Three Metaphors to Aid Interdisciplinary Dialogue in Public Health

Abstract

Within this journal, authors have recently called for or discussed the benefits of interdisciplinary collaboration. However, in practice such collaborations are extremely challenging, and little guidance is available to support researchers’ efforts to communicate with colleagues from other disciplines.

This article presents three metaphors from the sociology of scientific knowledge that can inform and support consideration and discussion of disciplinary issues. Disciplinary training acts as a “flashlight,” highlighting certain features of reality and leaving others in shadow. Our disciplinary sense of normal science is the metaphorical “box” into which we hope nature will fit, determining the manner in which we advance the frontier by recognizing the familiar in the unfamiliar. Finally, scientific training is a “lens” through which the world is perceived and understood.

In interdisciplinary and some multidisciplinary contexts, researchers are encouraged to (1) identify the set of fundamental concepts underpinning their approach to public health, (2) discuss methodological choices in terms that do not depend on familiarity with a common tradition of research excellence, and (3) maintain awareness that colleagues from other fields potentially hold different understandings of key public health concepts.

Public health is a multidisciplinary field, and modern public health problems increasingly require cooperation between researchers from different disciplines. In practice however, interdisciplinary collaboration is extremely challenging and some collaborative projects fail.¹ As public health research and practice become more specialized,² the importance of integration and engagement across disciplinary lines increases. Additionally, the emergence and popularity of multidisciplinary “double-degree” programs suggests a felt need for researchers who can confidently and effectively cross-disciplinary boundaries. It was during the completion of a combined undergraduate degree in biomedical science and economics that my own sense of the importance of disciplinary training (and the difficulties reconciling disciplinary ideas about health) emerged.

In 2017, within this journal, authors called for or alluded to the benefits of interdisciplinary collaboration.^3–6 But peer-reviewed strategies for overcoming the challenges of collaborative research (which include “discipline-based differences in values, terminology, methods, and work styles”¹) are limited. Improved awareness about the influence of disciplinary training in the design, conduct, and interpretation of public health research could enhance capacity for (and improve) sorely needed interdisciplinary work. However, to date this is a neglected area of empirical inquiry. Communication is understood to be key,⁷ but little detailed guidance is available to support researchers’ efforts to communicate with colleagues from other disciplines.

A relevant literature from the sociology of scientific knowledge (SSK) explores the ways scientific specialization shapes academic practice and output. Major strands of SSK include the Strong Programme⁸ (which considers “true” and “false” scientific statements of equal sociological interest), the application of discourse analysis to science,⁹ and reflexivity.¹⁰ A separate, related area is science and technology studies, which includes actor-network theory.¹¹ Conclusions drawn within these literatures could provide practical guidance for researchers engaging with other scientific specialties. As SSK draws on wider sociological and philosophical writings, some concepts and terms may not be familiar or immediately accessible to all researchers, and the explanation of these ideas via metaphor may be helpful. The use of metaphor and analogy within science is the subject of an established literature, and metaphor seems to be a natural format for communication of scientific ideas.¹² The aim of this article is therefore to outline practical strategies for considering and discussing disciplinary issues via the presentation of 3 metaphors selected from SSK, including illustration via examples from public health.

The three metaphors below—likening scientific specialization to a flashlight, a box, and a lens—were selected for their relevance to the activities of individual scientists, the range of sociological concepts they illustrate, and the extent to which they demonstrate not only what disciplinary training is, but what it does. This last point is crucial, because by considering how disciplinary training functions in practice, strategies for overcoming disciplinary boundaries emerge. The presentation of specific metaphors unavoidably emphasizes or conceals particular features of the research process; however, this perhaps further demonstrates how alternative ways of thinking can lead us to different ideas. The three metaphors that follow combine to support the broad conclusion of much SSK literature, that researchers from different disciplines inhabit different intellectual, cultural, and professional realities.^13,14 These metaphors can be employed in the meeting room, lunch room, or the classroom to shape productive discussion between disciplines, in the hope that miscommunication or misunderstanding might be anticipated or mitigated.

DISCIPLINARY TRAINING AS A FLASHLIGHT

Research is often presented in publications as a straightforward march from ignorance toward new knowledge, obscuring the messy reality. Bachelard¹⁵ challenged the presentation of science as a linear process, describing scientific inquiry as a three-dimensional struggle through concepts clustered together. Bachelard labeled this group of concepts a discipline’s “problematic,” the perspective from which a specialty develops its questions and the common starting point from which inquiry originates. A discipline’s dominant problematic does not spontaneously appear, but develops over time as a specialty identifies and struggles with relevant questions. Benton and Craib¹⁶ liken the problematic to the beam of a flashlight and note that while it casts helpful light on objects nearby, it leaves others in darkness.

