The Limits of Certitude

January 12, 2009

View as Single Page

Interacting with people in the sciences over my life one of the major issues I have noticed is an extreme hubris when it comes to their opinions in regards to non-scientific issues. Many individuals in the sciences consider all issues fundamentally scientific.  This engenders a certitude when it comes to public policy; there are no opinions, there are true descriptions of reality and false ones.  The underlying reason for this sort of outlook is obvious: science is an extremely powerful tool with which to model and predict phenomena of interest.  But for those with a scientific background this is no abstract theory, it is a lived experience. Granted, the edges and frontiers of science are fraught with hundreds of false leads for every sliver of new insight, but the basic background assumptions which new scientific endeavors are predicated upon have been validated to a high level of precision.  If the scientists at the Jet Propulsion Laboratory can plan unmanned spacecraft trajectories to Mars with error on the order of meters, what is something as simple as humanity?

This is where history helps, the idea that science can easily tackle social problems is not new.  In the wake of the incredible physical discoveries ushered in by Isaac Newton and his successors, in the 18th century social utopians naturally turned to rational and empirical methodologies to design a better society.  The failures of Robert Owen were not out of the norm.  While the utopians of the early 19th century looked to the physical sciences, those of the late 19th century lived in the wake of The Origin of the Species, a work which earned the admiration of Karl Marx and a range of Social Darwinists.  The few successes, and many failures, of the “scientific” method in social “engineering” should instill in us a natural caution.  The experimental data is in, and the theoretical models have not been fruitful.

The checkered history of scientism is not a weakness of the scientific methodology in social domains, but rather a function of the nature of human societies.  This is clear even within the natural sciences.  One of my professors, an evolutionary ecologist by training, told me once about a graduate level course in which he was enrolled that included a physics student.  At the end of the term when they handed in the results of their research project he noticed that the physicist’s report was far thicker than one would have expected.  It turned out that a physics graduate student was not habituated toward the enormous errors which are the norm in statistical ecology; he spent dozens of pages hypothesizing the exact sources of error where the other students simply accepted as normal a large amount of “noise.”  In a deep philosophical sense one might contend that all of biology is reducible to physics, but the reason for dividing these disciplines is that the heuristics and frameworks which are fruitful in physics are not always useful in biology.

Our modern world in which we are embedded is to a large extent an outgrowth of the ingenuity of scientific engineering.  The computer on which I am typing this post has more power than the whole air traffic control system of the United States in the 1980s.  The saturation of cell phones and iPods subtly change expectations and the nature of human experience.  The recourse to Google has also shifted the nature of returns to erudition.  But many of these technological changes are extensions upon physical science, arguably the most precise and powerful of the sciences.  With the rise of personal genomics the biological sciences are now also surfing the wave of technological advance, but there is a qualitative difference between biology’s contingent qualified implications and the clean certitude of physics.  Remember the end of infectious disease?  Evolution is adaptive, and victory in one battle does not win the war. We will always have to revisit this war every few generations as the pathogens adapt to the new techniques.  Contrast this with transportation, it isn’t as if the nature of friction changes to make our current jets less efficient in the future, all things equal.  The complex dynamics of biology are different from those of much of physics; diffusion equations are helpful in both materials science and evolutionary genetics, but the nature of the utility is not interchangeable.

Just as on a fundamental level biology is reducible to physics, so psychology is reducible to biology, and sociology and economics are reducible to psychology.  Though I am one who believes that this unity of nature is real and deeply significant, it is important to recall the differences in the utility of theoretical frameworks across the various disciplines.  Newtonian Mechanics is an imperfect approximation, superseded in the 20th century, but its imperfection is of a qualitatively different degree than that of the rational actor model in economics.  While Newtonian Mechanics is wrong only on the margins of the human scale of existence, some scholars have made the respectable argument that the rational actor model does not explain most human behavior!  Like the life sciences the social sciences are noisy and must make recourse to abstruse statistical tools, unlike biology there are obvious ethical boundaries as to the reach of experimentation.

All these limits to the power of knowledge, and its precision, is an argument against theory and for experience.  Societies are complex, and no human truly understands them on any deep level.  It is likely that there are many latent functions or variables which we do not discern.  Even today most social scientists are at a loss to explain the cross-cultural rise and subsequent decline in crime over the last two generations in developed nations.  The most radical of the Enlightenment philosophers saw much that was “useless” in human culture which detracted from “utility.”  This is not an original viewpoint, the Chinese philosopher Mozi expressed the same sentiment when noting the plethora of customs and traditions which scaffolded the society of his day.  It is perhaps notable that the philosophical school which emerged victorious out of the great debates of Mozi’s era, the Confucians, accepted that rites were an essential aspect of human existence, and served to underpin a social-political system which persisted for 2,000 years.  The rise of functionalists such as Émile Durkheim offered a utilitarian rationale for rites, but perhaps one might be thankful that most humans do not require a reasoned justification for every single act.

This is not to say that theory has no utility, or never will have utility, in the domains of the human sciences.  On the contrary, I have hopes that we are living on the precipice of a new age of integrative social science which combines the insights of psychology, the biological sciences and economics.  But the theories that we have now, in particular those which are still in early stages of gestation, should be treated with caution and skepticism, as is the idealized stance of a scientist to new ideas.  We can always rebuild a bridge which collapses, but who will be there to rebuild society if it collapses?

For Email Updates