1

What Is Transformative Research?

The central purpose of all research, whether basic or applied, is to create new knowledge. Research in the domain of the geographical sciences is generally driven by a desire to generate new knowledge about that specific domain; that is, about the relationships among space, place, and the anthropogenic and non-anthropogenic features and processes of the Earth. Some research exceeds these modest aims, with impacts that extend well beyond them; other research creates new opportunities for further research, affects the process of knowledge acquisition more broadly, or changes the way other researchers in a domain think about the world and go about their business. In such instances, research is capable of transforming a field. Thus, the very concept of transformative research is self-referential: By a process of positive feedback, a specific research effort can have impacts on an area of research that are much greater in magnitude than might normally be expected.

The term transformation implies the existence of something to be transformed and suggests Thomas Kuhn’s concept of normal science: a steady state in which science proceeds by continuous, incremental accumulation of knowledge (Kuhn, 1962). Transformative research introduces a phase of what Kuhn termed revolutionary science, in which normal science is interrupted, disrupted, or transformed by new ideas, technologies, or questions. Many have argued that Kuhn’s two-phase model is overly simplistic, implying for example that large areas, or even all of science, are interrupted by these transformative events. Although like all models, Kuhn’s model does simplify a more complicated process, there is no doubt that new research directions have indeed emerged in the geographical sciences from time to time as a result of a variety of transformative stimuli. Because of its positive feedback and impact, transformative research can be regarded as inherently more valuable than conventional research.

In this chapter, the committee cites examples of funding agencies that clearly regard transformative research as something to be encouraged and funded through special programs. This chapter also explores the concept of transformative research in detail, examining definitions that have been advanced, related terms, and funding programs aimed at stimulating, encouraging, and fostering transformative research. Thus, this chapter provides the foundational context for the later sections of the report. In Chapter 2, the committee focuses on several examples and presents a series of committee findings regarding how transformative research has developed in the past. Chapter 3 examines the current climate for research within the United States, especially with respect to funding, and compares it with the situation in the past and in other countries. In Chapter 3, the committee argues that a new emphasis on transformative research may be a logical response to shrinking funding and increasing international competition. In Chapter 4, the committee makes recommendations for fostering transformative research.

DEFINITIONS

In 2007, the National Science Board (NSB) issued the report Enhancing Support of Transformative Research at the National Science Foundation (NSB, 2007). Within this report, NSB adopted the following definition of transformative research:

The language of this definition references both the results of transformative research (e.g., discoveries or creation of a new paradigm or field of science) and the inputs to such research, (e.g., the tools used). Notably, what is transformed can include current understanding, an existing concept, or an established practice. The definition is lengthy, and clearly attempts to be as inclusive as possible, using “or” no less than seven times.

In 2012, the National Science Foundation (NSF) sponsored the workshop “Transformative Research: Ethical and Societal Implications” (Frodeman and Holbrook, 2012). Chapter 3 of this report expands on this broader context for transformative research in the United States and its current importance, situating it within the major trends and developments that are impacting higher education, the research enterprise, American competitiveness, the level of funding for research, and the intensity of competition among researchers for that funding.

In the United Kingdom, the Economic and Social Research Council (ESRC), roughly equivalent to NSF’s Directorate for Social, Behavioral & Economic Sciences, settled on the following definition of transformative research: “research ideas at the frontiers of the social sciences, enabling research which challenges current thinking. We regard transformative research as that which involves, for example, pioneering theoretical and methodological innovation... novel developments of social science enquiry... [or] an element of risk.”

While the definition may seem more succinct, especially when quoted selectively as here, one should note that it has been constructed specifically for the domain of the social sciences and may need to be broadened if it is to be applied to all domains of NSF, or even that of the geographical sciences, which include physical geography.

RELATED DEFINITIONS

The theme of transformative research as inherently risky provides a link to several other closely related definitions. The National Institutes of Health (NIH) recognizes “high-risk, high-reward” (HRHR) research. In the words of NIH director Dr. Francis Collins, “High-risk research isn’t for the faint of heart. It’s for fearless researchers who envision and develop innovative projects with unconventional approaches that, if successful, may yield great leaps in our understanding of health problems and/or biological mechanisms. It takes nerve and creativity to conceive such projects—and, often, special support to bring them to fruition. And, as the name implies, there is a significant chance of failure” (Collins, 2013). The European Research Council (ERC) prefers the term frontier research, which “reflects a new understanding of basic research. On one hand, it denotes that basic research in science and technology is of critical importance to economic and social welfare. And on the other, that research at and beyond the frontiers of understanding is an intrinsically risky venture, progressing in new and the most exiting [sic] research areas and is characterised by the absence of disciplinary boundaries” (ERC, 2015).

No single definition of transformative research exists or is likely to emerge in the near future. NSB’s defini-

tion of transformative research has clearly tried to accommodate all domains of science, technology, engineering, and mathematics (STEM), and to include the many ways in which research progresses. There is a temptation to view the lack of a single, simple definition as a failure of language rather than of the concept the language is trying to capture, and to fall back on “I know it when I see it” (Lal and Wilson, 2013).