In other words, when considering an empirical question, specialized knowledge illuminates particular features of reality. One consequence is that other features remain unseen, and unexamined. For an epidemiologist, the question “is the presence of X linked with Y?” evokes a set of connected concepts for consideration: population definitions, sampling strategies, the evidence pyramid, Bradford–Hill criteria, P values. This group of ideas—this problematic—fleshes out the statement “linked with Y” in scientifically operational terms. The consequences of beginning with this particular array of ideas are that investigation is guided toward deciding to measure, to measure certain attributes, to analyze measurements in a particular way, and to express conclusions in particular terms. This is the epidemiological flashlight at work, and the same phenomenon may look quite different when lit from a different disciplinary perspective.

When addressing the question “does income influence cardiovascular disease?” a colleague from another discipline may not take the phrase “influence cardiovascular disease” to mean “alter the statistical probability of mortality from cardiovascular disease within an appropriately defined population,” as an epidemiologist perhaps might. This phrase may conjure a different set of concepts, which in turn point to a different set of practical considerations. A qualitative sociologist might naturally focus on the lived experience of deprivation and hypothesize about the importance of agency, power, and class. If the problematic is the conceptual launchpad for inquiry, and researchers from different disciplines or specialties approach a project with different problematics, genuine collaboration will be difficult unless time is devoted to the consideration and articulation of the direction(s) from which participants approach research about health. In an interdisciplinary context this is time well spent, since our training, acting as a flashlight, inevitably leads us to focus on particular aspects of the phenomena we study.

DISCIPLINARY TRAINING AS A BOX

In addition to brightening certain aspects of phenomena under investigation, disciplinary training influences which phenomena are considered worthy of study in ways akin to the proverb “when you’ve got a hammer everything looks like a nail.” A different metaphor employed by Thomas Kuhn provides a detailed account of what swinging one’s disciplinary “hammer” involves, and how we determine when and where to “swing.”

Academic training familiarizes students with a pattern of investigative practice.^17,18 As a result, we tend to search for opportunities to apply the problem-solving toolkit our training provides. Kuhn proposed¹⁷ that scientific communities are identifiable via a shared “disciplinary matrix,” a group of elements including accepted laws, values, and symbolic representations. The most important element is a common set of puzzle-solving procedures that are employed as models or pedagogic examples. In epidemiology, such procedures include the randomized trial, cross-sectional survey and cohort study. This set of procedures originates from each discipline’s tradition of exemplary past achievement or paradigm. Epidemiological work generally aims to compare a health outcome across two or more populations so that something can be learned about the outcome’s distribution or determinants. This approach to empirical work did not occur to me de novo. I absorbed it during my training via familiarization with landmark studies (or materials demonstrating methods pioneered by studies) such as Doll and Hill,¹⁹ The Framingham Study,²⁰ and the Medical Research Council Streptomycinin trial.²¹ These studies represent exemplars within risk-factor epidemiology²² and form that discipline’s paradigm.¹⁷ Kuhn’s notion of the paradigm is conceptually much larger than a “box” into which nature can be bundled, and a full treatment of the concept is outside the scope of this article. But Kuhn himself used this metaphor to illustrate how the paradigm is utilized during what he called “normal science.”¹⁷

Kuhn noted that most science is not revolutionary, eureka-type moments. More typical is normal science the hunt for puzzles representing opportunities to apply the paradigm—methods absorbed during scientific education—in the image of previously influential studies. Undergraduates and professors alike demonstrate aptitude by showing that the novel can be viewed as analogous to the known, “the familiar in unfamiliar uniform.”¹⁸ For example, the question “how much cardiovascular disease risk is genetic?” was once unanswerable using traditional epidemiological methods. Today, genetic epidemiologists quantify the risk of disease associated with genetic variation in large-scale studies.²³ This is more than purely technological advancement; the demonstration that genetic risks could be studied in the same manner as other risks (that the genome-wide association study is the familiar case-control design in the unfamiliar uniform of data at genetic scale) expanded the domain of normal epidemiological science.

If normal science is the search for opportunities to deploy paradigmatic methods, and most science is normal science, research might “seem an attempt to force nature into the preformed and relatively inflexible box that paradigm supplies.”¹⁷ This is not to suggest a “cookie cutter” approach to public health, because each study presents idiosyncrasies and analytical challenges. Furthermore, normal science is not easy. Detecting the familiar within the unfamiliar can be extremely difficult, requiring lateral thinking and intellectual flexibility. But the goal is nevertheless straightforward: to attack a new problem by identifying similarity with a previously solved puzzle or successful study.

To demonstrate and clarify this process, sociologist Barry Barnes¹⁸ employs the example of a simple problem from geometry, with which many readers will be familiar (Figure 1).

FIGURE 1—

The Paradigm, a Known Problem Solution, Acts as a Resource in Devising a Solution to a New, Unseen Problem

Source. Simplified from Barnes,¹⁸ Figure 3.2.

