India science suffers because we don’t nourish ecosystems by creating clusters of excellence. Here’s what our institutes need to do to change this.
There has been an explosion in the quality and quantity of scientific research output from India in the past decade. Despite this, the overall “heft” of scientific research from India remains microscopic. At the table of world science leaders (eg. the USA, UK, Germany, Japan, now China) we don’t have a seat. But this article isn’t about the problems in Indian science. (Indeed, there is much to be positive about.) Instead, we focus on one visible lacuna, and how it can be filled. There is a “scaling while increasing quality” problem in Indian science, and a severe lack of clusters of excellence with enough critical mass. Almost everyone in Indian science acknowledges this, but then what is the best way to scale up? And are there lessons we can learn from other successful examples elsewhere in the world?
The idea of scale (in the Indian scientific context) is abused and misunderstood. Sure, we know that there are many woefully underequipped, poorly administered universities. There is an urgent need to strengthen our university system (to tackle the far bigger problem of improving our abysmal education system). However, most of the advanced research in the country happens in smaller research institutes spread across the country, mostly supported by a slew of scientific agencies in India (the DST, DBT, ICMR, CSIR, DAE and so on), along with a few “institutes of national importance,” like the older IITs or IISc. Many are good, and some now world class. For historical reasons, as well as because governments have an implicit mandate to be equitable, these are spread across the country. So this means that any one geographical region might have a handful of researchers in a broad area of research, a handful of whom are at the highest capability, and only a handful of these might be world leaders in a field. For a folksy perspective, if Birbal was around today, and was asked by Akbar to estimate the number of competent independent research groups in say “biological sciences” in India, he would say 2000, with a few coming in and going out, and wouldn’t be very wrong. Since these individuals are distributed across the country, there is little chance that they will regularly interact, meet each other, bounce off ideas, collaborate or even compete.
Separately, the history of the scientific and industrial revolution, the growth of scientifically advanced societies, the emergence of tech hubs, and the general economics of scale all tell us that when a large group of talented individuals congregate, the whole effect is usually greater than the sum of the output of individuals. This is true even for fundamental science research. The great science hubs of the world today, like Boston, San Francisco, London, Paris, Tokyo, the whole of Israel and more, all contain multiple institutes of size and quality. This means that the scale of ideas churning in each geographical spot is massive. For perspective, say the San Francisco area might have over 2000 capable research groups in the “biological sciences”. This creates science ecosystems: great academic research, lots of breakthroughs, this becomes a magnet to attract more capable researchers, industry grows to supply this academic enterprise, a fraction of academic researchers get together to start new companies, and a whole knowledge economy grows.
So coming back to Indian science, even the government recognizes the need for clusters. We have a small group of clusters of science with a chance to scale with excellence, in Bangalore, Delhi, Pune, Hyderabad, even Kolkota. The largest of these, Bangalore and Delhi, could probably muster up ~200 capable researchers in a given broad discipline (say again “biological sciences”), across a dozen institutes in each place. This is good, because there are some effects of the cluster (as discussed above) already visible. But how can any of these go from 200 to 2000? Especially given the Indian institutional context, where we expect government largesse to build or expand all existing institutes? Is it even fair to expect the government to expend its energy in expanding these institutes, at the risk of appearing partisan? One might argue that the government has already exceeded its necessary role – it has started and generously supported these hubs, thus fulfilling its role in seeding excellence. Also, the government must continue to strengthen our school and university education systems more urgently. Perhaps part of the solution to scaling advanced scientific research in India is not to look at the government, but beyond (while continuing to benefit from government support).
Here’s a more detailed depiction of the problem: an institute has been sanctioned say 30 independent group positions. This means that the government will support these groups, along with the running costs of the institute. This also means that an institute cannot grow beyond this (as conventionally supported). But then this means that the scaling problem is irresolvable in India! Digressing just a bit, there is the story of Alexander the Great and the Gordian knot. The short of it was that there was an intricate knot tying down an ox cart in Phrygia, and legend said that whoever untied the knot would become king of Asia. Alexander, like everyone before him, tried and failed to untangle the knot. But never one to easily give up, he reasoned that it didn’t matter how the knot was loosened, as long as it came loose. So he took out his sword and sliced the knot open. Perhaps our institutional scaling problem is a metaphorical Gordion knot, and the solution isn’t to try to loosen government strings, but to find another solution. But if not the government, who?
The Way Forward
The solution might lie in three parts. The first is that the institutes need to get out of their insularity, or their perceived hierarchies of standing, and come together to leverage their strengths. This means that everything should be on the table, ranging from shared PhD programs and joint academic appointments, creating new, multi-institute centres, shared (or shared access to) facilities, and putting up a somewhat united front to external agencies (be it government or private) when it comes to bigger issues like the need to invest more for research. Of course, this means institutes have to get off their high horses, and talk a lot more to each other with some mutual respect, and an understanding of a greater common cause.
The second is that nothing stops institutes from working harder to get private philanthropy and scale in different ways. If the government cannot sanction or support a position, there is nothing stopping institutes from working with private agencies to support endowed positions. Of course, the scale of endowments should be well beyond just individual positions, and also bring with them flexibility and the chance of doing something new. This means institutes must lose their ivory tower and socialistic mindsets, and instead of behaving like entitled children, understand how to interest and engage with private philanthropy.
The third part is to work far better to leverage the support of the state (and not central) government. Institutes must work so much harder to convince the state of the value in having them. (For example, it is a no-brainer that Bangalore’s tech industries have benefited enormously from the presence of IISc, NCBS, JNCASR, RRI and a slew of other institutes.) The states spend almost nothing in their budgets supporting advanced research, but there is no reason they cannot increase it to a larger pittance. When all of this happens, institutes can creatively create new positions, clusters of excellence (sometimes even in specific areas of research), multi-institutional programs and more, attracting new researchers (with competitive resources) to set shop.
“The fault Brutus lies not in our stars, but in ourselves that we are underlings”, Shakespeare’s Caesar is quotable for almost anything, but especially here. Its high time Indian science leadership thought hard about creative ways to break its Gordian knots, and solve its scaling problem. The self-amplifying benefits that follow will be worth the effort.