FUNDING INITIATIVES

The NSB report (NSB, 2007) discussed earlier recommended a Transformative Research Initiative within NSF that was “distinguishable by its potential impact on prevailing paradigms and by the potential to create new fields of science, to develop new technologies, and to open new frontiers” (NSB, 2007, p. v). Subsequently, NSF introduced the cross-directorate Creative Research Awards for Transformative Interdisciplinary Ventures (CREATIV) and the Integrated NSF Support Promoting Interdisciplinary Research and Education (INSPIRE) funding programs specifically aimed at providing opportunities for transformative research; several awards have since been made under both programs.

NIH has four award programs in its HRHR portfolio—the Early Independence Award, the New Innovator Award, the Pioneer Award, and the Transformative Research Award—with each drawing on the NIH Common Fund, and thus above the programs of the individual institutes and all closely associated with the director. The Transformative Research Award is described as being “created specifically to support exceptionally innovative and/or unconventional research projects that have the potential to create or overturn fundamental paradigms. These projects tend to be inherently risky and may not fare well in conventional NIH review. As compared to the other NIH Director’s Awards... the primary emphasis of the Transformative Research Awards initiative is to support research on bold, paradigm-shifting, but untested ideas” (https://commonfund.nih.gov/tra/description).

The reference to “not faring well in conventional NIH review” suggests a possible motivation for such funding programs, particularly the belief that the conventional review process is inherently conservative and favors normal science over revolutionary science. According to this view, reviewers tend to assess proposals against conventional practices, and may be unable to appreciate the potentially transformative aspects of new ideas. They may tend to react negatively to what are informally referred to as “trust me” proposals that lack the methodological detail needed to make an independent assessment of viability. And perhaps reviewers occasionally overlook the fact that research results are inherently, and by definition, unknowable.

Support for transformative research may require a novel approach to the review process. The CREATIV program at NSF, for example, allows a program manager to make decisions without external peer review on proposals valued up to $1 million, a limit more than an order of magnitude higher than NSF’s traditional practices for decisions based only on internal review and subject to the requirement that at least two directorates be involved. The ESRC’s Transformative Research Call, which is in its third annual cycle, is another example of a program that uses an unconventional mechanism for review. After an initial assessment by a panel of academic experts, “shortlisted” principal investigators (PIs) are invited to present their proposals at a “Pitch to Peers” workshop, with awards made shortly thereafter. Arguably, peers who have submitted their own transformative ideas may be better able to assess the transformative ideas of others. Other novel approaches include the “sandbox” or “sandpit,” in which the generation of novel ideas for research occurs in a workshop setting. During these workshops, ideas are proposed, discussed among the group, and then voted on, with reasonable assurance that the sponsoring agency will make an eventual award to the originator without further external review.

Another alternative to the all-or-nothing nature of traditional funding practice is what might be termed as progressive funding, in which PIs with promising ideas would first be awarded small seed grants through a

streamlined review process. If the research results prove to be promising, a subsequent proposal could be made for a second, larger phase of funding. Keeping the initial award small would ultimately reduce the overall risk to the funding agency.

On the other hand ERC’s support for its “frontier research” is obtained through its regular programs: “Support for investigator driven ‘frontier’ research can be obtained by individual researchers through the European Research Council (ERC)’s competitions for funding” (ERC, 2015).

Any revision in the traditional review process must be considered carefully, lest it lead to a more elitist, top-down pattern of funding at the expense of the broader community. To date, such revisions have involved only comparatively tiny fractions of the total funding stream, and many, such as “Pitch to Peers,” might be considered more community-based than traditional practices, by involving a larger panel. Nevertheless, funding agencies would be wise to anticipate resistance from sectors of their research communities if evidence suggests transformative research is funded at the expense of normal science.

Many agencies have developed activities designed to encourage and support transformative and HRHR research in their respective domains. Among these agencies are the U.S. Department of Defense’s Advanced Research Projects Agency, the Office of the Director of National Intelligence’s Intelligence Advanced Research Projects Activity, and the National Institute of Standards and Technology’s Technology Innovation Program. Other comparable programs exist in the national laboratories and in the private sector, as well.

PROGRAMMATIC ASSESSMENT

With significant effort being made to promote and encourage transformative research, and substantial (though small in proportion to the total funding stream) sums being awarded to projects, there is clearly interest in assessing the efficacy of such programs. According to Johnston and Hauser (2008, p. A12):

These questions bear a remarkable similarity to the three questions of the committee’s charge (see Box 1.1 later in this chapter). The question of whether “the predictors of transformative research are completely unstudied” is addressed further in Chapter 4.

