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.

‘Hunters’, sci-fi and pseudoscience

One of the ways in which pseudoscience is connected to authoritarian governments is through its newfound purpose and duty to supply an alternate intellectual tradition that subsumes science as well as culminates in the identitarian superiority of a race, culture or ethnic group. In return, aspects of the tradition are empowered by the regime both to legitimise it and to catalyse its adoption by the proverbial masses, tying faith in its precepts with agency, and of course giving itself divine sanction to rule.

The readers of this blog will recognise the spiritual features of Hindutva that the Bharatiya Janata Party regularly draws on that fit the bill. A German rocket scientist named Willy Ley who emigrated to the US before World War II published an essay entitled ‘Pseudoscience in Naziland’ in 1947, in which he describes the sort of crazy beliefs that prepared the ground with other conditions for the advent of Nazism.

In Hunters, the Amazon Prime show about Jewish Nazi-hunters in 1970s America, Edward Bulwer-Lytton’s sci-fi novel The Coming Race (1871) finds brief mention as a guiding text for neo-Nazis. In the novel, a subterranean race of angelic humanoids has acquired great power and superhuman abilities by manipulating a magical substance called Vril, and threatens to rise to the surface and destroy the human race one day.

Bulwer-Lytton also wrote that Vril alludes to electricity (i.e. the flow of electrons) and that The Coming Race is an allegory about how an older generation of people finds itself culturally and political incompatible with a new world order powered by electric power. (At the same time, he believed these forces were a subset of the aether, so to speak.) In a letter to John Forster on March 20, 1870 – precisely 150 years ago in twelve days – Bulwer-Lytton wrote:

I did not mean Vril for mesmerism, but for electricity, developed into uses as yet only dimly guessed, and including whatever there may be genuine in mesmerism, which I hold to be a mere branch current of the one great fluid pervading all nature. I am by no means, however, wedded to Vril, if you can suggest anything else to carry out this meaning – namely, that the coming race, though akin to us, has nevertheless acquired by hereditary transmission, etc., certain distinctions which make it a different species, and contains powers which we could not attain through a slow growth of time’ so that this race would not amalgamate with, but destroy us.

And yet this race, being in many respects better and milder than we are, ought not to be represented terrible, except through the impossibility of our tolerating them or they tolerating us, and they possess some powers of destruction denied to ourselves.

The collection of letters is available here.

In Bulwer-Lytton’s conception, higher technological prowess was born of hereditary traits. In a previous letter, dated March 15, Bulwer-Lytton had written to Forster:

The [manuscript] does not press for publication, so you can keep it during your excursion  and think over it among the other moonstricken productions which may have more professional demand on your attention. The only important point is to keen in view the Darwinian proposition that a coming race is destined to supplant our races, that such a race would be very gradually formed, and be indeed a new species developing itself out of our old one, that this process would be invisible to our eyes, and therefore in some region unknown to us.

So this is not a simple confusion or innocent ignorance. Bulwer-Lytton’s attribution of the invention of electricity to genetic ability was later appropriated by interwar German socialists.

This said, I’m not sure how much I can read into the reimagination of technological ability as a consequence of evolution or racial superiority because another part of Bulwer-Lytton’s letters suggests his example of electricity was incidental: “… in the course of the development [of the new species], the coming race will have acquired some peculiarities so distinct from our ways … and certain destructive powers which our science could not enable us to attain to, or cope with. Therefore, the idea of electrical power occurred to me, but some other might occur to you.”

Now, according to Ley, the Society for Truth believed Vril to be a real thing and used its existence to explain how the Britons created their empire. I don’t know how much stock Adolf Hitler and his “shites of the round table” (to quote from Hunters) placed in this idea but the parallels must have been inescapable – especially so since Ley also writes that not just any pseudoscientific belief could have supported Hitler’s rise nor have acquired his patronage. Instead, the beliefs had to be culturally specific to Germany, pandering to local folklore and provincialism.

Without commenting on whether this conclusion would apply to Fascism 2.0 in a world with the internet, civil aviation and computerised banking, and in naïve spite of history’s fondness for repeating itself and the politico-corporate-media complex, I wonder what lessons there are here – if any – for science educators, a people already caught between political anti-intellectualism and a stronger sense of their purpose in an intellectually debilitated society.

Science v. tech, à la Cixin Liu

A fascinating observation by Cixin Liu in an interview in Public Books, to John Plotz and translated by Pu Wang (numbers added):

… technology precedes science. (1) Way before the rise of modern science, there were so many technologies, so many technological innovations. But today technology is deeply embedded in the development of science. Basically, in our contemporary world, science sets a glass ceiling for technology. The degree of technological development is predetermined by the advances of science. (2) … What is remarkably interesting is how technology becomes so interconnected with science. In the ancient Greek world, science develops out of logic and reason. There is no reliance on technology. The big game changer is Galileo’s method of doing experiments in order to prove a theory and then putting theory back into experimentation. After Galileo, science had to rely on technology. … Today, the frontiers of physics are totally conditioned on the developments of technology. This is unprecedented. (3)

Perhaps an archaeology or palaeontology enthusiast might have regular chances to see the word ‘technology’ used to refer to Stone Age tools, Bronze Age pots and pans, etc. but I have almost always encountered these objects only as ‘relics’ or such in the popular literature. It’s easy to forget (1) because we have become so accustomed to thinking of technology as pieces of machines with complex electrical, electronic, hydraulic, motive, etc. components. I’m unsure of the extent to which this is an expression of my own ignorance but I’m convinced that our contemporary view of and use of technology, together with the fetishisation of science and engineering education over the humanities and social sciences, also plays a hand in maintaining this ignorance.

