Way back in December of last year I wrote a post about developing an ethic for enhancing diversity in the STEM fields. In it, I focused on one dimension of the case for enhancing diversity among scientists: that post-positivism in science provides a coherent epistemology and ethic on which to ground diversity work.

Yet there remain a number of other dimensions of the case for diversity in STEM. One more pragmatic is the argument that diversity improves the ability of scientists to solve problems within their research projects. Diverse people bring their different experiences, expertise, and ways of thinking to bear on a given problem, and this can bring about novel solutions. This idea is discussed in a recent episode of the highly recommended podcast Reply All. Although the discussion focuses on the problem of diversity in Silicon Valley tech companies (where the workforce may be even more homogenous than in many STEM fields), the insights into how diversity can improve problem solving and productivity are highly relevant to academic STEM disciplines as well.

(Note, the first part half of the podcast episode is about an obscure area of the “Twitter-sphere”. It’s pretty funny and kind of interesting but not particularly relevant; skip to the second half if you’re in a hurry. Also note: ironically, the hosts of the podcast completely ignored Rosalind Franklin’s contribution when they touched on the discovery of the structure of DNA. They have a follow-up episode where they atone)

The story in the podcast made me think about the dynamics in the Almeida lab. We are a disciplinarily diverse lab, with expertise in molecular biology, genomics, plant pathology, entomology, macro-ecology, and modeling. We are also a multinational and multicultural lab. It’s easy to see how diverse disciplines can improve scientific work, particularly when addressing environmental or agricultural problems. My own work has undoubtedly benefited from working with molecular and micro-biologists in the lab. While it remains less clear how other dimensions of our identities (e.g., nationality, race, class, sexual orientation) could improve our work, it is at least a worthwhile question to pursue. And there is only one way to find out.

When US President Barack Obama nominated Sonia Sotomayor to the US Supreme Court, he did something unprecedented. While he praised her jurisprudence just like previous presidents had spoken about previous nominees, he also spoke at length about her life story, arguing that her background as a Latina growing up in working class South Bronx would be an asset to her as a Supreme Court Justice. As Thane Rosenbaum of NYU’s Forum on Law, Culture, and Society says , “judges were supposed to be robbed robots, mechanical in their application of the law, no one had ever introduced a justice and actually said that their life experience is what would make them a really great Supreme Court Justice.” This represented a monumental shift in how the Supreme Court Justices in particular and judges in general were represented to the general public.

Diversity in government institutions and work places is a big deal these days. The term is often poorly defined, and when left undefined is often understood as specifically signifying racial diversity. (Clearly, people differ in many different ways that can be relevant for the discussion at hand, but the common understanding is sufficient for the time being.) Like “sustainability”, the combination of wide usage and vague definitions has, in some contexts, eviscerated the term of much of its meaning.

In the STEM fields (science, technology, engineering, mathematics), there seems to be increasing concern about the lack of diversity. To be sure, the low representation of scientists of color in many scientific societies is severe. For instance, black entomologists (African and African-American combined) make up only 2% of the membership of the Entomological Society of America¹. Likewise, lack of representation also extends to those identifying as LGBTQ, those in lower economic classes, and women. In response to this lack of representation in the ranks of scientific societies, numerous programs have been developed to enhance diversity, such as SEEDS from the Ecological Society of America.

At the same time, I feel like there is also a general malaise—or even derision—around diversity programs in the sciences. The zeitgeist goes something like: It’s required that we be able to write or talk about enhancing diversity, but one shouldn’t take it too seriously. Or alternatively, scientists seem to genuinely care about diversity but aren’t really sure why.

I argue that we should take the lack of diversity in the sciences seriously. I also argue that we need to think more deeply about why we should do so. While very much a work in progress, below is my attempt to begin developing more robust philosophy for enhancing diversity in the sciences.

Historically, science has operated from a positivist philosophy; that is, a view that a single set of coherent laws define reality and that these laws can be discovered through empirical hypothesis testing. Scientists then can and should perform their work “objectively and value-free, [because] the laws or generalizations exist independently of social and historical context”². However, decades of sociological and historical work on the practice of science has exposed the socially and historically contingent nature of the work that scientists perform. Such views have been bolstered within the environmental sciences particularly as more scholars recognize the inadequacy of science alone to resolve environmental problems (e.g., Ulrich Beck’s risk society theory)³.

