In the Mailbox
It occurs to me, following the literal meaning of the suffix "-tomy" ("cutting, incision") that contextomy may be considered part of a larger class of fallacious arguments that "cut" or "incise" context. Here is a hypothetical example. A writer argues that the Mafia is an unjustly persecuted group of citizens by compiling a long list of alleged police brutality, prosecutorial misconduct, and prison mismanagement, all involving Mafiosi as victims. Let us assume that all his facts are more or less accurate, but that the author never once mentions crimes committed by the Mafia. I think that such an argument would amount to contextomy, i.e. cutting out or incising relevant context. Similar arguments easily come to mind: describing the bombing of Dresden during WWII without mentioning WWII; the bombing of Hiroshima without, again, mentioning WWII. What do you think?
Professor Emeritus of Sociology
University of British Columbia
I first learned the word "contextomy" from Boller and George's book They Never Said It, which I quote:
…[S]ome quoters are not satisfied with gaffemanship; nor do they regard themselves as bound by the customary rules of civilized discourse. They engage in what writer Milton Mayer once called "contextomy": cutting a statement out of context…in order to give a completely misleading impression of what some famous person believes.
I've never been able to track down where Mayer used this term, but perhaps it was in conversation (if anyone knows otherwise, please let me know). I've mostly followed Boller and George's use of the term, though I don't limit contextomies to famous people, since anyone who is quoted may be quoted out of context in a misleading way. In my usage, "contextomy" refers to such a quote, and not to the fallacy of arguing from such a quote. Instead, I say that a fallacious argument based on a contextomy commits the fallacy of quoting out of context.
You make a good point about the importance of context beyond the narrow issue of quotation. I call the more general fallacy of arguing based on only some of the evidence "one-sidedness" or "slanting", while others call it―more revealingly perhaps, if more awkwardly―"suppressed evidence" or "ignoring the counterevidence". It could also be called "contextomy", but I think that I'll continue to use that word in the narrower sense, for the sake of consistency.
As I mention at the end of its entry, one-sidedness is one of the most insidious and misleading forms of propaganda, since it may be impossible to tell that relevant information has been omitted, unless one has access to other sources of information. Moreover, it violates the logical requirement of total evidence in inductive reasoning, which is enshrined in the legal requirement of "the truth, the whole truth, and nothing but the truth". A one-sided account is not "the whole truth" even if, as in your example, it is "the truth" and "nothing but the truth". The worst type of bias is often not in what is said, but in what is unsaid.
Source: Paul F. Boller, Jr. & John George, They Never Said It: A Book of Fake Quotes, Misquotes, & Misleading Attributions (1989), p. viii.
Update (5/11/2011): Coincidentally, philosopher Julian Baggini has an article discussing a new book called Born Liars: Why We Canít Live Without Deceit by Ian Leslie. The article is actually from last month, but I just found it thanks to Arts and Letters Daily.
One thing that Baggini discusses is the legal formula "the truth, the whole truth, and nothing but the truth". When someone doesn't speak the whole truth, we call what they say a "half-truth". Now, a half-truth isn't half true, it's wholly true―it's in fact part of "the truth" that the formula bids us speak. Rather, a half-truth is less than the whole truth, that is, it is some fraction of the whole truth―probably not exactly half―that gives a false impression. Here's an example Baggini gives of this kind of "economy with the truth":
The art of describing a home for sale or let is only to say true things, while leaving out the crucial additional information that would put the truth in its ugly context. In other words, no "false statement made with the intention to deceive"―St Augustineís still unbeatable definition of a lie―but plenty of economy with the truth. … To say that the truth requires accuracy does not mean simply that everything you say must be 100 per cent correct, but that it must include all the relevant truths. … Accuracy requires us to say enough to gain an accurate picture; not telling lies only requires us to make sure what we do say is not false.
Nonetheless, a half-truth can be as misleading as a lie, and people frequently use them with the intent to deceive. Case in point:
…[O]ne of the most famous "lies" of recent decades is not a lie at all, but objectionable nonetheless: Bill Clintonís famous "I did not have sexual relations with that woman, Miss Lewinsky." As many people have pointed out, to a Southern Baptist, this could indeed be interpreted as being strictly true. "Sexual relations" is, in many parts, a euphemism for coitus, not any other sexual acts between two people. …[C]learly, he was not sincerely trying to convey the truth of his situation.
