Priyanka Pulla has a new blog, and in her first post, she writes about her report in Mint on nitrosamine contamination in some Indian drugs and answers two follow-up questions she received from readers. One of these answers contains the following portion, describing what exactly a drug-maker needs to do to ensure a drug in question doesn’t have any dimethylnitrosamine (also shortened as NDMA).
To do this, it must invest in expensive apparatuses – maybe a liquid chromatography-high resolution mass spectroscopy system (LC-HRMS) that costs up to Rs 2 crore. Then it must find a skilled professional to operate it, and that skilled professional must develop an appropriate method to test for NDMA in the drug. This is because a pharma company cannot directly copy publicly available testing methods: each drug and formulation requires a custom test.
This is such a good example of how lack of investment in R&D and of attention to problems in higher education can affect our research output. Many articles (including some of my own) allude to such connections but we don’t often come across such clear examples that explain how, say, not being able to access a chromatograph or mass-spectrometer in the classroom or not specifically training students to devise new solutions to new problems leads directly to low-quality drugs that endanger the lives of Indians. Priyanka’s article as well as her blog post touch on the fact that investing in an LC-HRMS device plus a suitable person to operate it will increase the price of the drug, which would be anathema in a highly price-sensitive market like India – so drug companies cut corners. But simply harping on keeping costs low doesn’t make sense when, as the nitrosamine story alone shows, there are at least two more things we can do but aren’t: a) encouraging local innovation on laboratory devices towards reducing their costs; and b) extending state subsidies to manufacturers and their quality control practices, in a way that will improve access to and affordability of high-quality drugs.
“It would not be unusual for finger-stick testing to be met with skepticism,” says a spokesman for Theranos. “Patents from that period explain Elizabeth’s ideas and were foundational for the company’s current technologies.”
Vanity Fair received this statement from Theranos, the company entrepreneur Elizabeth Holmes founded claiming to revolutionise healthcare but ended up being sued by investors, employees and patients for fraud, in response to a query from the magazine presumably asking about how/why Holmes thought her idea would work despite many medical experts telling her it wouldn’t. The idea in question: to use just a pinprick of blood from each patient to check for more than 200 conditions/diseases/etc. using a portable machine. In effect, Holmes, and Theranos, were attempting to shrink the blood-testing process, make it cheaper and more automated. It would have revolutionised healthcare if it weren’t for two things: the machine didn’t work, and Holmes/Theranos raised capital and made promises to investors, patients and US government institutions to the effect that it did. Holmes founded the company in 2003, reached great (Silicon-Valley-esque) heights around 2014-2015, and was dissolved in September 2018. Holmes’s trial began on August 31, 2021, earlier this week. Her colleague Ramesh Balwani is also to stand trial, and that’s expected to begin early next year. In case you’d like to catch up too, I recommend watching the HBO documentary about Holmes and Theranos, The Inventor, and reading articles by John Carreyrou (Wall Street Journal) and Nick Bilton (Vanity Fair) published between 2015 and 2018.
Towards the end, The Inventor dwells for a bit on Holmes’s state of mind: at a time when Theranos was besieged by allegations of fraud, conspiracy and knowingly subjecting its customers (technically, patients) to dysfunctional medical tests that endangered their lives, and when nobody believed its blood-testing machine, called ‘Edison’, could ever work as promised, Holmes carried on as if nothing was wrong and, in fact, according to people still at Theranos at the time, she exuded hope and confidence that the company was on the verge of a turnaround. She was clearly swindling people – diverting the money they’d invested and paid into supporting a lavish lifestyle – but seemed to believe she wasn’t. The Inventor offers (only) one explanation, that Holmes was a zealot: “a person who is fanatical and uncompromising in pursuit of their … ideals”. Without knowing more about what went on inside Theranos, especially since it’s downfall began, it’s hard to dispute this characterisation. (However, CNBCreported on August 28: “In a bombshell revelation just days before her criminal fraud trial, defense attorneys for Elizabeth Holmes claim she’s suffered a ‘decade-long campaign of psychological abuse’ from her former boyfriend and business partner Ramesh Balwani.”)