Lal and Wilson (2013) report on efforts to address some of these questions through systematic, quantitative research. In work partly supported by NIH, they analyzed 35 of the awards given in the first 3 years of HRHR funding, using a variety of data sources including publications resulting from the work and biographical characteristics of the PIs. When examining potential predictive indicators of success in transformative research, they found that:

• Researchers conducting transformative research tend to have a similar number of publications as compared with similarly excellent researchers;

• Researchers conducting transformative research do not tend to be younger when compared to mainstream researchers;
• Transformative research is perceived to be risky by peers, but the degree of risk at the time of award does not seem strongly associated with impact years after award; and
• Transformative research tends to be no more interdisciplinary or collaborative than similarly excellent research.

Note, however, that Lal and Wilson’s (2013) null hypothesis was accepted for a small sample (i.e., 35 awards) and may have been rejected for a larger sample; in other words, these results, which are dominated by acceptance of the null hypothesis, may reflect Type II statistical errors. In the committee’s view, these results concerning pre-award indicators are inconclusive because of the small sample size, the broad basis of the questions asked, and the limited domain of the sampled research. The conclusions and recommendations presented in Chapter 4 balance this provisional evidence from the Lal and Wilson study with the views expressed during a workshop organized by the committee and responses from an online questionnaire.

With respect to post-award indicators, Lal and Wilson (2013) found that transformative research:

• Tends to follow more innovative research approaches as compared with similarly excellent researchers;
• Tends to have a greater impact as compared with similarly excellent researchers; and
• Garners less disagreement among peers and does not take longer to be accepted by the community as compared with similarly excellent researchers.

In the case of post-award indicators, then, Lal and Wilson (2013) were able to reject the null hypothesis, lending greater weight to their conclusions than in the case of pre-award indicators.

Thus, the results are mixed. It appears that proposals that are selected by programs designed to foster transformative research have greater impact, and are more innovative on reflection, than the work of similarly excellent researchers that is funded by more conventional programs. Yet, there is ample anecdotal evidence of the highly controversial nature and long-delayed acceptance of many transformative ideas, such as anthropogenic climate change (discussed in Chapter 2), evolution, or continental drift. Additional systematic research using larger samples and addressing problematic cases is clearly needed, particularly where ideas determined to be potentially transformative at the proposal stage turned out not to be and where transformative research was not identified as such at the proposal stage. As the committee discusses elsewhere in this report, such evidence would be useful in addressing the three questions of the study charge.

COMMITTEE’S APPROACH

The committee was asked to provide insight into how transformative research in the geographical sciences evolved in the past so that it can be encouraged in the future. The charge asks that the committee take a historic approach by reviewing how transformative research emerged in the past, what its early markers were, and how it can be nurtured in the future (see Box 1.1). The charge refers repeatedly to the term geographical sciences, so a brief note of clarification might be useful at this point. The committee recognizes the term as overlapping substantially with the discipline of geography but differing in two respects. First, not all researchers who identify with geography would be content defining their work as science; and second, the methods and principles of the geographical sciences are used and advanced in numerous disciplines, from engineering to the humanities. The

committee uses the term geographical scientist, meaning someone engaged with the geographical sciences, when it seems appropriate. Finally, the committee also uses the term geographer, while recognizing that the term is distinctly ambiguous: To be a geographer, must one have one’s final degree in geography or work in a department of geography, or is it simply a matter of self-identification? There are no simple answers to these questions, and it is therefore left to the reader to attach whatever meaning the context suggests.

The Statement of Task (see Box 1.1) instructed the committee to hold a workshop as its principal information-gathering activity. This workshop was held on August 5–6, 2014, in Irvine, California, with approximately 30 invited participants from a broad cross-section of the geographical sciences and affiliated disciplines, as well as experts in assessing research outcomes (see Appendixes B and C for the contributors and the agenda, respectively). Two keynote lectures started off the workshop, followed by a series of moderated panels on (1) society, polity, and economy; (2) methods, models, and geographic information systems; (3) environmental sciences; and (4) being transformative.

Each panelist provided a short white paper that was distributed to participants prior to the workshop. In developing these papers, the panelists were asked to describe one or two transformations that have occurred in their field of interest and to consider the following questions to whatever extent possible:

1. How did the transformative research emerge and how did it become transformative?
2. What were the early markers of the transformative research and how did it become possible to identify its transformative character?
3. What has helped nurture and bring transformative research to fruition in your field and how can it be fostered?
4. Is there past research that should have been transformative (in your estimation), but in hindsight was not?

The first three questions reflect the committee’s Statement of Task, while the fourth was intended to encourage panelists to explore the equally important question of why some promising research does not become transformative. These white papers provided a starting point for discussions, as well as a rich source of information for the committee to draw on as it wrote this report. The first three questions produced useful input to the report. The fourth question proved to be more problematic; while many participants could cite examples from personal experience, nothing emerged in the way of general and useful principles. Hence, there is clearly room for a more extensive investigation of what is surely an important and interesting question.

The committee also chose to collect information via a questionnaire that was distributed online (see Appendix D) to include the ideas of those who were unable to attend the workshop and to bolster the committee’s other information-gathering activities. The questionnaire was not designed for nor subjected to statistical analysis, but the responses were reviewed and considered by the committee in the writing of this report.

This page intentionally left blank.