As positivism has waned, post-positivist science has gained more ground. Post-positivism, according to Frank Fischer, sees “empirical inquiry—from observation and hypothesis formation through data collection and modes of explanation—to be influenced or shaped by the theoretical assumptions of sociocultural practices in which they are manifested”². Under post-positivism, either the universe is not governed by immutable laws or, even if they such laws exist, we are unable to build a coherent theory that reflects those laws because we can’t escape the social, cultural, and political worlds through which we move. Importantly, none of this is to imply that science should not be taken seriously. But rather than regarding “facts” as they are conventionally understood under positivism, we should take the position that, “objective fact is in effect the decision of a particular community of inquirers and the theoretical presuppositions to which they subscribe”².

If scientists are trapped in their social, cultural, and political worlds, then who gets to be part of our special “community of inquirers” matters a great deal more. Post-positivism suggests that science could be different if practiced by different groups of people; President Obama in his nomination of Sotomayor would seem to go even farther—perhaps science would be not just different but better. And who would dare call the US President a liar?

It’s unclear to me if a more diverse scientific community would produce facts and theories closer to the immutable laws of the universe. But that’s not the only way that science could be “better”. Many of us in entomology, plant pathology, and ecology are at least partly motivated to produce socially relevant knowledge, to make the world just a little better. In this case, scientific knowledge that is produced by a diverse community of inquirers, that reflects their diversity, and that is more relevant to a diverse society would constitute “better” science. (Interestingly, there is also some empirical support for the idea that diversity makes science better, discussed at the new Climate Grrrl blog here.)

To accept that diversity could make scientific production of knowledge better is then to endorse a post-positivist view of scientific practice. This is a radical proposition. On the other hand, to advocate for diversity in science under a positivist philosophy represents a trivial and banal political correctness. Such advocacy wastes valuable resources and does a disservice to scientists from under-represented backgrounds.

So what does this all mean for applied entomology? What does a post-positivist plant pathology look like? As Carolyn Finney has argued, the stories that people tell about the environment differs by their social positions⁴. Without a diverse set of scholars, we run the risk of missing important narratives about the impacts that environmental or agricultural changes have on different groups of people and, in turn, how actions of different groups of people determine the environmental and agricultural impacts of the pests and diseases we study. Do African-American vineyard managers experience Pierce’s Disease differently than white managers? Do potato farmers experience Zebra Chip Disease differently from immigrant farm workers? Do these differences matter for management? More radically, do these differences matter for our biological understandings of these systems? Would plant pathologists from different backgrounds pose and answer these questions differently? I don’t know, but I think it’s worth finding out.

1. Memoirs of Black Entomologists: Reflections on Childhood, University, and Career Experiences. (Entomological Society of America, 2015).

2. Fischer, F. Citizens, Experts, and the Environment: The politics of local knowledge. (Duke University Press, 2000).

3. Beck, U. Risk Society. 1st, (Sage Publications, 1992).

4. Finney, C. Black Faces, White Spaces: Reimagining the Relationship of African Americans to the Great Outdoors. (UNC Press Books, 2014).

For many vector-borne diseases, land managers and policy-makers —including farmers, ranchers, wildlife conservation officials, and public health officials—aim to reduce pathogen spread by suppressing vector populations. And a straight-forward method of vector control is insecticide use. Globally, the World Health Organization estimates that over 6,248 metric tons of insecticides are applied every year to control vectors of human diseases¹; this does not include vector control for diseases of plants, livestock, and wildlife. The simple hope here is that lower vector populations will reduce the spread of disease. It’s an intuitive and compelling idea—fewer mosquitoes cause less malaria. But how good are insecticides at reducing disease? Both theory and empirical evidence suggest that the record for insecticides are mixed.

15 seconds of fame for R0. Kate Winslet breaks it down in the movie Contagion.

The reasons for this mixed success are (perhaps unsurprisingly) complicated. First, epidemiological theory—in the form of mathematical models that simulate disease spread—predicts that the relationship between vector population size and disease spread is not linear. In such models, we can estimate the rate of disease spread by the term R­₀ . Called the pathogen reproduction number, the value of R₀ represents the number of new infections that the average infected host will cause: if R₀ is greater than 1, then each infected host will infect 1 or more new hosts and the pathogen will spread very quickly (exponentially). But the pathogen will not spread when R₀ is less than 1, because each infected host fails to infect more than 1 new host. In these cases, the pathogen will die out. If we plot R₀ vs. vector population size in a standard epidemic model, we see that, while R₀ gets smaller as vector abundance decreases, the vector population has to become very small before the value of R₀ goes below the threshold of 1 and the epidemic dies out. So theory predicts that it can be difficult to stop a disease purely by reducing the abundance of vectors, whether by insecticide or other means.