The new problem “calculate the area of a rhombus” has multiple solutions, including via subdivision into triangles. Prior training (the paradigm) does not instruct how to solve the problem—does not instruct where to draw lines—but assures that if the rhombus can be reduced to triangles, the problem is solvable.

Similarly within public health, epidemiological training does not help develop assays for measuring genetic variability, but promises that if reliable measurements are obtained, and individuals classified reliably as “cases” and “controls,” normal epidemiological science can proceed. Likewise, when studying a rapidly spreading outbreak of food poisoning, epidemiological training does not direct an investigator precisely where to collect data. But if a case definition can be established, and a database of foods consumed by enough individuals assembled, previous work²⁴ promises that understanding will advance.

Our scientific training establishes and develops our sense of what is known and familiar. Generally speaking, to advance the frontier, new scientific terrain must be connected with established and accepted problem formats within the discipline. Economists have demonstrated great capacity for this kind of disciplinary expansion, by outlining ways in which, for example, the distribution of health can be understood as a variation of the core economic problem of utility maximization.²⁵ By producing such demonstrations, previously uncharted terrain becomes a variation on normal science.

In interdisciplinary contexts it is important to reflect upon what kind of familiar we search for, to ask what our own normal science looks like and why we value its features. Researchers from different disciplines stand on the shoulders of different traditions of excellence—different paradigms—and are searching for a different kind of familiar, fitting nature into a different kind of box. The solution is not to abandon disciplinary understandings of “good science.” However, frequently missing from interdisciplinary dialogue is an articulation of the provenance of particular ideas and methods, the identification of the research tradition to which a particular method belongs, and the merits of linking the present project to that tradition. Truly interdisciplinary work annexes one piece of intellectual territory to multiple disciplinary domains; and this cannot be meaningfully achieved when methodological choices are discussed in terms which assume that researchers share a single, common paradigm.

DISCIPLINARY TRAINING AS A LENS

In the previous section, researchers were presented as making different choices, guided by the norms of their discipline. A stronger claim is that disciplinary training is more than simply a resource for decision-making, but a means by which perception of the world is organized.^14,17,18 From this perspective, researchers from different backgrounds differ in ways more fundamental than the contents of their metaphorical toolbox; they see the world differently, as if through a different lens. The support for the claim that we experience a different reality from interdisciplinary colleagues is an awareness that disciplinary training provides the raw materials with which we construct our research findings, and give them meaning.

Barnes¹⁸ provides another illuminating general example. One cannot know or teach what a “swan” is without existing knowledge about what “plumage” is, what “white” is, and what “bird” is. A conversation about whether a particular object is “a swan” can only proceed between individuals who attribute common meanings to these other terms. In just the same way, within public health one cannot know or teach what a “treatment effect” is without relying on understanding about related concepts such as “study,” “control,” and “exposure.” Likewise, a conversation between qualitative researchers about whether a set of interviews has “saturated” relies on common understandings of “validity” and the process by which qualitative data are “generated.” It is during scientific education that the meaning of such terms is acquired. These meanings are not universal; words with apparently obvious meanings may not have agreed definitions across disciplinary lines. For example, in epidemiology and biostatistics, the word “group” refers uncontroversially to a collection of individuals receiving the same intervention (e.g., “control group”). In the social sciences the existence, nature, and status of “groups” is an important theoretical debate. The idea of a “group” is laden with associated concepts and perspectives, and its empirical application is not casual. When we say that colleagues from other disciplines do not “speak our language,” what we mean is that for our colleagues, the raw ingredients of science take different forms, have different names, or possess different meanings. Underlying, elementary ideas are very rarely explained during the conduct of normal science. But in an interdisciplinary environment this is essential, because colleagues from other disciplines—viewing nature through their own lens—arrive at different understandings.

CONCLUSION

Three metaphors from SSK were presented with the aim of describing the enduring influence of disciplinary training in research conduct, including as a potential source of difficulty during interdisciplinary collaboration. Disciplinary training acts as a flashlight, highlighting certain features of nature while leaving others in shadow. Specialization also influences which problems are deemed worthy of study and the form answers to those problems are expected to take. Disciplinary “normal science”—the box into which we hope nature will fit—determines the manner in which we advance the frontier of understanding, by recognizing the familiar in the unfamiliar. Finally, scientific training provides a lens through which nature itself is understood, by identifying and attaching meanings to elementary concepts. Public health’s status as an interdisciplinary field is frequently asserted and celebrated, but the practical challenges of reaching across disciplinary boundaries remain to be fully explored. Exposure to the SSK literature has helped me to understand why, as an undergraduate, I found studying health from multiple disciplinary perspectives so challenging. Programs intending to prepare researchers for interdisciplinary or multidisciplinary work may benefit from exposure to the ideas presented in this commentary. If nothing else, these metaphors encourage us to reflect upon the way our training shapes the choices we each make every day in our research.