Until the end of The Inventor, and the stories by Carreyrou and Bilton, Holmes holds her ground that Theranos’s revolutionisation of healthcare is only a day away. If we assumed for a moment that Holmes really didn’t believe, in any corner of her mind, that she’d knowingly cheated people and had known that ‘Edison’ and Theranos were both part of one big sham, we’re confronted with some discomfiting questions about how we define our successes. Did Theranos conflate scepticism with impossibility – i.e. “this can’t work because the laws of nature don’t allow it” versus “this can’t work because it is disruptive”?
The Theranos story is about many things — one of them is that it highlights a highway Silicon Valley has built to its arbitrarily defined form of success that starts from one of the same points from which many success stories in the rest of the world, the real world but especially the world of scientific research, begin: “I wonder why that doesn’t work”. So it’s easy to get confused – as many journalists, investors and Holmes’s fellow entrepreneurs did – and to believe that you’re taking one highway when you may just be starting on the other. And I wish I could say the rest of the real-world highway has some checks and balances to kill bad ideas, and these the Silicon Valley highway lacks. Problems in scientific publishing, including and leading up to the replication crisis across subjects, would prove me wrong; in fact, these parallels are quite important, if only for us to reflect on why reputation-based measures of success exist. One of my favourite examples in history is that of Dan Shechtman, described here. A common example from India would be any institute that attempts to evaluate scientists’ application for promotion based on the journals in which they’ve published their papers, instead of the papers’ contents. A common and more global example: ‘prestige’ journals’ historic preference for papers with sensational results (over all papers with reliable results). And a more recent example: the Australian Resarch Council’s announcement last week that it wouldn’t consider preprint papers towards scientists’ applications for many fellowships it funds.
According to one of Bilton’s articles: “On the Friday morning that they gathered in the war room, Holmes and her team of advisers had believed that there would be one negative story from the [Wall Street Journal], and that Holmes would be able to squash the controversy. Then it would be back to business as usual, telling her flawlessly curated story to investors, to the media, and now to patients who used her technology” (emphasis added). Such ‘curation’ had allowed Theranos to be valued at $9 billion (her stake at $4.5 billion), count Henry Kissinger as a board member, Walgreens as a partner, a prominent investment firm as an investor and Joe Biden as a supporter.
This said, there’s still one big difference between the two highways: one has a better, if still quite inchoate, understanding of failure. Failure in science comes in many forms, but I know of at least two ways in which the research enterprise often ‘moves on’. One of course is retractions — and there are more scientists today than there were in decades past who are coming on board the idea that retractions are a good thing, not something to be stigmatised. The other is an increasingly deeper understanding of research fraud, the different circumstances in which it manifests, and the steps scientists and science administrators must take to prevent them from recurring. For all its lucre, the Silicon Valley highway in Theranos’s case didn’t appear to offer Holmes the opportunity of a graceful exit, so much so that it wasn’t a highway to success so much as a railroad to zealotry. That even when your product fails, you haven’t failed until you can raise no more money, until you can keep up the appearance of being successful and have a shot at being actually successful. This is also why Carreyrou, among others, has said: “It’s going to be a wake-up call for venture capitalists and young entrepreneurs in Silicon Valley. If you go too far, if you push the envelope and hype and exaggerate to the point of lying, it becomes securities fraud.” It fails to surprise me that even ‘pushing the envelope’ – presumably a euphemism for ‘smaller’ lies – is okay and that it becomes wrong/bad only when it grabs the SEC’s attention (even when most of us outside the American billionaire class aren’t likely to forget the house of cards that was the 2008 financial disaster).
Hopefully Holmes’s trial and eventual conviction will be the moment Silicon Valley stops stigmatising failure, begins to disconnect the appearance of success from success itself, and ultimately allows companies to fail without condemning their leaders at the same time. And yes, I know how ridiculous such hope sounds.
Featured image: Elizabeth Holmes in 2013. Credit: US Department of Defence, public domain.
Thus far, the composition of claims in my pieces has followed a simple pattern, even a rule: I break down a claim into a series of reasons that, when processed in serial fashion, leads up to the final thing. This has made writing pieces easy. As long as I had a claim, and deemed it to be good by whatever parameters, all I had to do was break it down into a linear chain of reasons, to be understood in sequence.
