’s existence? They may have, they may have not, just as we contemplate what the future might look like from today, but sometimes presume to anticipate – even though we really can’t – the full breadth of what lies in store for humankind. Can we even say if the rotary screen printer will still be around?

Featured image: An artist’s rendering of spaceships hovering above a city. More importantly, this image belongs to a genre quite popular in the 2000s, perhaps the late 1990s too, when image-editing software wasn’t as versatile as it is today and when the internet was only just beginning to democratise access to literature and videos, among other things, so the most common idea of first contact looked a lot like this. Credit: Javier Rodriguez/pixabay.

The climate change of bad news

This post flows a bit like the 1987 film Full Metal Jacket. As one friend put it, “It starts somewhere and then goes in a different direction.”

This year hasn’t been beset by the same old steady drizzle of bad news we have every year – but has borne the brunt of cyclonic storms, each one distinctively episodic and devastating. The latest of these storms is l’affair Rukmini Callimachi. To the uninitiated: Callimachi is a reporter with the NYT who shot to fame from 2015 or so onwards for her inside reports of the Islamic Caliphate; she later dramatised her efforts to produce these stories in a podcast called Caliphate. And in this time, she raked up four Pulitzer Prize nominations (although I don’t set much store by prizes in general).

I haven’t read or listened to her work, so when a friend shared a link to the NYT’s own report, by its media columnist Ben Smith, discussing the charges against Callimachi and their newfound, but evidently delayed, efforts to reevaluate her work, I wasn’t guilty of not having criticised her myself. (If you think this is a tall order: the headline of Jacob Silverman’s review of this storm for The New Republic describes, in a few words, how quickly her house of cards seems to fall down.)

However, these days, a successful journalist is two things: she is the producer of stories that have changed the world, and which continue to live lives of their own, and she is a role-model of sorts. Her output and her resolve represent what is possible if only one tried. An even greater example of such work is that of the journalists at the Miami Herald – especially Julie Brown – who exposed Jeffrey Epstein and brought on, among other changes, a reckoning at various universities around the US that had knowingly accepted his money and overtures.

But now, with Callimachi’s articles seemingly teetering on the brink of legitimacy, both the things she stood for are on the edge as well. First, the good thing: her stories, which – if Smith’s account is to be believed – Callimachi seems to have composed in her head before moving in to report them, often, if not always, with the spiritual and material support of many of NYT’s senior editors. Second, the bad: her legacy, such as it is – erected as a façade at which we could all marvel, at least those of us who unquestioningly placed our faith and hope in the greatness of another. This is the guilt I feel, a fractured reflection of what Callimachi’s coverage of the Islamic Caliphate at the NYT is itself going through right now.

However, I will also be quick to shed this guilt because I insist that as much as I’m tasked – by my employer, but the zeitgeist, so to speak – to be wary, cautious, skeptical, to fact-check, fact-check, fact-check, to maintain cupfuls of salt at hand so I’m never taken for a ride, just as much as I’m behooved to stand on guard, I’m also fortifying an increasingly small, and increasingly precious, garden in a corner of my mind, a place away from the bad news that I can visit in my daydreams, where I can recoup some hope and optimism. Today, the winds of l’affair Callimachi blew away her articles and podcasts from this place.

Make no mistake, I will still call out everything that deserves to be called out: from the multiple red-flags Silverman spotlighted to the anti-oriental undertones of Callimachi’s methods, of her claims and even of the self-recrimination bubbling up around her, to a lot of which Rafia Zakaria has (repeatedly) called attention. I’m only saddened, for now, by the unstoppable eradication of all that is good, such as it is, and by the guilt for my part in it. As a political being, in this moment I deem this march upon ignorance to be necessary, but as a human one, it is deeply, and to my mind unforeseeably, exacting. A cognitive dissonance for the times, I suppose, although I’m sure I will cope soon enough.

Fortunately, perhaps in a counterintuitive sense, the Callimachi episode is personally not very hard to recover from. While it is true that what Callimachi and her collaborators have (still largely allegedly) done is quite different from, say, what Jonah Lehrer did, they were both motivated by a common sin: to print what could be instead of what is (and even these words might be too strong). More specifically, reporting on war brings with it its own seductions, many of them quite powerful, to the extent that some – as Zakaria implied in her piece for The Baffler – may choose to believe Callimachi et al’s failings are still the failings of an institution vis-à-vis conflict journalism. But no, the problem is pervasive.

