The Demon-Haunted World: Science as a Candle in the Dark

Sagan discusses his feelings of grief after the death of his parents, to whom he was very close, and his natural longing to believe that aspects of them remained in one form or another after death. He then disparages those who capitalize on this human need, such as mediums and channeling spiritualists, before providing an exhaustive list of what he calls “cases of proved or presumptive baloney” (209), in which people knowingly deceived someone else by assuming beliefs to which they do not actually subscribe. To counter this, Sagan presents what he calls the “baloney detection kit” full of “Tools for skeptical thinking” (210), which provide a foundation of falsifiability against spurious claims, such as abduction by aliens:

• seek independent confirmation of facts
• encourage debate on the evidence from experts of opposing points of view
• construct multiple working hypothesis
• do not overly emotionally attach to your own hypothesis
• quantify (put into numbers) data as often as possible
• every link in the chain of argument must work
• adhere to Occam’s Razor, the idea that the simplest solution is often the correct solution
• always ask if the hypothesis is falsifiable

After providing tools for proactive thinking, Sagan next reveals what not to do by listing 20 logical fallacies, which are common errors in reasoning that weaken or invalidate arguments. A good place to look for such fallacies, Sagan reminds, is in religious and political arguments, as their practitioners are “often obliged to justify two contradictory positions” (212). The logical fallacies Sagan lists are:

• ad hominem, in which one attacks the arguer’s character rather than their argument
• argument from authority, in which authority figures make claims which are unchecked simply because they are authorities
• argument from adverse consequences, in which it is believed that extreme consequences are necessary for regulating human behavior
• appeal to ignorance, in which one argues that if something has not been proven false, then it must be true, or vice versa
• special pleading, in which the arguer claims special or unsharable knowledge, to support their proposition
• begging the question, in which an argument’s premise assumes the truth of its conclusion, rather than supporting the truth
• observational selection, in which only flattering or supportive data is presented, rather than all relevant data
• statistics of small numbers, in which the results of small studies are generalized into conclusions about wider populations
• misunderstanding the nature of statistics
• inconsistency
• non sequitur, in which there is a lack of connection between the premises and the conclusion that follows
• post hoc ergo propter hoc, in which one assumes one thing is caused by another, simply because the first thing came first
• meaningless question, in which the logic of the question invalidates its own premises
• excluded middle, or false dichotomy, in which only the extremes of a proposition are considered, rather than the continuum of possibilities
• short-term vs. long-term, in which short-term and long-term concerns are made to stand against each other even though they require wholly different approaches
• slippery slope, in which unrealistic and unproven consequences are claimed to result from simpler propositions
• confusion of correlation and causation, in which unrelated occurrences are thought to be cause-and-effect rather than simply concurrent
• straw man, in which a position is oversimplified or caricatured in order to make it easier to attack
• suppressed evidence, in which evidence contrary to the arguer’s point is ignored, or occluded
• weasel words, in which harmless-sounding euphemisms are used to cover up harmful or unpopular actions and policies

After describing the proactive steps for thinking and an awareness of the common logical fallacies, Sagan is careful to warn that an unevaluated adherence to his suggestions goes against the principles of critical reasoning: Mindless reliance can cause the toolkit to be “misused, applied out of context, or even employed as a rote alternative to thinking” (216). To reinforce the vital effectiveness of his toolkit, Sagan gives the example of tobacco companies proffering false science and data denial as a literal way that “Gullibility kills” (218).