The hidden complexity of simple models, or Why theorists of all kinds should be troubled by unmodeled variables having dynamical lives of their own

Can ecological theory generate principles that could be usefully generalized across ecological situations? Particularism has been a perennial attraction in ecology, but a new source of doubt gained momentum by the end of the 1980s after theorists started looking at “indirect interactions”—effects mediated through the populations not immediately in focus, or, more generally, through “hidden variables” that have their own dynamics. How much do indirect effects confound principles derived on the basis of observing the direct interactions among populations? My exploration of this question should challenge not only ecologists, but theorists in all fields that make use of models of any kind of sub-system elevated from the complexity in which the sub-system is actually embedded.

From complexity to construction to intersecting processes: Puzzles for theoretical and social inquiry

Although stable systems may be extremely rare as a fraction of the complex ecological systems being sampled (as shown in the 1970s theoretical work of Robert May), they can be readily constructed over time by the addition of populations from a pool of populations or by elimination of populations from
systems not at a steady state. The implications identified in this paper of such a constructionist perspective could challenge not only ecologists, but also theorists in all fields that make use of models without a process of construction over time of the complexity of the situation studied. This paper centers not so much on advancing this perspective, but on two consequent puzzles: Why does the constructionist view seem difficult for theorists to take up? What social implications should be drawn from the resulting view of complexity, especially to the extent that critical events cannot be predicted?

Hayekian accounts rest on a self-contradictory assumption about natural selection

This post emerges from my puzzling over the overlap between Hayek’s neoliberal critique of attempts to model complexity well enough to make predictions and economic policy and my view that “knowledge, plans, and action [have to] be continually reassessed in response to developments — predicted and surprising alike” (as described in a 2011 post).

Intersecting Processes: May 2017 New England Workshop on Science and Social Change

Location: Old Fire Station, Woods Hole MA, USA
New Dates May 2017, 8am Sat May 27 – 2.30pm Weds May 31

In this five-day workshop participants will create spaces, interactions, and support in formulating plans to extend our own projects of inquiry and engagement around “intersecting processes.”

On ecological philosophy

Introduction to a forthcoming entry in the Oxford Bibliographies in Ecology:

Ecological philosophy does not have a single meaning. To some social thinkers it signifies a worldview that invokes ecology in promoting environmental protection; to others ecology is invoked in relation to a wider realm of social action; to philosophers of science, ecological philosophy may seem to be a synonym for the part of their academic field that focuses on ecology. For this entry in a bibliography in ecology, however, the term is taken to refer to conceptual frameworks in ecological and environmental science (hereon: ecology), and as such combines theorizing in ecology with some contributions from philosophy of ecology. Concepts and theorizing in ecology can be viewed in relation to the challenge faced by all ecologists (taken hereon to include environmental scientists) of dealing with the complexity of ongoing change in the structure of situations that have built up over time from heterogeneous components and are embedded or situated within wider dynamics.

A constructionist perspective on the structure of ecological complexity, follow-up questions

It would be interesting to investigate why the constructionist perspective on ecological complexity (see below) is overlooked. One answer is that people haven’t come across what has been written on that perspective by me and others. But I’m more interested in why hasn’t it been discovered and enunciated by others for themselves and why they don’t discuss its implications once they know about it. This post presents the idea again (quoting from a 2010 post, which draws from Taylor 2005, 3-17) then reviews Robert May’s response to it over the last 30 years.

Historical and autobiographical origins of an ongoing project on scientific and social complexity

From Taylor, Unruly Complexity (2005, xiv ff):

The sequence of cases [in the book] should help researchers and students in this wide range of fields appreciate more acutely the limitations of assuming that ecological, scientific, and social complexity can be delimited into well-bounded systems.  My hope is that readers will then take steps—on their own and in collaboration with others—to reconstruct the unruliness of complexity without suppressing it, to link knowledge-making to social change, and to wrestle with the potential and limitations of critical reflection as a means to redirect practice.  In the words of Raymond Williams (1980, 83), I want to encourage others not to “mentally draw back [and be] spared the effort of looking, in any active way, at the whole complex of social and natural relationships which is at once our product and our activity.”

Historical origins

Why undertake a project that addresses complexity and change across the different realms of science, interpretation of science, and critical reflection on practice?  One answer would be that the realms are already always connected, but concepts and practice are shaped to make the realms seem separate.  This is a position that can only emerge after the book has worked its way through many steps.  A shorter answer that might suffice in the meantime derives from the project’s historical origins, which can be located in the intersection of two kinds of ecology during the 1970s.

A century earlier Ernst Haeckel had defined “ecology” as the study of the complex interrelations among animals, plants, and their living and non-living environments (Allee et al. 1949).  The meaning of the new term soon stretched to refer to the complex interrelations themselves as well as the scientific study of them.  Starting around 1970, “ecology” (and the prefix “eco-“) also became associated with actions responding to the degradation of the environment of humans and other species.  The array of endeavors that have come under the umbrella of ecology-as-social-action is vast: preventing pollution, ozone holes, global climate change, future catastrophe; advocating radical social change, environmental activism, recycling, simpler lifestyles, unrefined foods; preserving nature, biodiversity, endangered species; promoting balance and interdependency.