However, looking on this shitshow from not-so-distant India, two bells have been quick to go off. First, this is very old wine in a new bottle, in which, to borrow Zakaria’s words, “the greed for catching terrorists” is pressed into the service of making “white journalists’ careers”; you could replace ‘terrorists’ with anything else that has been touched, at any point in its history, by a colonist or invader. Also read Priyanka Borpujari’s 2019 essay in the Columbia Journalism Review, in which she writes:

The title ‘foreign correspondent’ has long been synonymous with whiteness, maleness, and imperialism—journalists fly in from North America, Europe, and Australia to cover the poverty and wars of the non-Western world. In recent years, a push for diversity has meant that more women are pursuing stories in what was once the domain of men—conflict zones and fractured democracies—or in traditionally private female spaces. But the opportunities for journalists in non-Western nations to tell their own stories in international outlets have not been as great. Overwhelmingly, foreign reportage still relies on a model of Western, and largely white, reporters hiring local journalists in subservient roles.”

And thanks to biases in the way technology is constructed, used as well as located around the world, the problem extends to the consumption of journalism as well. To quote from an older post:

Where an app [that amplifies content] was made matters because nobody is going to build an app in location A and hope that it becomes popular in faraway location B. Pocket itself is San Franciscan and the bias shows: most recommendations I’ve received, or even the non-personalised trending topics I’ve spotted, are American. In fact, among all the tools I use and curation services I follow, I’ve come across only two exceptions: the heartwarming human-curated 3QuarksDaily and Quora. I’m not familiar with Quora’s story but I’m sure it’s interesting – about how a Q&A platform out of Mountain View came to be dominated by Indian users.

I notice a not insignificant number of articles and essays, in English, to this day emerging from blogs and publications in Central, South and Southeast Asia, South America and of course Africa that will never go viral on Twitter, make it to the list of ‘most read’ articles on Pocket or be cited by even the most quirky columnist – even as the same ideas and arguments will virtually ‘break the internet’ the moment they emerge from The Atlantic or New Yorker a few months later.

None of the writers of The Atlantic or New Yorker can be blamed, at least not most of the time, for something quite hard to discover in the first place, but that doesn’t mean Big Tech isn’t distorting our view of who is doing good work and who isn’t. And many Indian journalists and writers are often at the wrong end of this discovery problem.

In this light, what Callimachi and the NYT did is not new at all but in fact further widens, or accentuates, the divide between being non-white, non-Western and being white and Western. This is a divide that I and many others, perhaps especially the others, have been habituated to ignore – especially when the crime at hand appears to be victimless but in fact quietly sidelines those who have already been historically, and today structurally, displaced from the ‘mainstream’.

On the other hand, what the NYT has perpetrated here is akin to what many in India (myself included) have done and, to different degrees, continue to have a part in. Specifically, the second bell that goes off has to do with my privileges, one product of which is that I will always be a parachute-journalist in my own country – a member of the top 1% who claims to understand the problems of the 99%.

Journalism professor Justin Martin gently defended parachute journalism in a 2011 essay, deeming fluency in “one of the main local languages” to be a prerequisite of parachuting well. I am not likely to speak any other languages than the four I already know, and less literally, I can never know, in any meaningful sense, what it means to be poor, transgender, tribal, of a lower caste; that lived experience will stay out of reach, and my assessment of what is right will always be inferior to those of, say, a desperate job-seeker, a transgender activist, a member of a tribe, a Dalit scholar when, for example, the topic at hand is poverty, gender, Indigenous people’s rights and caste.

As Martin also admits, “Hiring correspondents who live in the countries and regions they cover … is ideal”, and my higher social status in India does place me in a country other than the one I’m writing about. Although I may not be guilty of allowing information sources I haven’t vetted enough to feed exaggerated stories that I can’t prove in any other way to be true – that is, although we may not all be Rukmini Callimachis ourselves – the composition of our newsrooms means we are only one illegitimate source away, only one moment of weakness for what could be in place of what is away, from creating the next storm.