A new take on the research funding debate
J. R. Saylor, J. D. Summers, G. M. Mocko
The relative importance of basic and applied research is the subject of fierce debate in academic circles. The intensity of this debate is, while understandable, somewhat peculiar since researchers of all types tend to agree that science and engineering research ultimately should, and does, benefit society. This being the case, it would seem that the debate should not be about basic research versus applied research, but rather about how, and how fast, the knowledge obtained in basic research should be, and can be, transitioned toward applied research and the development of new technologies and products that benefit society. In other words, the important question is not how much funding basic research receives, but rather "how quickly should we expect these basic research results to be transitioned into products?"; "who should do the transitioning?"; and "what is the most effective way to make this transition?"
Efforts by funding agencies to improve the effectiveness and speed of the transition from basic research to products inevitably take the form of highly structured funding programs designed to connect a basic researcher with an entity (usually a company) interested in taking that one, clearly-defined step needed to move the research into the realm of a product. An example of this is the NSF Small Business Technology Transfer Program which is focused on bridging "…the gap between performance of basic science and commercialization of resulting innovations." This is a laudatory goal, but these programs only provide funding when it is relatively obvious how the basic research will result in a product. This point is typically reached long after the basic research was conducted – viz too long, in our opinion. Such programs also close off avenues of exploration that arise during the conduct of the research/development, but are unrelated to the proposed work. This is a significant limitation, since curiosity is a very significant motivation for the basic researcher (indeed, NSF often refers to basic research as "curiosity-driven research").
We think there is much room for improvement in the speed and effectiveness with which basic research results are transitioned into products. Herein we propose an alternative funding model. The proposed model involves two PIs (or groups of PIs), a basic researcher(s) and an applied researcher(s), in an academic setting. The goal of the proposed program is to: (1) minimize the time between attaining a new basic research result and the development of a product, while (2) expanding the degree of intellectual freedom for both the basic and applied researcher, thereby (3) turning the academic research enterprise into a product development incubator. The hallmark of the proposed program, and we expect the controversial aspect of it, is the absence of a commitment by the PIs to a specific research topic, or to a specific product to be developed, as long as both fall somewhere within the funding agency’s portfolio. The PIs would be expected to commit only to the development of new basic knowledge (by the basic researcher), and the development of a technology based on that research (by the applied researcher). This would be done in a co-evolutionary fashion, whereby early results from the basic research are evaluated by the applied researcher and together, both researchers modify (if appropriate) the direction of the research so that a product actually results. A critically important aspect of the proposal is a detailed, well-researched, and thoughtful means for interaction between the applied and basic researcher. Another critically important aspect is an annual workshop/review, wherein funding is withdrawn if satisfactory collaboration between the PIs is not explicitly demonstrated.
Basic and applied researchers both face challenges in justifying their work. Basic researchers must justify spending taxpayer dollars on work that a priori, cannot be guaranteed to result in a useful application. Applied researchers, on the other hand, benefit from being able to explicitly state how their research will benefit society, but in so doing tend to establish the limits of their work. That is, by showing how their research will solve a specific problem, applied researchers limit the scope of their work. In a nutshell, basic research has the potential to transform society, but with low probability and in ways that cannot be predicted, while applied research has a benefit that can be easily predicted, but is almost guaranteed not to be transformative. How can these two types of research be combined to eliminate their respective weaknesses while maintaining their strengths?
As noted above, attempts to address the aforementioned problem typically result in funding programs where researchers who do basic research are expected to partner with those who do applied research or, in some cases, to go it alone and transform their basic research results into something more applied. Sometimes this is in the form of technology transfer, where a basic researcher teams with or forms a startup company. In other cases, funding programs are developed seeking to address a specific problem, and basic researchers are encouraged to apply their knowledge to this problem. Other approaches have been explored as well. However, in our opinion, the weakness of all these programs is the specificity of the research. A specific type of basic research is to be transformed into a specific type of technology. There are milestones, goals and deliverables that are all laid out before a dollar is spent. This, we think, is a problem.
The National Science Foundation has a research portfolio that spans all aspects of science. Research on metallurgy, cancer, and arctic environments all exist within the purview of a single agency. This notwithstanding, there is little cross-fertilization within the NSF. The NSF would almost certainly disagree with this statement and would cite its long list of interdisciplinary programs. However, regardless of the interdisciplinary program, the following remains true: one has to write a proposal with specific deliverables. At first blush this makes nothing but good sense; the NSF is, and should be, committed to good stewardship of taxpayer dollars. However, it seems to us that there is room within the NSF for at least one program where the deliverables need only fall within the overall scope of the agency. The elimination of specific deliverables would eliminate a huge obstacle to true innovation. Accordingly, we propose a new program where a (self-selected) basic and applied researcher team together with the only commitment being that they (1) do research in an area that falls within the domain of the NSF; (2) that important basic research is done; and (3) that the basic research result in intellectual property that is economically viable at the end of the funding period. Such a program would increase the freedom of the basic researcher, but with the restriction that s/he works with an applied researcher in such a way that the research results in something with concrete benefits.
The obvious risk of the proposed program is that, once funded, the two researchers will each head in their own direction, doing what they want to do, but without interacting in any meaningful way. To prevent this from occurring, the proposed program would require attendance of both PIs at an annual workshop run by the program manager. The PIs would be required to make a presentation that focuses on the characteristics of their interaction. Their research would also be presented, but the emphasis of the presentation would be on how they interacted. For example, questions to be addressed would be, "how is each PI’s research informing the other?"; and "how has the direction of the research evolved during the funding period?" The presentations would serve to both provide the program management panel with information needed to determine whether to continue funding, as well as to provide lessons-learned for the other attendees of the workshop.
The proposed program would be of three years duration and would include the usual final report. However, a potential fourth year of funding would be included for the top research groups (say the top 25%) who demonstrated exceptional promise at the annual workshops. These groups would be given the fourth year of funding for purposes of patenting and further developing the technology.
Like all funding programs, the one proposed above is imperfect (and incomplete, as many details would need to be added). The aforementioned checks notwithstanding, it is still possible that the PIs will not work together. It is likely that the program itself will have to evolve to develop methods for effectively ensuring that those groups who are not collaborating have their funding eliminated. Other problems may arise. In spite of these risks, we feel that the proposed program, or something similar to it, is worth an attempt. If innovation and technologies are what we hope to obtain from federal dollars, then such a program should be explored.
John R. Saylor, Ph.D.
Department of Mechanical Engineering
Clemson, SC 29634-0921
Joshua D. Summers, Ph.D.
Department of Mechanical Engineering
Gregory M. Mocko, Ph.D.
Department of Mechanical Engineering
These contributions have not been peer-refereed. They represent solely the view(s) of the author(s) and not necessarily the view of APS.