Baggini touches on the issue of whether it is ever possible to tell "the whole truth":
…[T]ruthfulness―the whole truth if you like―requires more than just true things being said, while acknowledging that there really is no such thing as "the whole truth" anyway. Full disclosure is never possible. Truthfulness is largely a matter of deciding what it is reasonable to withhold.
One might also wonder whether it is ever possible to meet the inductive requirement for total evidence. Do we ever have all the evidence? Probably not, but the point of the rule is that one should not withhold relevant evidence that one does have, that is, evidence that would affect the probability of a hypothesis. We can't include in our reasoning evidence that we don't have, but we should include what we do have. Of course, if we lack some evidence, then our reasoning may be affected for the worse, but that's just the nature of induction: there's no guarantee that you'll get the right answer. However, if you leave out evidence that you do have, then you're almost certain to get the wrong answer. It would be sheer good luck if you drew the right conclusion from incomplete evidence.
Finally, I disagree with Baggini's take on Steve Fuller:
…[P]erhaps the most interesting counter-example to the twin virtues of sincerity and accuracy was proposed by the sociologist Steve Fuller, who has been widely condemned for suggesting that intelligent design theory merits a hearing. Many of Fullerís colleagues know he is a smart guy and canít understand why he persists with this kind of argument. The answer is perhaps to be found in a piece he wrote in the spring 2008 edition of The Philosopherís Magazine explaining his modus operandi. The idea that one should always say what one truly believes is narcissistic nonsense, he argued. The role of the intellectual is to say what they think needs saying most at any given time in a debate, not to bear testimony to their deepest convictions. Although this might involve some dissembling, it serves the cause of establishing truth in the long run better than simply saying the truth as you see it. What matters is how what one says helps build and expand the widest, most expansive truth―not whether as a distinct ingredient it is more or less true than another.
First of all, how we are supposed to know, if Fuller is willing to be insincere, whether this is a sincere description of his method? Baggini continues:
I find Fullerís argument very persuasive. Indeed, it fits with my own tendency to want to talk more about the virtues of religion around atheists than with believers, or to question the value of philosophy with philosophers. The quest for truth requires a constant critical edge. In the case of intelligent design, I think Fuller is sharpening the wrong blade, and a dangerous one at that. But the idea that the contemporary consensus needs some shaking from its dogmatic slumber is not such a stupid one, and may justify a suspension of sincerity in the name of furthering debate.
I don't find Fuller's argument at all persuasive, either as a general method or in the particular case of intelligent design. Of course, it's perfectly fine to argue as the devil's advocate, so long as people realize that that's what you're up to, and so long as the cause that you're advocating for needs representation. I don't think that there's any danger that experts in the field of evolutionary theory will be misled by Fuller, but it's possible that laymen may take it as some evidence that there is real dispute among the experts over the truth of evolution.
However, "intelligent design" (ID) has been given a hearing, including back when it went under the name "creation science". The theory that life was designed by an intelligence is what the theory of evolution replaced. How many times does it have to be refuted? As I've pointed out previously, the arguments that ID advocates make are just stylistic variants of the "design argument" refuted by Hume centuries ago, prior to Darwin.
Creationism is one of those brainless monsters that are killed at the end of horror movies, but keep coming back in sequel after sequel after sequel….
Source: Julian Baggini, "The Whole Truth", Prospect Magazine, 4/20/2011
Mr Panetta also told the network that the US Navy Seals made the final decision to kill bin Laden rather than the president.
Source: Steven Swinford, "Osama bin Laden dead: Blackout during raid on bin Laden compound", The Telegraph, 5/4/2011
Via: Eugene Volokh, "Location, Location, Location", The Volokh Conspiracy, 5/5/2011
An Iffy Puzzle
The following iffy passage from a life insurance policy explains who should get the benefits when the insured person dies:
One-half to his mother, if living, if not to his father, and one-half to his mother-in-law, if living, if not to his mother, if living, if not to his father. … On the one-half payable to his mother, if living, if not to his father, he does not bring in his mother-in-law as the next payee to receive, although on the one-half to his mother-in-law, he does bring in the mother or father.
This is hard to understand, especially at first reading, but is it nonsense? Can you determine who gets what if the mother is living when the policy holder dies but the mother-in-law is not (no jokes, please!)? Also, how would you reword the policy so that it is easier to understand?
