A common logical error I observe on DKos is the use of boolean "black-and-white" thinking as opposed to arithmetic "shades of gray" thinking. A detailed explanation of how the problem often manifests itself follows below the fold.
The word "arithmetic" in the title is an adjective, pronounced "air-ith-MET-ic", and it refers to thought processes that rely on numbers rather than simple black-and-white evaluations. This does not necessarily require the use of mathematics; indeed, in most cases, it involves nothing more than noting the difference between two numbers.
I have noticed that some people here on DKos are handicapped by an inability to apply arithmetic thinking to their political analyses. This often yields silly results.
Here's a fictional case: the group is discussing energy policy and comparing different sources of energy. Mr. Boolean declares that Source A is unacceptable because it emits nastious oxide. He further declares that Source B also emits nastious oxide, and is therefore just as bad as Source A.
Mr. Arithmetic argues that Source B emits only one-hundredth as much nastious oxide as Source A, and so is nowhere near as bad as Source A, and nastious oxide in small quantities is not so dangerous.
Mr. Boolean rejects Mr. Arithmetic's argument because nastious oxide is just bad, bad, bad and no amount of nastious oxide is acceptable.
That's boolean thinking. Presented this way, I'm sure that you'll think that nobody could be as dumb as Mr. Boolean. I assure you, I see this kind of thinking every day on DKos.
Another example is polarization of non-polarized values. Here's an example, not fictional but taken from the pages of history:
The 2000 election, you will recall, was extremely close: Mr. Gore got more popular votes but Mr. Bush got more electoral college votes because he won Florida by a few hundred votes. However, there was a third candidate: Mr. Nader. Thousands of progressives chose to vote for Mr. Nader instead of Mr. Gore. They did so because they felt that Mr. Gore was not progressive enough for their taste. This was due to a boolean error on their part. They evaluated candidates in black and white terms: Mr. Bush was black, Mr. Gore was black, and Mr. Nader was white. They denied that there was any significant difference between Mr. Bush and Mr. Gore, when in fact there were many substantial differences between the two candidates.
However, had they used arithmetic thinking, they would have seen Mr. Bush as dark grey, Mr. Gore as middle-gray, and Mr. Nader as light grey. The realization that Mr. Gore was more desirable than Mr. Bush, combined with the certainty that Mr. Nader could not win and that Mr. Bush could win should have made the decision to vote for Mr. Gore obvious -- but not to people who think in terms of black-and-white. Those votes for Mr. Nader were a small contributing factor in the invasion of Iraq, the deaths of several hundred thousand Iraqis, nearly two trillion dollars in overall economic costs, the loss of American prestige, the reduction of financial regulation leading to the economic crisis and the deepest recession since the Depression, millions of people losing their jobs -- that's what happens when you don't think clearly.
Here's another actual example: the stupid debate about "nature versus nurture". Is human behavior controlled by our genes or our upbringing? People have wasted enormous amounts of energy arguing this question, when the problem lies in the black-and-white nature of the question itself. Human behavior is not controlled by any single factor; it is influenced by many factors. When you think of it in black-and-white, either-or terms, you get stuck in a fruitless argument. When you start thinking arithmetically, of which behaviors are influenced to what degree by which factors, then you start making some progress.
I will close with what is in my opinion the single most destructive case of boolean thinking in American political discussion these days: health care. The mistake lies in this dictum:
"Nobody should die because of lack of medical care."
At first glance, it looks like a reasonable statement. But its danger lies in its black-and-white representation of a problem that really has shades of gray. For example, consider a hypothetical 90-year old man with a bad case of cancer. It is certainly conceivable that we could extend his life by a year or two by devoting, say, $100 million worth of health care resources to the task. Denying him such resources would violate our dictum above -- he would die due to lack of medical care. Yet we can all agree that it would absurd to spend that kind of money to extend his life by a year or two.
The real question is not boolean but arithmetic. It's not a black-and-white issue of whether or not we should provide health care, but a greyscale issue of how much health care we should provide. This forces us to ask the following uncomfortable question:
"How much money should we spend to save a person's life?"
What makes it even more uncomfortable is the fact that different people deserve different amounts of expenditure, because they have different life expectancies. Why spend as much money on a senile person as we spend on a teenager? Other considerations complicate the matter even further.
Yes, such questions are considerably more difficult to answer than the simple black-and-white question. Yes, answering them requires us to make painful decisions. But our refusal to replace boolean thinking (as illustrated in the dictum above) with arithmetic thinking is a major factor in the exploding cost of health care.
Lastly, some upbeat observations: I don't think that boolean thinking is rampant here on DKos; it crops up occasionally, not frequently. Moreover, this kind of error still beats all heck out of the simple denial of reality that we observe so often among members of farther right.