WEBLOG
Previous Month | RSS/XML | Current | Next Month
December 31st, 2010 (Permalink)
Made with All Natural Ingredients!
I can honestly say that The Fallacy Files has no MSG. Never had it; never will! Also, it has no preservatives or artificial flavors, unlike some Frito-Lay chips.
Snack-maker Frito-Lay is introducing "natural" versions of its potato chips and other snacks. I saw bags of Lay's potato chips prominently displayed at the grocery store the other day bearing new "made with all natural ingredients" labels, which say that they have no MSG, preservatives, or artificial flavors. Frito-Lay is owned by Pepsico, the makers of Pepsi Cola, and longtime readers will remember Pepsi Natural, which also claims to be made from all natural ingredients, such as sugar instead of corn syrup. Why sugar is more natural than corn syrup remains a mystery.
In contrast, psychologist David Barash, who likes nature, recognizes that there is bad stuff in it:
Consider typhoid, cholera, polio, plague, and HIV: What can be more natural than viruses or bacteria, composed as they are of proteins, nucleic acids, carbohydrates, and the like? Do you object to vaccination? You'd probably object even more to smallpox. …I suspect that those well-intentioned people who admire "natural" raw milk have never experienced the ravages of Campylobacter, pathogenic E. coli, or bovine tuberculosis, each spread by the unpasteurized McCoy.
Now, they can wash down some Lay's potato chips with raw milk and find out! These thoughts lead Barash into deep philosophical waters:
In his A Treatise of Human Nature, in the 18th century, David Hume presented, and criticized, what has come to be known as the "is-ought problem"… Is there any way, he asked, that we can legitimately connect how the world "is"…with how it "ought" to be (including how we ought to behave)? Simply by raising the question, he so conclusively severed "is" from "ought" that the distinction―between the descriptive and the prescriptive, between facts and values―is called "Hume's Guillotine." His insight that it is fallacious to derive "ought" from "is" has become known as the "naturalistic fallacy," a term coined by the British philosopher G.E. Moore in his 1903 book, Principia Ethica.
True enough, but as I've pointed out before (see the Exposure section of the Appeal to Nature entry), this is not what Moore meant by "naturalistic fallacy". "Naturalistic fallacy" was an unfortunate choice of a name, since it sounds as if it ought to mean the appeal to nature. "Definist fallacy" has often been suggested as a better alternative, because Moore thought that "good" was a simple, indefinable concept. Thus, trying to define "good" in naturalistic terms would commit the definist "fallacy", but so would defining it in theological terms, for instance, defining "good" as "what God wills" would also commit the fallacy.
In any case, the "naturalistic" or "definist" fallacy is not a logical fallacy. For one thing, it's not a mistake in reasoning, but a supposed error in definition. For another thing, it's at best a mistake in ethics, rather than logic, and a controversial one at that.
At the same time that Barash is accepting an is-ought divide, Michael Shermer is trying to bridge it:
The "naturalistic fallacy," sometimes rendered as the "is-ought problem"―the way something "is" does not mean that is the way it "ought" to be―has for centuries been piously parroted from its leading proponents, philosophers David Hume and G. E. Moore, as if pronouncing it closes the door to further scientific inquiry. We should be skeptical of this divide. If morals and values should not be based on the way things are―reality―then on what should they be based?
It certainly is invalid to reason from the fact that something is the case to the claim that it ought to be the case; you can't turn around without tripping over a counter-example. However, Hume had more in mind: he thought it impossible to argue validly from any facts about the world―i.e., "is" statements―to an "ought" statement. This is not at all obvious, and is philosophically controversial. For that reason, I won't take a position here on whether there is such a "gap" between "is" and "ought" statements, though I agree with Shermer that skepticism is warranted.
Besides, I don't need to, because my objection to the appeal to nature is not based on this supposed gap, but on the vagueness and ambiguity of the concept of the natural. What is "natural" or "unnatural" is so unclear that it can't be the basis for any conclusion of consequence, whether about what snacks to eat or if its wrong to clone a human being.
