oudonquijote

Layman on Feynman on Motion – Sarah Smith

Advertisements

Richard Feynman explaining science in a way that even Letters majors could understand.

Feynman begins his description of motion with a discussion of description itself: the concept of motion is defined by the changes that occur in bodies over time, he says, and “to find the laws governing [these] changes, … we must be able to describe the changes and have some way to record them.” Feynman’s recognition of this descriptive necessity then launches him into the basics of the description of motion, that it can be described either in a table recording time and the distance travelled in that time, or in a graph relating these same factors. After this, Feynman discusses the basics of motion, including how we actually do describe it and how we calculate changes in it. But throughout the chapter, he refers back to this problem of description, namely regarding the degree of precision that can be reached when describing motion.

Feynman first refers to a problem that occurs between misguided philosophers who require a precise definition for every last term in an argument. In Feynman’s hypothetical, one philosopher says to another, “You don’t know what you are talking about!” and the other questions what the he meant by “you,” “know,” “talking,” and so on. In order to be constructive, philosophers—and physicists—must avoid such misplaced rigor of defining and accept each other’s axioms. In physics, or in Feynman’s discussion of motion particularly, the axioms that must be accepted are “time” and “space.” Precision in defining these terms is not necessary to understanding the basics of motion; the complexities and subtleties can only be discussed after the base topics are explained. Furthermore, complete precision in terms of describing measurements is impossible. In math language, certain measurements can be described, but the translation from math to human language is problematic in that human language is necessarily much less mathematically precise than math language.

The primary barrier to precision in science is humans, really, and this barrier is shared between the social sciences, obviously, and the “hard” sciences like physics. Feynman intimates the human barrier to precision in physics with contrasting examples of motion: a falling body and a lady driving a car. In the prior example, the motion can be described fairly precisely in formulas because basically the only variables involved are the falling body and gravity, but in the latter example, the variables include not only the speed of the car but the decisions of the lady driving it, not to mention all the factors contributing to her decisions. The latter example of the lady driving the car is inherently inconsistent because it relies on human decisions and interactions, which involve a prohibitively large number of variables and input sources.

Nate Silver (pictured) is the god of accuracy in the social sciences. Through big-data, a statistical expertise, and thousands of computer-generated models, Silver was able to predict the state-by-state outcome of the 2012 Presidential Election perfectly after having missed only one state in 2008. Dealing with humans remains tricky in every social science; however, with increasingly sophisticated technology and techniques, experts are coming closer and closer to hard-science-like precision. Silver covers the statistical side of politics on his blog: fivethirtyeight.

Ultimately, I think it’s significant for us to recognize the limitations of science where it deals with humans. Social scientists have been forced to realize these limitations much more explicitly because social science deals explicitly with humans. Social scientists have to account for the human problem in every instance of data collection; a frequent criticism of these scientists is that they fail to consider numerous possible alternative influences on their data results. For example, in any given survey, respondents may answer a given question in several ways, and the way they choose to interpret the question affects the answers they choose to give. To account for this potential inconsistency, social scientists must assume that the results of their data collections are generally rather than absolutely accurate.

We often assume that sciences like physics avoid these limitations entirely because they’re objective, empirical, mathematical, but physics does share these limitations where its subject matter intersects with that of the social sciences. In the same way that social science experiments account for human inconsistency and variability in results, physics experiments involving humans at all (as with measuring the motion of the lady-driven car) should account for those possibilities as well, and these experiments should be interpreted the same: as generally, not absolutely, accurate. Furthermore, a commonality shared by experiments done in any science is that the data must be interpreted by humans. While in some cases of interpretation this contention is more significant than others, in all cases the fallibility of the human interpreter should be taken into account, as data points can be missed or misread to alter the perceived results or implications of the whole experiment.

Overall, Feynman’s eighth chapter is ostensibly about motion, but peripherally, he brings up philosophies and problems inherent to the sciences as well. The main problem with all data collection is humans—we’re inconsistent, we’re fallible, and too many variables go into every decision we make to guarantee any high degree of accuracy. Ultimately, as interpreters of scientists’ interpretations of data, regardless of specific field, we should maintain more awareness of our potential for imprecision and inaccuracy when discerning truth.

Advertisements