Ecology-the-science promised to help address ecological concerns from a number of angles.  Researchers competent in using tools of ecological research could provide technical assistance on particular environmental problems.  Systematic environmental analysis and planning might be established so problems could be managed before they became the crises that provoke environmental campaigns.  General theories of ecological complexity might enlighten humans about the conditions for more harmonious relations among people and with other organisms sharing our environment.

The rise of ecology-as-social-action, however, also involved a serious critique of the scientific enterprise.  The presumption that scientific advances constitute Progress was challenged by peace and environmental activists, among others.  The destructive effects of science applied, for example, in military technologies and synthetic agro-chemicals made it hard to justify the pursuit of knowledge as a good thing for all.  The pertinent question was raised: Who benefits from scientific research, and who does not?  Such probing exposed science’s role in many forms of domination: developed nations over former colonies, military and security branches of the State over dissenting citizens, managers over workers, whites over other races, men over women, and humans over non-humans.  Some people saw science in the service of domination as abuse, not use, of science, but other critical commentators associated these tendencies with the nature of scientific inquiry itself.  Either way, science was not viewed as unfettered inquiry; instead, specific developments in scientific knowledge began to be interpreted in terms of the social priorities of the governmental bodies, military agencies, corporations, and individuals who sponsored, created, or applied them.

The critique of science also involved positive proposals for alternative processes of inquiry and alternative applications of the products of science.  To counter the inherent tendencies of science towards domination—or the recurrent abuses of science in that direction—these alternatives should revolve around cooperation and should not take the contributions of other people or species for granted.  Scientists were urged to accept local, democratically formulated input to their research.  Even among scientists who insisted on their freedom of inquiry (albeit within parameters set by their funding sources), there was wide recognition of the need to take more responsibility for how the knowledge they made would be applied.[endnote]

In short, ecology-as-social-action challenged ecological researchers not only to attend to ecological concerns through technical assistance, analysis for planning, or general theories, but also to shape their scientific practices and products self-consciously so as to contribute to transforming the dominant structure of social and environmental relations.  In retrospect, I would read in the broad terms of the critique of science an overoptimistic assessment of the potential, on one hand, for the social movements of the 1960s and 70s to bring about radical restructuring of social relations and, on the other hand, for people to transform their lives accordingly—including, in this context, for scientists to redirect their research.  Yet the 1970s critique of science was a key aspect of the context in which I first began to engage with the complexities of environmental, scientific, and social change together, as part of one project.  The challenge I take up in writing this book, then, is to build on the historical and personal origins of the project and to convey its subsequent evolution in terms that help other researchers engage with such complexities in the context of the early 21st century.

————————

Conceptual exploration: An autobiographical narrative

My decision to study ecology during the early 1970s stemmed from environmental activism in Australia that ranged from a collaboration with trade unionists opposing the construction of an inner-city power station to street theater exposing fraudulent, industry-sponsored recycling plans (Whole Earth Group 1974).  Ecology-the-science was the recommended choice for college students who sought programs of study in which to pursue their interests in ecology-as-social-action—if indeed any other choices were available.  I hoped my studies would lead to some kind of career that would take me beyond responding to one environmental issue after another and instead allow me to help in planning that prevented future problems from emerging.  I also hoped that understanding how to explain the complexities of interactions in life would lend support to less hierarchical and exploitative relationships, both within society and among humans and other species.

I had brought a mathematical disposition to my studies in ecology, so I undertook projects that advanced my skills in quantitative analysis and mathematical modeling.  I was excited to learn that some biologists and mathematicians were creating a specialty called theoretical biology (Waddington 1969).  This discovery was still fresh when I took a course for which E. C. Pielou’s (1969) text on mathematical ecology was assigned.  In the introduction she noted that organisms come from a range of species; within any species they differ in age, sex, genetics, experience, and so on; and any particular individual changes over its lifetime.  Any situation an ecologist might study is continually altered by births and deaths, by migratory exchanges with other places, and by seasons and climatic change.  Even so, ecological regularities persist long enough for most people to recognize some order, such as, an oak-maple forest or the sequence of plants encountered as one moves inland from the seashore (Pielou 1969, 1).  The processes could be simply described, yet the combination of them seemed theoretically challenging—how could ecologists account for order arising out of such complexity?…[continued through the book]

References

Allee, W. C., A. E. Emerson, O. Park, T. Park and K. P. Schmidt (1949). Principles of Animal Ecology. Philadelphia: Saunders.
Pielou, E. C. (1969). An Introduction to Mathematical Ecology. New York: Wiley-Interscience.
Taylor, P.J. (2005) Unruly Complexity: Ecology, Interpretation, Engagement. Chicago: U. Chicago Press.
Waddington, C. H. (Ed.) (1969). Towards a Theoretical Biology. Edinburgh: Edinburgh University Press.
Whole Earth Group (1974). Uncle Afrely’s Earth Guide (info)
Williams, R. (1980). “Ideas of Nature,” in Problems in materialism and culture. London: Verso, 67-85.

