“I bet @1amnerd disagrees with this” was how Kapil Subramanian’s piece in The Hindu today was pointed out to me on Twitter. Titled ‘India must look beyond neutrinos’, the piece examines if India should be a “global leader in science” and if investing in a neutrino detector is the way to do it. A few days ago, former Indian President Abdul Kalam and his advisor Srijan Pal Singh had penned a piece, also in The Hindu, about how India could do with the neutrino detector planned to be constructed in Theni, Tamil Nadu. While I wrote a piece along the lines of Kalam’s (again, in The Hindu) in March 2014, I must admit I have since become less convinced by an urgent need for the detector entirely due to administrative reasons. There are some parts of Subramanian’s piece that I disagree with nonetheless, and in fact I admit I have doubts about my commitment to whatever factions are involved in this debate. Here’s the break-down.
To raise the first question [Why must India gain leadership in science?] is to risk being accused of Luddite blasphemy.
This tag about “leadership in science” must be dropped from the INO debates. It is corrupting how we are seeing this problem.
How can you even question the importance of science we’ll be asked; if pressed, statistics and rankings of the poor state of Indian science will be quoted. We’ll be told that scientific research will lead to economic growth; comparisons with the West and China will be drawn. The odd spin-off story about the National Aeronautics and Space Administration (NASA) or the Indian Space Research Organisation will be quoted to demonstrate how Big Science changes lives and impacts the economy. Dr. Kalam and Mr. Singh promise applications in non-proliferation and counter terrorism, mineral and oil exploration, as well as in earthquake detection. But there has been a long history of the impact of spin-offs being exaggerated; an article in the journal of the Federation of American Scientists (a body whose board of sponsors included over 60 Nobel laureates) calculated that NASA produced only $5 million of spin-offs for $65 billion invested over eight years.
This is wrong. The document in question says $55 billion was invested between 1978 and 1986 and the return via spin-offs was $5 billion, not $5 million. Second, the document itself states that as long as it considered only the R&D spending between 1978 and 1986, the ROI was 4x ($10 billion for $2.5 billion), but when it considered the total expenditure, the ROI dropped to 0.1x ($5 billion for $55 billion). Here, government ROI should be calculated differently when compared to ROI on private investments because why would anyone consider overall expenditure that includes capital expenditure, operational expenses, legal fees and HR? Even as it is impossible to have an R&D facility without those expenses, NASA doesn’t have a product to sell either.
Update: The Hindu has since corrected the figure from $5 million to $5 billion.
If such is the low return from projects which involve high levels of engineering design, can spin-offs form a plausible rationale for what is largely a pure science project? The patchy record of Indian Big Science in delivering on core promises (let alone spin-offs) make it difficult to accept that INO will deliver any significant real-world utility despite claims. It was not for nothing that the highly regarded Science magazine termed the project “India’s costly neutrino gamble”.
That sentence there in bold – that’s probably going to keep us from doing anything at all, leaving us to stick perpetually with only the things we’re good at. In fact, we’re concerned about deliverables, let’s spend a little more and build a strongly accountable system instead of calling for less spending and more efficiency. And while it wasn’t for nothing that Science magazine called it a costly gamble, it also stated, “As India’s most expensive basic science facility ever, INO will have a profound impact on the nation’s science. Its opening in 2020 would mark a homecoming for India’s particle physicists, who over the last quarter-century dispersed overseas as they waited for India to build a premier laboratory. And the INO team is laying plans to propel the facility beyond neutrinos into other areas, such as the hunt for dark matter, in which a subterranean setting is critical.”
Even if it delivers useful technology, the argument that research spurs economic growth is highly suspect. As David Edgerton has shown, contrary to popular perception, there is actually a negative correlation between national spending on R&D and national GDP growth rates with few exceptions. This correlation does not, of course, suggest that research is a drag on the economy; merely that rich countries (which tend to grow slowly) spend more on science and technology.