Sources:
- David P. Barash, "Two Cheers for Nature", The Chronicle Review, 12/12/2010
- Gary Curtis, "It's Only Natural!", Fallacy Files Weblog, 5/3/2009
- Bruce Horovitz, "Frito-Lay to make snacks from natural ingredients", USA Today, 12/28/2010
- Michael Shermer, "The Science of Right and Wrong", Scientific American, 12/23/2010
- Don Teague, "Frito-Lay Targets All-Natural Market", CBS Evening News, 12/28/2010
December 19th, 2010 (Permalink)
New Book
December 14th, 2010 (Permalink)
Check 'Em Out
- (12/18/2010) Every time a new technology is introduced, someone will claim that it causes cancer or some other problem―for instance, there was the "cell phones cause brain cancer" claim. Most such claims are the result of fallacious post hoc causal reasoning, though there's always the possibility of a real effect hidden among all the spurious ones.
A recent claim of this type is that wi-fi networks cause damage to trees. Carl "The Numbers Guy" Bialik gives a thorough account of how this claim got started, how it was misreported by the news media, and blown out of proportion partly due to bad translations.
The study that was misreported, while preliminary, is the sort of thing that needs to be done to see whether wi-fi does damage trees. Just noticing tree damage after―"post hoc"―the introduction of wi-fi isn't sufficient, since there may be no more damage than before the wi-fi, and there are many other things that could damage trees in the urban environment. One scientist quoted by Bialik describes how to do it:
Chris Guy, a systems engineer at the University of Reading in the U.K.,…wrote in an email…"To have any meaning there would have to be a proper blind trial, with a control group. One group of trees exposed to radiation and one not, with the observer unaware of which group was which. They would have to be sufficiently close to experience the same soil and weather conditions but far enough apart for the Wi-fi signals to be close to one but effectively zero at the other. Then we might start to believe the results."
However, even that is not enough, as another scientist quoted by Bialik points out: "'Replication is critically important in experiments,' said John Stufken, head of the department of statistics at the University of Georgia." When a study of the type that Guy describes has been done and replicated, we can start believing.
Sources:
- Carl Bialik, "The Numbers Guy: Wi-Fi Threat to Trees Rooted in Shaky Stats", The Wall Street Journal, 12/11/2010
- Carl Bialik, "Trees and Wi-Fi May Co-Exist After All", The Numbers Guy, 12/10/2010
- Speaking of how not to establish causation, as I was in the previous two entries, Ben Goldacre discusses a case of jumping to a causal conclusion on insufficient evidence, and even in the face of counter-evidence. He details, among other things, how a company used patient anecdotes to advertise a medical device. The anecdotes are misleading since the device failed clinical tests of efficacy, but even if it hadn't failed the trial, such anecdotes provide no evidence at all that a treatment works.
The trial is a good illustration of why anecdotes are worthless evidence: three patients who received the treatment had positive results, while three who received a placebo treatment also had positive outcomes. Guess which three patients' anecdotes were used in the ads. However, it could just as easily have been the other three, and the three placebo patients may be sincerely convinced that they were helped by the treatment―unless they were told subsequently that they were in the control group. Such is the power of post hoc reasoning. Moreover, they probably would have been glad to promote the product. Keep this in mind the next time you see such an ad.
Source: Ben Goldacre, "A migraine-inducing libel case", Bad Science, 12/11/2010
December 9th, 2010 (Permalink)
Reader Response
Dave Bjornstad writes:
My question is this: What do you call it when A makes a statement, and B says that A's statement contains logical fallacy C, only it really doesn't?
I call it "fallacy abuse"; Howard Kahane called it "False Charge of Fallacy". Dave's question refers to the following example that used to be on the "Stalking the Wild Fallacy" page:
The Lies Behind Gun Control
"We need safe storage laws." False. States that passed "safe storage" laws have high crime rates, especially higher rates of rape and aggravated assault against women.
Back to Dave:
The explanation says that this is an example of Post Hoc, Ergo Propter Hoc; i.e., event C (the passage of safe storage laws) happened immediately prior to event E (high crime rates); therefore, C caused E. However, that's not what the statement says at all. It says nothing about the crime rates' increasing after the passage of safe storage laws. At best, it says that the passage of safe storage laws failed to prevent high crime rates, or failed to lower crime rates that were already high. Phyllis Schlafly and her buddies don't always exercise the clearest thinking. But this time she (or whoever wrote the article in the journal that bears her name) is innocent of the charge of fallacious reasoning.
In order to understand the example, a little background knowledge is needed: "safe storage laws" legally require that guns be stored in such a way as to make it difficult for children to get ahold of them. The rationale for such laws is to curtail accidental deaths and injuries resulting from loaded guns in the hands of kids.
