Example 1: "Judge Sandra Olson recused herself from presiding at the trial of her niece, Ms. Bernice Smith, who had been charged with embezzling from the Tri-County Children's Fund. Judge Olson said she wanted to avoid even an appearance of bias."
What does "bias" mean here? Does the term name a valid concept? If the concept is invalid, should I keep the term, but use it to label a valid concept of my own creation, or should I abandon the term?
DICTIONARY USAGE. Like most general-purpose dictionaries, mine--The Random House Dictionary of the English Language, 2nd ed., Unabridged--does not formally define concepts. Instead it records various conventional usages of terms. The usage of "bias" relevant to individuals is: having a "particular tendency or inclination . . . that prevents unprejudiced consideration of a question."
I see two points worth spotlighting. The first is the idea of "tendency" ("inclination"). Where does a "tendency" arise? It arises in the subconscious, an area not directly visible to me, the area where I may make decisions but without awareness of the process and the factors that cause me to reach a particular conclusion. (I can, of course, consciously test and evaluate any decision that emerges from my subconscious; and I can consciously identify my "tendencies" and correct for them.)
The second point is the negative phrasing "prevents unprejudiced consideration." What that wording is not saying is "bias undermines objectivity" (which is a logical relationship between facts of reality and ideas). Our cognitive goal should be objectivity, not merely avoiding prejudice.
Returning to the conventional usage of the term "bias," another question arises. Why would an individual have a subconscious "tendency" to reach nonobjective conclusions? The following example illustrates a typical situation in which the term "bias" would be used.
Example 2: "In civilian life, Mr. Blair ran the Institute for Defense Analyses, a nonprofit group . . . that does extensive work for the Pentagon. But he had to step down from the post in 2006 amid concerns that his positions on the boards of several defense contractors constituted a conflict of interest."
In this situation, Mr. Blair has an "interest" (extra income) that might lead him to favor a particular recommendation to the Pentagon because it would benefit him personally, not because the recommendation is the best for defending the country. For instance, Mr. Blair might, despite contrary evidence, recommend that the Pentagon purchase a certain new combat aircraft, one manufactured by a company of which he is a paid director.
Those who watch government are right to raise an alarm when consultants to the government are in a position to benefit indirectly from the advice they give. However, an alarm is not a proof. A particular individual, if he values objectivity, can have an apparent "conflict of interest" and yet render objective judgments. Not any particular short-term benefit, but the principle of objectivity as well as the virtues of rationality and honesty, can and should guide him. Only cynics--those who believe virtue is impossible--would automatically assume everyone in a supposed conflict of interest automatically chooses the particular interest over the principle.
Of course, a fully objective person--one who values reason and therefore objectivity--has no actual long-term "conflict of interests." He examines facts, draws conclusions logically from those facts, and then evaluates them objectively, in the full context of his philosophical as well as personal values. (Honest individuals can be confused, on the short-term, in selecting between particular values, but not between principles vs. particular values.)
CONCLUSION. In its main elements and charitably interpreted, the conventional meaning of bias is valid but poorly formulated. A proper definition is: a mental condition in which a value held in the subconscious might lead a person who does not supremely value objectivity to make a judgment not supported logically by facts of the case being judged.
APPLICATIONS. In a debate, if I hear the word "bias" used as an accusation, I take it to mean that the accuser thinks the accused person has a particular interest (value) that might be more important to him than objectivity. (Ironically a charge of bias, if unexplained and unsupported, can prejudice an uncritical audience. It is a smear against the accused person.) If such an accusation is being made as an attack on the person, then I would ask for a definition and proof.
On the other hand, if the debater is bringing up bias as a caution--as in examining the medical results of clinical studies whose conclusions favor the pharmaceutical manufacturers who financed the studies--then I would accept the caution but make a point of not leaving it hanging in the air. I would ask for evidence that the conclusions were or were not objective.
I will generally avoid the term "bias." I will try to be more specific by using "illogical," "prejudice," "rationalization," or a similar concept that directly identifies the problem, if there is one. I will also stress the positive, objectivity, as the standard.
Author of The Power and the Glory: The Key Ideas and Crusading Lives of Eight Debaters of Reason vs. Faith
 Mark Mazzetti, "Likely Pick for Intelligence Chief Would Face Task of Corralling Fractious Agencies," New York Times, Dec. 20, '08, online.  For examples of (subconscious) prejudice influencing decisions in science: David Harriman, "Electric Current" subsection of "Errors in Inductive Reasoning," The Objective Standard, Winter 2008-09.
Dec 26, 2008
What is bias?
Posted by Burgess Laughlin at 7:56 AM 3 comments:
Labels: bias, objectivity, prejudice, smear
Subscribe to: Posts (Atom)