Rich countries spend more – but India is spending too little. Second, the book addressed UK’s research and productive capacity – India’s capacities are different. Third, David Edgerton wrote that in a book titled Warfare State: Britain, 1920-1970, addressing research and manufacturing capacities during the Second World War and the Cold War that followed. These were periods of building and then rebuilding, and were obviously skewed against heavy investments in research (apart from in disciplines relevant to defense and security). Second, Edgerton’s contention is centered on R&D spending beyond a point and its impact on economic growth because, at the time, Britain had one of the highest state expenditures on R&D in the European region yet one of the lowest growth rates. His call was to strike a balance between research and manufacturing – theory and prototyping – instead of over-researching. As he writes of Sir Solly Zuckerman, Chairman of the Central Advisory Council for Science and Technology (in 1967), in the same book,
[He] argued, implicitly but clearly enough, that the British government, and British industry, were spending too much on R&D in absolute and relative terms. It noted that ‘a high level of R&D is far from being the main key to successful innovation’, and that ‘Capital investment in new productive capacity has not … been matching our outlays in R&D’.
In India, the problem is on both ends of this pipe: insufficient and inefficient research on the one hand due to a lack of funds among various complaints and insufficient productive capacity, as well as incentive, on the other for realizing research. Finally, if anyone expects one big science experiment to contribute tremendously to India’s economic growth, then they can also expect Chennai to have snowfall in May. What must happen is that initiatives like the INO must be (conditionally) encouraged and funded before we concern ourselves with over-researching.
Thus, national investment in science and technology is more a result of growing richer as an economy than a cause of it. Investment in research is an inefficient means of economic growth in middle income countries such as India where cheaper options for economic development are plentiful. Every country gets most of its technology from R&D done by others. The East Asian Tigers, for example, benefitted from reverse engineering Western technologies before building their own research capabilities. Technologies have always been mobile in their economic impact; this is more so today when Apple’s research in California creates more jobs in China than in the United States. Most jobs in our own booming IT sector arose from technological developments in the U.S. rather than Indian invention.
Subramanian makes a good point: that poor countries can benefit from rich countries. Apple gets almost all – if not all – of its manufacturing done in China – that’s thousands of jobs created in China and, implicitly, lost in the USA. But this argument overlooks what Apple has done to California, where the technology giant pays taxes, where it creates massive investment opportunities, where it bedecks an entire valley renowned for its creative and remunerative potential. In fact, it wouldn’t be remiss to say the digital revolution that the companies of Silicon Valley were at the forefront of were largely responsible for catapulting the United States as a global superpower after the Cold War.
It may have suited Subramanian to instead have quoted the example of France trying to recreate a Silicon Valley of its own in Grenoble, and failing, illustrating how countries need to stick to doing what they’re best at at least for the moment. (First) Then again, this presupposes India will not be good at managing a Big Science experiment – and I wouldn’t dispute the skepticism much because we’re all aware how much of a bully the DAE can be. (Second) At the same time, we must also remember that we have very few institutions that do world-class work and are at the same time free from bureaucratic interventions. The first, and only, institution that comes to mind is ISRO, and it is today poised to reach for blue sky research only after having appeased the central government for over five decades. One reason for its enviable status is that it comes under the Department of Space. These two departments – Space and Atomic Energy – are more autonomous because of the histories of their establishment, and I believe that in the near future, no large-scale scientific program can come up and hope to be well-managed that’s not under the purview of these two departments.
(Third) There is also the question of initiative. My knowledge at this point is fuzzy; nonetheless: I believe the government is not going to come up with research laboratories and R&D opportunities of its own (unless the outcomes are tied to defense purposes). I would have sided with Subramanian had it been the government’s plan to come up with a $224 million neutrino detector at the end of a typically non-consultative process. But that’s not what happened – the initiative arose at the TIFR, Mumbai, and MatScience, Chennai. Even though they’re both government-funded, the idea of the INO didn’t stem from some hypothetical need to host a large experiment in India but by physicists to complement a strong theoretical research community in the country.