The example lacks explicit causal language linking safe storage laws and high crime rates. However, the last part of the example reads: "States that passed 'safe storage' laws have…higher rates of rape and aggravated assault against women." This claim contains a dangling comparative, namely, "higher". Higher than what? There are two possibilities that occur to me:
- Higher than it was before the law was passed. If this is what is meant, the argument is obviously a post hoc.
- Higher than states that did not pass the law. In this case, the argument might be a cum hoc rather than a post hoc.
Why does the example address the issue of the crime rates in states with safe storage laws, and not the rates of firearms accidents involving children? As far as I know, no one who supports such laws suggests that they will lower the crime rate; rather, they are aimed at lowering the rate of accidents. As a consequence, if the example is really arguing that "the passage of safe storage laws failed to prevent high crime rates, or failed to lower crime rates that were already high", then it is attacking a straw man.
Rather than just being a red herring, I think the reason that crime rates are introduced is that there is a causal story of how safe storage laws could lead to more crime. Storing a gun in such a way that children cannot misuse it―for instance, by keeping it unloaded, storing it in a safe, or using a trigger lock―also makes it more difficult for the owner to use it in an emergency. Thus, the main objection to such laws is that they will make it hard for law-abiding gun owners to use their weapons to defend themselves, their families, or property from criminals.
It's plausible that safe storage laws might indeed have such a side effect, but plausibility is not enough to establish causation. Nor is the fact―if it is a fact―that the crime rate has risen in states that have adopted the laws, or is higher in states with such laws than in states without them. The fact that the example mentions the supposed "higher" crime rates leads me to believe that it is at least setting up a logical boobytrap for the reader, specifically, a Non Causa Pro Causa boobytrap. If that's not what's going on, then the example is either a straw man or a simple non sequitur.
I've removed the example from the wild fallacies page, even though I still think that the most plausible interpretation of it is as a post hoc argument. However, it's a much more complicated example than I originally realized, and thus needs more room for explanation than is appropriate for that page. It's also a lot more interesting than I first thought!
Sources:
- "Law Enforcement And Gun Violence Prevention Groups Urge Massachusetts High Court To Uphold Safe Gun Storage Law", Brady Campaign to Prevent Gun Violence, 6/29/2009
- "The Media Campaign Against Gun Ownership", The Phyllis Schlafly Report, Vol. 33, No. 11, June 2000
- "Safe Gun Storage in North Carolina", North Carolinians Against Gun Violence Education Fund
- Howard Kahane, Logic and Contemporary Rhetoric: The Use of Reason in Everyday Life (2nd Edition), p. 66
December 6th, 2010 (Permalink)
Check it Out
We know about optical illusions, and we’re familiar with the ways that our eyes can be misled. It would be nice if we could also be wary of cognitive illusions that affect our reasoning apparatus….―Ben Goldacre
Human causal detection abilities are oversensitive, and tend to produce a lot of false positives, that is, we often jump to causal conclusions on insufficient evidence. It's a lot harder to prove a cause and effect relationship than people seem to imagine.
In his latest "Bad Science" column, Ben Goldacre reports on new research into the formation of false causal beliefs and, perhaps more importantly, how such false beliefs can be minimized. Not surprisingly, when told about ill patients recovering after they received a drug, people infer that the drug cured the disease. Apparently, it never occurs to them that the person might have recovered without taking the drug. As the old joke goes, left to itself a cold will last seven days, but proper treatment will clear it up in a week.
Judging from Goldacre's description, neither group in the experiment received enough information to warrant drawing a causal conclusion. However, hearing about more cases of people recovering who did not receive the drug seems to lessen the tendency to impute curative powers to it. Now, if only people were able to draw the general conclusion that post hoc reasoning is fallacious.
Source: Ben Goldacre, "Illusions of Control", Bad Science, 12/4/2010
December 3rd, 2010 (Permalink)
Blurb Watch: The Next Three Days
Here's a blurb from an ad for the new Russell Crowe movie, together with its context from the review:
| Blurb | Context |
|---|---|
| "A SMART, GRIPPING, AND IMMENSELY ENTERTAINING SUSPENSE THRILLER!"
Scott Mantz, ACCESS HOLLYWOOD |
The payoff occurs when the jailbreak begins, kicking the last 30 minutes into high gear as a smart, gripping and immensely entertaining suspense thriller. |
| Source: Ad for The Next Three Days, The Dallas Morning News, 12/3/2010 | Source: Scott Mantz, "MovieMantz Review: ‘The Next Three Days’", Access Hollywood, 11/19/2010 |
Maybe you should watch just the last half hour.