I annually teach a second-year course at Lewis & Clark College called Environmental Analysis (ENVS 220). While the course is ostensibly limited to methods and research design, as one of several core ENVS courses in the major it also helps promote some key scholarly virtues we encourage our ENVS students to cultivate, and on such virtue involves recognition of many shades of green, i.e., the spectrum of ways that people make sense of, and take action on, environmental issues.
I don’t simply mean the spectrum of environmental issues—e.g., that some people focus on climate change while others focus on biodiversity. What I mean is that people who care about environmental issues don’t come at them in some homogeneous “green” way. This flies in the face of simplistic discourse about choosing “green” or “eco-” alternatives (say, in a consumer product or urban planning policy), and reminds us that, even if we do want to think/act “green,” there are still important considerations and choices to be made.
So, what are the many shades of green? One would think that there is some settled scholarly taxonomy…yet there isn’t. There are indeed many attempts, some quite simple (e.g., Alex Steffen’s bright/dark/light green alternatives), and others much more involved—the current winner, by far, may be the Integral Ecology appendix appropriately titled “200+ Perspectives on Nature.” A question I posed to my students: in what ways are these taxonomies defensible, i.e., how do the authors justify them? This led to a quick discussion around coherence vs. correspondence theories of truth, and we ultimately recommended both: any attempt we make to map out those many shades of green should both make sense (coherence) and be based on some empirical evidence (correspondence).
This led to a two-course, two-semester project my students undertook with me, based on a survey they designed and administered primarily involving their Facebook networks (thus the sample ended up being predominantly white, young, upper-middle class, and female). The survey was quite broad, but in this post I just want to focus on some interesting “many shades of green” results obtained via factor analysis. Factor analysis is a data reduction technique that attempts to combine a given number of initial variables into a smaller number of factor variables. Without factor analysis, for instance, those 200+ perspectives on nature are just that: 200 separate perspectives, with no discernible pattern. But maybe those perspectives have common points of agreement/disagreement: these then would be their underlying factors. I’ve invoked factor analysis to argue for two broad domains of trust in authority among Americans…and I’ve critiqued factor analysis in a study suggesting sustainability as a unifying principle in environmental studies and sciences. It’s no magic wand, and must be applied thoughtfully, but perhaps factor analysis can help move us as we seek greater correspondence between green taxonomies and survey data.
For our EcoTypes survey, we decided to create scales—sets of common questions around a key theme—for five areas students were interested in:
- The scale of environmental action, specifically whether change can effectively occur incrementally via many individual-scale actions, or whether effective change requires larger-scale institutional and structural actions.
- Fear of environmental dystopias, suggesting greater or lesser concern over whether the state of the earth is getting worse and approaching some sort of calamity.
- Support (or not) for inclusive environmental solutions, which would draw upon and benefit from a wide range of participants and perspectives.
- Belief (or not) in the purity of nature, by which nature is understood to be separate from, and generally degraded by, human actions.
- Belief in science and technology as a potential force for good (or evil) in future—a longstanding point of difference among environmentalists.
For our very short survey, students came up with five candidate statements for each of the above themes based on existing surveys and their own creativity; we then reduced each scale to three items based on a presurvey and statistics such as Cronbach’s alpha. Students administered the survey in November and December 2015, with respondents agreeing or disagreeing with these EcoTypes statements on a six-point Likert scale.
In all, we received 114 responses. This is a relatively small N for factor analysis (e.g., my earlier work on trust in authority among Americans had over 1000 responses), and the literature on ideal N is ambiguous, but it’s best not to load too many initial variables onto a small-N factor analysis, so for the analysis below I calculated the arithmetic mean score for each 3-item scale (inverting negatively worded items as needed), and used these averaged scales for factor analysis, not their component items. Here are the descriptive stats on these five scales (with potential Likert range 1–6, thus midpoint 3.5).
[table id=5 /]
The descriptive results above suggest, among other things, that our respondents generally strongly feel some sense of dystopian dread, and generally support small-scale action as effective. The least variance (disagreement) was over inclusive solutions; the most concerned large-scale action and whether science/technology is good.
Now, to the factor analysis. To keep things simple, I ran the analysis so as to obtain two orthogonal results (Varimax)—the sort of result one can (readily) visualize as two right-angle axes. The results for these five EcoTypes areas are below—think of these numbers as correlations, where strong loadings approach +/- 1.00.
[table id=6 /]
Remember that these results are constrained by our survey! But, based on these five areas alone, some interesting patterns emerge. Factor 1 (which explained 30% of variance in the five scales), combined dystopian dread, nature as pure, and doubt in science/technology (thus the negative loading). Factor 2 (explaining 23% of variance) focused on large-scale action and discounted inclusive solutions.
Factor 1 reminds us of mainstream late-20th century American environmentalist discourse: things are bad and getting worse, people are destroying nature, and science/technology is not to be trusted. We can call this factor “classic green”—if by classic we mean a particular time and place and politics. That this factor becomes the major axis of differentiation among our respondents (remember that factors point out main areas of agreement/disagreement) may suggest that these classic threads in U.S. green discourse still strongly resonate among some participants in the survey, yet resonate weakly or not at all among others (possibly some of our students?), who have been influenced by more recent, post-classic environmentalist notions such as ecomodernism or political ecology.
Factor 2 is, frankly, a coherence-criterion puzzle to me: what logical connection is there between large-scale action and non-inclusive solutions?? The two seem contradictory, since any larger-scale action by definition requires an inclusive approach given the reality of multiple points of view. There is some possibly useful theory on all this: the notion of cosmopolitanism, for instance, grapples with the reality of diversity (at least in certain versions, e.g., that of Ulrich Beck) set against the necessity for common action. Perhaps, then, factor 2 expresses the partly anti-cosmopolitan sentiment that large-scale action requires common goals and common values—or, more negatively, that emphasis on diversity ultimately makes larger-scale action impossible…as some cosmopolitans may argue in the context of arguments deriving from interpretations from classic liberalism that foreground individual human rights. So, let’s call factor 2 “whole-earth,” championing a common global identity and the need to act in common, and hearkening back to such green roots as the Whole Earth Catalog and early 1970s whole earth image.
There you have it for now: our students’ Facebook friends exhibit many shades of green, ultimately boiling down to differences over (a) classic green and (b) whole-earth notions of environmentalism. What would be next? Of course, a larger and more representative N; perhaps more scales derived from analyzing scholarly and popular green (and anti-green) discourse; and interviews to see if the sense we make of factors makes any sense to respondents—this indeed was how I explored the factors identified in my study of trust in authority. Lots to be done in coming up with a more robust notion of our many shades of green!; I look forward to what our Lewis & Clark students and others contribute as we move forward in getting beyond that simplistic green/not-green binary.