This sort of communication has an obvious disadvantage: it doesn’t lend itself to the composition of long ideas. By these I mean claims that can’t be understood by parsing one reason after another, like pulling on Ariadne’s thread. Instead, they require an Ariadne’s weave, so to speak – multiple reasons understood at once. Think of it like a particularly long series of instructions, and that to understand each instruction, the reader should be expected to have only two pieces of information: that contained in the immediately preceding instruction and that contained in the current instruction. I think I can do this well. What I’ve struggled to do, and in fact have frequently avoided (by reconfiguring what I’m trying to say), is to require the following: to compile a long series of instructions with three pieces of information at a time – that contained in the immediately preceding instruction, that contained in the present one and that contained in an arbitrary prior instruction. It’s hard for me to construct such claims or arguments fundamentally because I don’t fully understand how the reader might cognate them. (I assume here, of course, that making sense of claims made one after another is the simplest way to cognate complex ideas.)
For example, in the movie Arrival, the aliens communicate using circular logograms. Each logogram is equivalent to a full sentence written in English. But while an English sentence constructs meaning by placing down one word after the next (so that the order of words can change the overall meaning as well as that the claims made towards the end of the sentence are overemphasised, by virtue of being more recent in time, and thus memory), the aliens’ logograms make meaning by presenting all the parts of each ‘sentence’ at once, forcing their interlocutor to cognate them at once, in parallel as it were. There is an analogy in a post I published recently (‘Climate: The US needs to do more – and India needs to, too’, August 31, 2021). Here, I write that understanding claim X* requires us to consider two sub-claims at once; let’s call them P and Q*. In my post, I specify P and Q, and then I explain Q before explaining P. I did this so that the post would have flow – of the narrative moving (as) seamlessly (as possible) from one argument to the next. (You may notice that most articles in the news, especially those published by Indian mainstream English newspapers, almost always reject flow in favour of laying out P and Q, whatever they are, in that order.) Which of the two ways is better? Neither, in my view; instead, I’d prefer a visual layout that more faithfully reflects the structure of the argument:
* X = despite its rating on Climate Action Tracker, India’s climate actions are insufficient; P = climate change will affect India more than it will affect most other countries; Q = ‘they aren’t cutting emissions, so we won’t either’ is a real, if misguided, argument.
This way, the reader doesn’t have to consider P and Q one after the other but can in parallel. (Of course, literary purists may consider this to be an abdication of the writer’s duty to write in such a way that the reader isn’t at all confused about the relative weight of two sets of arguments (P and Q), but if we had to pacify purists to begin with, I wouldn’t be writing this post. (This said, I must say that I don’t like listicles for the same reason: each one of them represents a failure on the writer’s part to not give a damn about things like flow, structure, etc. – and each instance is in effect an abdication of the writer’s responsibility to write. However, on this slippery slope, both listicles and visually representing the writer’s intended location of arguments in the reader’s psyche are higher up than demanding that we must fully embrace our immutable linearity of the human condition at all times.)(Strained argument, I know.))
I’m writing all of this down because I recently composed a long idea, and noticed it as I was doing so. The long idea was for the post ‘They’re trying to build a telescope’ (published August 19, 2021) – specifically, the following portion:
And now, astronomers in China have published a paper expressing their excitement about having spotted a new location at which to mount a telescope, themselves overlooking considerations of whether the people who are already there might be okay with it. As a result they may have effectively shut one option out. This is an important factor because, as Rao has written (see excerpt below), many people seem to think that Hawaiians’ resistance to the TMT and others of its kind on the islands is fairly recent; this is not true. They expressed their opposition how they could; the rest of us didn’t pay attention.
Here, I’m talking about how astronomers didn’t allow Hawaiians to say a telescope couldn’t be erected at a particular site by framing the terms on which they commenced negotiations in a way that precluded the option of not having a telescope. Making my point here required me to draw on the conclusion of the post until that point in the narrative and, more importantly, a point Rao builds up to in the excerpt (from his article) that follows. As a result, I ended up effectively telling the reader: this is what I’m saying; Rao’s words will prove my point, but unfortunately, you’ll have to read them yourself; I won’t be able to guide you to the end; once you finish reading the excerpt from his article, I hope you will be able to see what I’m trying to say. Alternatively, if I had to represent this visually, the narrative diagram would look something like this:
That is, Q follows P, but at the same time they ought to be considered together in order to understand what follows. I don’t know about you, but psychologically, I’d find any argument presented this way to have greater potential to be misconstrued than an argument that is entirely, and straightforwardly, linear. Axiomatically, I edit any text to ensure that an idea it contains that is already likely to be misunderstood (due to certain historical connotations, say) is couched in a narrative that is linear to the extent possible; any bit of non-linearity will allow readers to order reasons the wrong way – particularly, placing effect before cause – and construe a claim that isn’t being made at all. Finally, I acknowledge that this post may seem wholly confused, in which case I apologise for wasting your time; I also didn’t conduct a literature review before I started, and am more than likely to have recreated something scholars already know, may have articulated better, and in fact may have debunked as well. If you’re aware of any such things, please let me know (by email or on Twitter, where I’m @1amnerd).