The expression of (2) is also quite uncommon, especially in India, where the government’s overbearing preference for applied research has undermined blue-sky studies in favour of already-translated technologies with obvious commercial and developmental advantages. So when I think of ‘science and technology’ as a body of knowledge about various features of the natural universe, I immediately think of science as the long-ranging, exploratory exercise that lays the railway tracks into the future that the train of technology can later ride. Ergo, less glass ceiling and predetermination, and more springboard and liberation. Cixin’s next words offer the requisite elucidatory context: advances in particle physics are currently limited by the size of the particle collider we can build.

(3) However, he may not be able to justify his view beyond specific examples simply because, to draw from the words of a theoretical physicist from many years ago – that they “require only a pen and paper to work” – it is possible to predict the world for a much lower cost than one would incur to build and study the future.

Plotz subsequently, but thankfully briefly, loses the plot when he asks Cixin whether he thinks mathematics belongs in science, and to which Cixin provides a circuitous non-answer that somehow misses the obvious: science’s historical preeminence began when natural philosophers began to encode their observations in a build-as-you-go, yet largely self-consistent, mathematical language (my favourite instance is the invention of non-Euclidean geometry that enabled the theories of relativity). So instead of belonging within one of the two, mathematics is – among other things – better viewed as a bridge.

No country for new journalism

(Formatting issues fixed.)

TwitterNgoodThrough an oped in Nieman Lab, Ken Doctor makes a timely case for explanatory – or explainer – journalism being far from a passing fad. Across the many factors that he argues contribute to its rise and persistence in western markets, there is evidence that he believes explainer journalism’s historical basis is more relevant than its technological one, most simply by virtue of having been necessitated by traditional journalism no longer connecting the dots well enough.

Second, his argument that explainer journalism is helped by the success of digital journalism takes for granted the resources that have helped it succeed in the west and not so much in countries like India.

So these points make me wonder if explainer journalism can expect to be adopted with similar enthusiasm here – where, unsurprisingly, it is most relevant. Thinking of journalism as an “imported” enterprise in the country, differences both cultural and historical become apparent between mainstream English-language journalism and regional local-language journalism. They cater to different interests and are shaped by different forces. For example, English-language establishments cater to an audience whose news sources are worldwide, who can always switch channels or newspapers and not be worried about running out of options. For such establishments, How/Why journalism is a way to differentiate itself.

Local v. regional

On the other hand, local-language establishments cater to an audience that is not spoiled for options and that is dependent profoundly on Who/What/When/Where journalism no matter where its ‘reading diaspora’. For them, How/Why journalism is an add-on. In this sense, the localism that Ken Doctor probes in his piece has no counterpart. It is substituted with a more fragmented regionalism whose players are interested in an expanding readership over that of their own scope. In this context, let’s revisit one of his statements:

Local daily newspapers have traditionally been disproportionately in the Who/What/When/Where column, but some of that now-lost local knowledge edged its ways into How/Why stories, or at least How/Why explanations within stories. Understanding of local policy and local news players has been lost; lots of local b.s. detection has vanished almost overnight.

Because of explainer journalism’s reliance on digital and digital’s compliance with the economics of scale (especially in a market where purchasing power is low), what Doctor calls small, local players are not in a position to adopt explainer journalism as an exclusive storytelling mode. As a result of this exclusion, Doctor argues that what digital makes accessible – i.e. what is found online – often lacks the local angle. But it remains to be seen if this issue’s Indian counterpart – digital vs. the unique regional as opposed to digital vs. the small local – is even likely to be relevant. In other words, do smaller regional players see the need to take the explainer route?

Local-level journalism (not to be confused with what is practiced by local establishments) in India is bifocal. On the one hand, there are regional players who cover the Who/What/When/Where thoroughly. On the other, there are the bigger English-language mainstreamers who don’t each have enough reporters to cover a region like India thanks, of course, to its profuse fragmentation, compensating instead by covering local stories in two distinct ways:

as single-column 150-word pieces that report a minor story (Who/What/When/Where) or

as six-column 1,500-word pieces where the regional story informs a national plot (How/Why),

—as if regional connect-the-dots journalism surfaces as a result of mainstream failures to bridge an acknowledged gap between conventional and contextualizing journalism. Where academicians, scholars and other experts do what journalists should have done – rather, in fact, they help journalists do what they must do. Therefore, readers of the mainstream publications have access to How/Why journalism because, counter-intuitively, it is made available in order to repair its unavailability. This is an unavailability that many mainstreamers believe they have license to further because they think the ‘profuse fragmentation’ is an insurmountable barrier.

There’s no history

The Hindu and The Indian Express are two Indian newspapers that have carved a space for themselves by being outstanding purveyors of such How/Why journalism, and in the same vein can’t be thought of as having succumbed to the historical basis that makes the case for its revival—“Why fix something that ain’t broken?”. And the “top-drawer” publications such as The New York Times and The Washington Post that Doctor mentions that find a need to conspicuously assert this renewal are doing so on the back of the technology that they think has finally made the renewal economically feasible. And that the Times stands to be able to charge a premium for packaging Upshot and its other offerings together is not something Hindu or Express can also do now because, for the latter couple, How/Why isn’t new, hasn’t been for some time.

Therefore, whereupon the time has come in the western mainstream media to “readopt” explainer journalism, its Indian counterpart can’t claim to do that any time soon because it has neither the west’s historical nor technological bases. Our motivation has to come from elsewhere.