NRF: What research should be funded?

Part I of this article (Business Standard, July 20, 2023) said the National Research Foundation was a landmark initiative. It emphasised that we must adhere fully to the two announced principles: Firstly, exclusively funding research within the higher education system, with researchers at autonomous government laboratories eligible only in partnership with an academic researcher. And, secondly, funding researchers in both private and public academic institutions.

I argued also that public funding must provide the entire Rs 50,000 crore budget without any reliance on private funding, and that effective governance demands a professional (non-government) board that would be accountable to the Cabinet. This outlined the how for the NRF to be done right, and to be truly transformative for both higher education and scientific research. But what research should be funded? That is the focus of this article.

Pursue excellence, don’t seek relevance: Official statements on the NRF cite its similarity with the US National Science Foundation (NSF).  The NSF maintains a database of over 450,000 peers and assesses proposals on their intellectual merit. The singular objective is excellence. Only peers — meaning other academics, operating in the same field — judge whether a proposal is worth funding or not. And the funding is spread widely: Last year’s NSF funding of $8.8 billion (Rs 70,000 crore) was spread across 11,000 awards, in 1,800 colleges and universities. The NSF claims this directly supported 350,000 researchers, postdocs, trainees, teachers, and students.

The most insightful observers of how technology progresses have for long explicitly warned against seeking relevance or utility in research. Economist Keith Pavitt said that distinctions between “blue-sky” (probably useless) and “strategic” (potentially useful) research should be made with great caution and treated with great scepticism. Economic historian Nathan Rosenberg, in a paper called “Uncertainty and Technological Change”, concluded that major leaps often happen when one is searching for something else. As such, attempts to select or promote research on the basis of relevance are misplaced. Instead, an effective research policy would open many windows and provide incentives to the private sector to explore the opportunities they see (however faintly) through those windows. A research portfolio should be deliberately diversified.

That is also a reading of several key discoveries.  Take the development of two great scientific discoveries: Thermodynamics, by the engineer Sadi Carnot in 1824, and bacteriology, by Louis Pasteur in 1865. Carnot was interested in a very practical problem: Why were high pressure steam engines more efficient than atmospheric engines? The science of thermodynamics resulted from his practical technological enquiry.  Pasteur was interested in an equally practical (and very French) problem: What is it that leads to better quality wine as it ferments in the bottle? 

Or consider two famous inventions where the development process was anything but planned: If the scientist Alexander Fleming had not gone on holiday for two weeks, a petri dish of bacteria would not have played self-destructing football with the mould that later became penicillin, the world’s first antibiotic. And if the 3M scientist Spencer Silver had not been searching for a stronger adhesive, he would not have found the weaker one that became Post-Its.

All these stories have two things in common. All demonstrate how the path of technical and scientific progress is often uncertain and unpredictable. And all demonstrate the potential in letting researchers pursue their own intellectual journeys. They may not reach the destination they set out to, but where they end up often makes the world a far better place than what they actually sought. 

Fund science, not Science: Scientific research delivers an enhanced understanding of nature in two ways.  Large scale scientific research, or Big Science, usually involves massive investment in equipment.  Particle physics is Big Science at its biggest: The Stanford Linear Accelerator is housed in the world’s longest building and has received decades of generous funding from the US Department of Energy; CERN, outside Geneva, is so expensive that it is funded by a dozen countries.  Big Science matters to global scientific advance; India can be part of such efforts as we get wealthier. I am not going to enter the debate of whether or not we have the capacity to do Big Science well. Let’s assume we can. We should still not do it, as one unit of Big Science would displace thousands of smaller efforts. So for now, we should be much more focused on science, with a small “s”, supporting and funding the work of hundreds of thousands of academics working across our higher education system, and a million master’s and doctoral students in a thousand public and private institutions.  That should claim all of the Rs 50,000 crore NRF funding.

Technical change is evolutionary:  But how can Pavitt be right, that we should not seek relevance, utility and application in the research we fund? It depends on our understanding of the innovation process.  Another economist of innovation, Richard Nelson, has long argued that technical change is best seen as an evolutionary process.  Evolutionary processes have two essential elements: The generation of variation and a selection mechanism that decides which variation survives. Variation is a natural part of biological processes.  Natural selection happens through the survival of the fittest. And the fitter variations get coded, over generations, into genetic structures.

For technical change, technological and scientific experiments generate alternatives. The market selects the best alternative. In “Capitalism as an engine of progress”, Nelson said that capitalism’s power is derived from innovation. Its power to innovate came from its generation of many, many alternatives. This process is inherently wasteful — much, maybe most, technical work ends up being discarded.  But it is precisely this that also makes it powerful; the market selects from many alternatives, each competing aggressively to get better than the others.  Nelson was writing at a time when the Soviet Union still existed. His argument still resonates 40 years later: Government-selected innovation sought efficiency and so delivered weakness, while market-determined innovation ended up being powerful by being wasteful. The role of the state in research, then, is to fund academics widely and well, on the basis of peer review. The state should not decide what or who to fund, or try to focus on a few big areas. Let firms look through the windows that public research opens through academic research. Let firms hire the talent that comes out of the public and private institutions where liberal public funding from the NRF results in students trained in high quality and intellectually meritorious work. And then let industry focus on its own particular variation, and compete for the products and services that people wish to buy.

ndforbes@forbesmarshall.com . The writer is co-chairman, Forbes Marshall; past president, CII; Chairman of Centre for Technology Innovation and Economic Research and Ananta Aspen Centre. His book, The Struggle and the Promise , has been published by HarperCollins