U.S. Vets Mock Nuclear Weapon
It was just an A-bomb. They wouldn't have dared mess with an H-bomb.
Source: Jonathan Chait, "Headline of the Day", The New Republic, 4/5/2011
Acknowledgment: Thanks to Michael Koplow for drawing this to my attention.
It's not nonsense. Here's a clearer, if longer-winded, way of expressing the meaning of the first sentence from the quote:
If the mother of the policy holder is living at the time of the holder's death, she shall receive one-half of the policy's benefits. If she is not living, then the holder's father will receive one-half of the benefits. If the holder's mother-in-law is living, she shall receive the other half of the benefits. If she is dead, then the holder's mother will receive the other half of the benefits. If the mother is also dead, then the father will receive the other half of the benefits.
You can check this for accuracy against the second sentence of the quote. Drawing upon this translation, you can then easily determine who shall receive what under each eventuality:
|Mother||Mother-in-Law||First Half||Other Half|
So, the mother gets everything if she is alive and the mother-in-law dead.
The first sentence of the quote is a conjunction of two simpler statements. One confusing aspect is that when it talks about "one-half" of the policy's benefits in the first conjunct―before the "and"―it's talking about a different half than in the second conjunct; that is, the first conjunct talks about what happens to the first half of the benefits, then the second conjunct goes on to explain what becomes of the remaining half. This is made clear in the second quoted sentence.
Here are the two conjuncts:
- One-half to his mother, if living, if not to his father.
- One-half to his mother-in-law, if living, if not to his mother, if living, if not to his father.
The second conjunct is the more complicated, so let's start with the first. The first conjunct has the form "q, if p, if not r", where the component statements are:
- q: The first half of the benefits will go to the mother of the policy holder.
- p: The mother is living at the time the policy holder dies.
- r: The first half will go to the father of the policy holder.
Let's call this type of statement an "if/if-not statement", and abbreviate it: [p,q,r]. Hopefully, a little thought will show that this statement form means the same thing as: "if p then q and if not-p then r". Let's try this on for size with the first conjunct: "If the policy holder's mother is alive at the time of his death then one-half of the benefits will go to her, and if she is not living then one-half of the benefits will go to the holder's father." This is a plausible interpretation of the conjunct, and it makes sense as a provision in a life insurance policy.
Given that if/if-not statements are equivalent to conjoined conditional statements, you can easily determine the truth table for the ternary if/if-not connective, which is the same as that for: (p → q) & (¬p → r)
You can solve the part of the puzzle pertaining to the first conjunct using this truth table as follows: first, eliminate every row in which the if/if-not statement is false, since we're only interested in those cases where the policy is followed. Also, we can eliminate any remaining lines in which both q and r are true, since it's not possible for the first half of the benefits to go to both the mother and father. This leaves only the second and penultimate rows, which tell us what happens when the mother is alive and when dead.
The second conjunct is actually two if/if-not statements, one nested inside the other: [p,q,[r,s,t]], where the letters stand for the simpler statements:
- q: The second half of the benefits will go to the mother-in-law of the policy holder.
- p: The mother-in-law is living at the time the policy holder dies.
- s: The second half will go to the holder's mother.
- r: The mother is living.
- t: The second half will go to the holder's father.
The remaining part of the puzzle can be solved with a truth table in a similar fashion to that of the first part. However, the complete table has 32 rows, so I'll leave it as an exercise for the reader. Oddly, the policy seems to make no provision for what happens if the father dies before the policy holder.
Source: Stuart Chase, "Gobbledygook" in Language Awareness (Second Edition), edited by Paul Eschholz, Alfred Rosa & Virginia Clark (1978), p. 41.
Update (5/5/2011): Readers who have some knowledge of computer programming may be familiar with what I have called an "if/if-not" statement―following the insurance policy language―as an "if-then-else" command. An "if p then q else r" command tells the computer to do q in case condition p is true and to do r in case p is false. For programming, this may work better than a binary conditional, which only tells the computer what to do if the condition is true and not when it's false. As a result, what the insurance company wrote may be more easily understood by a computer than a human being, which could have been its intent.
Resource: William J. Rapaport, "The Logic of the Ternary Sentential Connective 'If-Then-Else'" (1997). A short, mildly technical article in PDF format. If you can understand the above Technical Appendix, you can probably understand this article.