Shortly after the IPCC published the first installment of its AR6 report, The Wire Science produced a short video explaining the report’s salient points. It swiftly met with some backlash from some scientists, who were miffed that the video spoke about India reducing its carbon dioxide emissions without emphasising that the US and many European nations needed to commit to greater reductions than others.
I’m wary that repeatedly stressing that point could lead to a mindset that if the US, the UK, Germany, etc. don’t reduce their emissions, India has a free-pass to not reduce its emissions either. From a bird’s eye view, this ‘free pass’ might seem like a distant possibility considering, according to Climate Action Tracker, India is on course to do its bit to keep the world’s average surface temperature increase below 2º C over pre-industrial levels – the Paris Agreement line. However, there are two issues here that should dispel this sense of satiation.
First, climate change will affect India more than it will affect most other countries, and what India needs to do to stave off the worst of these effects is not something Climate Action Tracker or any other global monitor measures. Second, ‘they are not doing it, so we won’t either’ is a not-so-distant possibility because it has already turned up in some narratives – but especially ones concerned with getting people out of poverty.
The latter, we are told, is a carbon-intensive exercise, but we must also consider how and to whom the benefits of such development accrue, considering arguments that India should be allowed to emit some more carbon dioxide for some more time typically emerge when a hydrocarbon extraction project in Tamil Nadu, a transshipment project in Nicobar, an iron-ore mine in Goa, a railway line in Maharashtra, an oil pipeline or sand-mine in Assam, a solar-power plant in Rajasthan or a diamond-mine in Bundelkhand is at stake.
As M. Rajshekhar has written in Seminar, one big difference between the UPA I/II and the BJP I/II governments is that the former was corrupt and sought profits, while the latter is corrupt and seeks rent. Under the BJP, Adani, Reliance, Essar and a few other corporate groups have benefited inordinately to the exclusion of most others, as a result oligopolising a swathe of the country’s natural resources, including forests, mountains, water bodies and non-agricultural land. This is not sustainable development and can’t possibly lead to it either.
In this sense, Rajshekhar wrote for CarbonCopy, “the country’s inability to lift its people out of poverty shouldn’t become an unlimited pass to pump greenhouse gases into the air.” That is, if eliminating poverty is taking the form of allowing Adani, Reliance, Essar, etc. to pad their bottom lines by building roads, airports and railway tracks (often to the rejection of all ecological wisdom), then the emissions resulting from these activities don’t deserve to be excused. And considering the incumbent government has made a habit of accelerating and approving such projects, the added pressure of having to cut emissions is a good thing.
On the first count, that climate change will affect India more than most: climate policy expert Kapil Subramanian broke down the IPCC report’s predictions in three different emissions scenarios for the South Asia region, and found that “the 1º C difference in warming in South Asia between the SSP 2-4.5 and the SSP 1-2.6 scenarios is more than worth fighting for.” (This is the difference between global mean surface temperature rise of 1.8º C and 2.7º C.)
As examples, he discusses projected weather patterns – which we ought to consider as conservative ‘estimates’ – over India corresponding to the scenarios: more days of extreme heat, more flood-causing rainfall and longer summer monsoons, and extreme events that happened once a year likely happening 4-6 times a year. Climate Action Tracker or any other similar entities that take a big-picture view of India’s actions are blind to these considerations, to which India alone can, and must, respond.
So while emphasising that the US and some European countries should do more at every turn is important in some fora, we don’t have to do it at every turn all the time. Instead, we need to flip our own demands, bearing in mind that ‘cutting emissions’ – i.e. mitigation – isn’t the full picture. India needs to cut its own emissions, irrespective of how much the US, the EU, etc. are cutting, while transitioning to sustainable development (long fricking shot but we must demand it), and, on a related note, make its adaptation policies better and more just.
The US needs to do more; India needs to do more as well.
