Writing Analytically, 6th Edition - Rosenwasser, David & Stephen, Jill [116]
—Jack Gambino, Professor of Political Science
The preferences of different disciplines for certain kinds of evidence notwithstanding, most professors share the conviction that the evidence you choose to present should not be one-sided. They also understand that the observation and use of evidence is never completely neutral.
A useful example for thinking about evidence-gathering in this context is Werner Heisenberg’s famous formulation, the Uncertainty Principle. A theoretical physicist, Heisenberg hypothesized a subatomic particle orbiting the nucleus of an atom that could be observed only when it passed through a concentrated beam of light. At the instant of its illumination, however, the direction of the particle would necessarily be skewed by the beam. From this model, Heisenberg concluded that the act of observation invariably alters whatever is observed.
This insight has made its way across the academic disciplines. In anthropology, for example, Clifford Geertz has written extensively on the ways that researchers into other cultures not only impose their own cultural assumptions onto their subjects, but also, by their very presence, cause change in the behavior of the people they are observing.
The challenge of determining what counts as evidence is also at issue when you start with a given problem or question and then must decide what you should look at. Say you are looking into the causes of child abuse. How do you decide what to look at? How do you even define what it is you are studying, since what conceivably constitutes child abuse now might have been considered normal child-rearing practices in the past? If you are searching for causes, what is the important evidence? In the past, the physical environment lay outside what sociologists usually considered, but what if the height of buildings in which child abuse occurs provides better data than the size of families? As this hypothetical example suggests, the relationship between cause and effect is always slippery, and the assumptions about what is and isn’t evidence are potentially blinding.
We are, to a significant degree, a society obsessed with evidence—from UFO fanatics to conspiracy theorists to those who avidly follow the latest leaks in the press about the peccadilloes of the famous. This raises the question of the kinds of evidence that can be used to support claims.
MORE THAN JUST THE FACTS
As we have been suggesting for most of this chapter, evidence is virtually never simply a matter of “the facts.” It is no accident that one often hears the phrase “questions of evidence” because evidence is perennially subject to question—for its accuracy, its veracity, and so forth. When we hear that mellifluous voiceover in the TV commercial assuring us that “3 out of 4 dentists recommend a fluoride toothpaste,” how are we to take that remark? Why are lie detector tests “inadmissible as evidence” in some cases?
Nor is established practice a guarantee that the evidence is reliable. About 15 years ago, sportswriter Bill James levied a powerful attack on the accepted way of providing data for assessing a baseball player’s fielding—the fielding average, which was a ratio of the total number of balls hit to a fielder (his “chances”) as against his errors. James pointed out that fielding percentage didn’t take into account the greater number of balls that a superior fielder might get to in the first place, adding that the more difficult the chance, the more likely the player was to make an error. So the number of errors was not a reliable index of a player’s proficiency. Consequently, James invented a new statistical measure, the range factor, which more heavily weights the number of chances and devalues the number of errors. In effect, he redefines the pool of evidence.
What follows are some passages from student and professional writing, each of which illustrates