Is the INO the best way forward for Indian science?
One may cite better uses (sanitation, roads, schools and hospitals) for the $224 million that is to be spent on the most expensive research facility in Indian history; but that argument is unfashionable (and some may say unfair). However, even if one concedes the importance of India pursuing global leadership in scientific research, one may question if investing in the INO is the best way to do so.
Allocation of resources
Like many other countries, India has long had a skewed approach to allocating its research budget to disciplines, institutions and individual researchers; given limited resources, this has a larger negative impact in India than in the rich countries. Of the Central government’s total research spend in 2009-10, almost a third went to the Defence Research and Development Organisation, 15 per cent to the Department of Space, 14 per cent to the Department of Atomic Energy (which is now in-charge of the INO project) and 11 per cent to the Indian Council of Agricultural Research. The Department of Science, which covers most other scientific disciplines, accounted for barely 8 per cent of the Central government’s total R&D spending. Barely 4 per cent of India’s total R&D spending took place in the higher education sector which accounts for a large share of science and technology personnel in the country. Much of this meagre spending took place in elite institutes such as the IITs and IISc., leaving little for our universities where vast numbers of S&T professors and research scholars work.
Spending on Big Science has thus been at the cost of a vibrant culture of research at our universities. Given its not so insubstantial investment in research, India punches well below its weight in research output. This raises serious questions as to whether our hierarchical model of allocating resource to research has paid off.
Subramanian’s right, but argues from the angle that government spending on science will remain the same and that what’s spent should be split among all disciplines. I’m saying that spending should increase for all fields, and developments in one field should not be held back by the slow rate of development in others, that we should ensure ambitious science experiments should go forward alongside increased funding for other research. In fact, my overall dispute with Subramanian’s opinions are centered on the concession that there are two broad models of economic development involved in this debate – whether a country should only do what it can be truly competitive in, or whether it should do all it can to be self-sufficient and protect itself. I believe Kapil Subramanian’s rooting for the former idea and I, for the latter.
It may be argued that to gain leadership in science, money is best spent in supporting a wide range of research at many institutions, rather than investing an amount equivalent to nearly 16 per cent of the 2015-16 Science Ministry budget in a very expensive facility like INO designed to benefit a relatively small number of scientists working in a highly specialised and esoteric field.
We need to invest in nurturing research at the still-struggling new IITs (and IISERs) as well as increase support to the old IITs (and IISc). More generally, we need to allocate public resources for research more fairly (though perhaps not entirely equitably) to the specialised bodies and educational institutions, including the universities. Besides raising the overall quality and quantity of our research output, this will allow students to experience being taught by leaders in their discipline who would not only inspire the young to pursue a career in research, but also encourage the small but growing trend of the best and the brightest staying back in India for their doctorate rather than migrating overseas.
Unquestionably true. We need to increase funding for the IITs, IISERs, and the wealth of other centrally funded institutions in our midst, as well as pay our researchers and technicians more. However, what Subramanian’s piece overlooks is that particle physics research, definitely one esoteric discipline of scientific research in that its contribution to our daily lives is nowhere as immediate as that of genetics or chemical engineering, in the country has managed to become somewhat more efficient, more organized and more collaborative than many other disciplines sharing its complexity. If managed well, the INO project can lead by example. The Science Ministry may have been screwing with its funding priorities since 1991 but that doesn’t mean all that’s come of it has been misguided.
Finally, like I wrote in the beginning: my support for the INO was once at its peak, then declined, and now stagnates at a plateau. If you’re interested: I’m meeting some physicists who are working on the INO on Monday (June 29), and will try to get them to open up – on the demands made in Subramanian’s piece, on the legal issues surrounding the project, and they themselves have to say about government support.
(Many thanks to Anuj Srivas for helping bounce around ideas.)