Day 1, 2015?—an attempt to address various comments from the taking stock session in a course on gender, race & the complexities ofscience, technology

Course: grst.wikispaces.umb.edu

Getting going on learning tools to use to make connections and contributions to a topic

Understanding angles of action in relation to climate change

Reading Mike Hulme’s (2009) Why we disagree about climate change (Cambridge UP) for a new discussion group leads me to say that I’m most interested in hearing how the other members of the group have pursued their various angles of action in relation to climate change.  Two lines of thinking underlie this interest:

1. I expect the group members already appreciate that scientific knowledge is not established/accepted/disputed solely on the basis of its correspondence to physical reality.  Many diverse practical considerations have to be addressed in trying to make or dispute knowledge and these considerations span intersecting social worlds (lab technicians, computer programmers, funders, journal editors, media portrayals, policy makers, owners, workers,…).  The challenge is where and how to engage within the resulting complexity with a view to changing what is accepted as knowledge and necessarily simultaneously modifying our own social situatedness (where social is a shorthand for the intersecting social world in which each of us makes the many practical decisions).  In addressing this challenge we draw on resources that include aspirations (a sense of the place we want to head towards) and values-talk (what we say to others and ourselves –or even viscerally feel– about what should guide individual and group decisions.

Where I diverge from Hulme is that I do not position aspirations or values as drivers or foundational.  Several reasons for rejecting that:

a. Empirical–Abundant cases of people acting in ways that depart from their expressed values, or shifting their professed values when push comes to shove.

b. Methodological–How does a researcher demonstrate that values are drivers of the diverse practical decisions?

c.  Explanatory weakness–Do we know why people disagree after reading a book that highlights aspirations and values better than after reading Merchants of Doubt, which documents the active work done to discredit scientific results around smoking, ozone, climate change, etc.?

d. Practical, in specific situations–When trying to make changes in some specific situation, it is necessary to address a wide range of practical considerations.  Values talk is a small and often distracting part of that work.

e. Interpretive–The author emphasizing deep drivers can be viewed in terms of his particular positioning in their own specific situation.

f. Political blowback–Movements that emphasize deep drivers and discount the diverse practical considerations facing diverse knowledge-making agents end up producing unintended and undesired consequences.

2.  To the extent that any of us emphasize “diverse practical considerations facing diverse knowledge-making agents” we have to develop frameworks that help us navigate and negotiate a multi-stranded and -layered complexity, more complex in many ways than the supercomputer models of climate change.

Free will discussion often miss a crucial point 2

Let me entertain, then explore the consequences of, one answer to the previous post‘s questions about the free will debate, which was:

Could someone else point to conditions 1,2,3,… (e.g., biochemical or neurological or unconscious) at or before time t that imply that B and C were not actual choices I could have made, that is, that I was mistaken in saying they were?  What method of finding out about the world would show this inadmissibility of choices I had pointed to?

Actually, the point is probably deeper: Could someone with this knowledge of the inadmissibility of choices B and C convey the knowledge to me in a way that would influence my (mis)understanding of the choices I faced?  And, if they could, is this a way of engaging with each other that we want to foster?

Fried et al. (2011) “report progressive neuronal recruitment over ∼1500 ms [=1.5 seconds] before subjects report making the decision to move.”  If (and that’s a big if) that research developed to the state that the firing neurons could be linked to which decision, say choice A, I was about to be able to report that I had just then made, then the neuronal firing would be “the conditions 1, 2, 3, … at or before time t that imply that B and C were not actual choices I could have made.”  The next challenge would be to convey that information, namely, that I had already decided on A, back to me within the 1.5 seconds.  If that became do-able, we would then have to see if conveying that information influenced the decision.  If not, then we would still be left with the last question: “[I]s this a way of engaging with each other that we want to foster?”

It could be argued that the short time—1.5 seconds—is not an issue.  If neurons fire before I am aware of my decision, then it’s neurons all the way back (by analogy with “turtles all the way down“).  Trace the neurons back and someone would have time to provide information that contradicts my view that I have more than one choice.  My response:  Of course there are neuronal firings that precede the final ones 1.5 seconds in advance.  (That is, I am not denying a mechanistic view of living organisms.)  I doubt, however, that firing neurons could be linked forward to which decision would later appear to being made—the computational and statistical complexity of discerning associations among masses of neurons over time is too great.

The traditional rejoinder to doubts about what future research will be able to show is to say we have to wait and see empirically, implying that my response is not a strong argument conceptually (especially given the doubters of the past who have turned out to be very wrong).  This rejoinder fails to address the computational and statistical complexity issue in general or how it plays out in specific cases, such as linking genomes to diseases and behaviors.