The relationship between vector abundance and R_­0­ is non-linear. As a result, vector populations must (in theory) be very small before R­₀ < 1 (dotted line) and the disease does not spread.

These predictions from theory are also backed up by studies of on-going disease problems. Tom Perring and colleagues from UC Riverside reviewed the evidence on whether insecticides were effective at reducing the spread of plant viruses. They found that in 29% of cases, insecticides were ineffective at reducing pathogen spread ² . Which cases were successful and which were not depended heavily on the biological interactions between the insect vector and its associated virus—insecticides tended to be successful when insects had to feed on a plant for a long time to acquire the virus and to then infect a new plant (so called “persistently transmitted” viruses), whereas it was hard to control plant viruses that could be picked up and transmitted by insects quickly (“non-persistent” viruses). What’s more, insecticides aren’t always effective at even suppressing vector populations. The mosquito species that transmit the Dengue fever virus are much more difficult to manage with insecticides alone ³ . Rather, an integrated approach is best—incorporating multiple strategies such as removal of stagnant water and distribution of bed nets, in addition to insecticides.

In the late 1990s, new outbreaks of Pierce’s disease devastated many grape vineyards in southern California. These new outbreaks were caused by a new vector, the glassy-winged sharpshooter (Homalodisca vitripennis). In response, state agriculture officials implemented an area-wide management program aimed at reducing populations of the vector, primarily through coordinated insecticide applications in citrus trees—one of the vector’s favorite host plants⁴. This program has successfully reduced populations of vectors as well as spread of Pierce’s disease in vineyards.

At the same time, vineyard managers have also been spraying insecticides in their vineyards. A recent analysis led by Matt Daugherty at UC Riverside (and with help from Sarah O’Neill, Frank Byrne, and myself), showed that these within-vineyard insecticide applications had little benefit for reducing disease ⁵. We think this is because most new infections of grape plants occur when glassy-winged sharpshooters disperse from citrus groves to nearby vineyards. Thus the window of opportunity for managers to kill the insects once they’ve arrived in the vineyard, but before they’ve infected a grape plant, is quite narrow.

Clearly, insecticide use for disease management is, in some cases, grossly ineffective. Importantly, any insecticide use should always be balanced with the cost of treatment, the damage they can do to beneficial insects, and the risks to human health. For example, DDT use against malaria mosquitoes remains widespread throughout much of Africa, and it’s unclear if the benefits of DDT use for public health outweigh the harms^1,6. (On a tangential note, I highly recommend this blog post on DDT, GM foods, and the scientific difficulty in determining if something is safe or harmful.) The good news is that we can understand which vector-borne diseases are likely to be effectively managed with insecticides and which may require alternative approaches by understanding the ecology of the system—the particular interactions of a pathogen, its vector, and the host.

1. van den Berg, H. et al. Global trends in the use of insecticides to control vector-borne diseases. Environ. Health Perspect. 120, 577–582 (2012).

2. Perring, T. M., Gruenhagen, N. M. & Farrar, C. A. Management of plant viral diseases through chemical control of insect vectors. Annual Review of Entomology 44, 457–481 (1999).

3. Erlanger, T. E., Keiser, J. & Utzinger, J. Effect of dengue vector control interventions on entomological parameters in developing countries: a systematic review and meta-analysis. Medical and Veterinary Entomology 22, 203–221 (2008).

4. Toscano, N. C., Gispert, C., Snyder, J. & Mulherin, R. Riverside County glassy-winged sharpshooter area-wide management program in the Coachella and Temecula Valleys. in Proceedings of the Pierce’s Disease Research Symposium 7–10 (2004).

5. Daugherty, M. P., O’Neill, S., Byrne, F. & Zeilinger, A. Is vector control sufficient to limit pathogen spread in vineyards? Environmental Entomology (2015).

6. Curtis, C. F. & Lines, J. D. Should DDT be Banned by International Treaty? Parasitology Today 16, 119–121 (2000).