Monday, November 12, 2007

L/C Presentation

Fallibility and L/C Minds

I'm going to give a talk at our local philosophy club this coming Friday. I would once again like to bounce some ideas off of you folks and see what you think.

During the first half of the talk, I'll talk about fallibility. I'll start with some quotes from Thomas Jefferson, who mistrusted his own judgment when it differed from those around him. I'll talk about the fact that on huge issues like religion the world is highly divided signaling us that a large number of brains are wrong. The high divorce rate is another demonstration that the human brain is a faulty decision maker. Many people make low level decisions concerning job selection, car choice and home selection and are later full of regret. How is this possible?

I'll then review some of the sources of error in human thinking and decision making process such as bad statement of the problem faulty data in, incorrect modeling of the world and incorrect execution. In fact there are so many opportunities for error that one is surprised that we ever get anything right. We are fortunate that most choices are binary and we have a 50/50 chance of being right despite the brain's manifold inadequacies.

We might ask why we have this strange situation. Why is it that nature has screwed up so badly? I believe that the answer is simple. We are not using our brains for the purpose they were intended by nature. That's an anthropomorphic way of saying that our brains evolved to fulfill a life or death survival function, and we use them in ways that are mere "spandrels" or unintended consequences. I'll describe the importance of the brain as a tool to predict the intention of a potential mate, friend or adversary. I'll also discuss the value of the brain in the prediction of the movement and location of game, weather and plant life. All these things have direct implications for survival of the species. Great works of art, literature and philosophy have an effect on the survival of the very few creators involved, but little effect on the propagation of the dna of the creators throughout the population. Therefore, we should not expect a great ability for complex thinking and decision-making to be a common trait in our species. I'll split the audience into small groups and have them see if they can come up with an L/C Mind trait that is non-pejorative.

I'll then move into all this stuff about L-C Minds that we have previously discussed. My objective is to get people to think a little harder about the point of view of others i.e. to show respect. My definition of "to respect" is to act as though there is a possibility that you are wrong and your intellectual adversary is right, no matter how totally farfetched that might seem on the surface.

Any suggestions would be appreciated. With respect -Joel

25 comments:

Howard Pattee said...

For primitive survival it is clear that the choice of an action is not as important as the commitment to carry it out. For example, whether I choose to run or fight, my survival depends on how fast I run or how hard I fight. If there is any uncertainty in execution I lose. Isn’t this also the case for war and marriage?

Is it reasonable to say that C-minds tend to be better at commitment than L-minds? C-minds therefore appear stubborn to L-minds. On the other hand, maybe L-minds tend to think more about the possible choices, thereby appearing soft-headed to C-minds.

Ira Glickstein said...

Very insightful Howard!

I agree with you that L-minds appear soft-headed to me. They often seem to allow their (soft) heads to rule their minds, and change their previous decisions when the blood flows and reality shocks them.

C-minds, I admit, are definitely more stubborn, often to the point of denying some aspects of reality.

We have been IMHO properly socialized by the childhood fables of "The Little Engine that COULD" (and eventually DID!) or "The Little Red Hen" (who went on planting and harvesting and threshing and grinding and baking the wheat into bread even though no one would volunteer to help her - except for the last part, the eating of the bread), and so many more.

Ira Glickstein

joel said...

Howard said:

Is it reasonable to say that C-minds tend to be better at commitment than L-minds? C-minds therefore appear stubborn to L-minds. On the other hand, maybe L-minds tend to think more about the possible choices, thereby appearing soft-headed to C-minds.

Joel responds:

Thanks for the help. I think I'll bring up this in my talk and let the group work out the details. I think you and Ira have an important point. It's the impression of C-Minds that L-Minds make problems so complex that they become insolvable. On the other hand, L-Minds think that C-Mind solutions are simplistic, because they don't understand the problem. In fact, Ronald Reagan as governor of California remarked, "For many years now, you and I have been shushed like children and told there are no simple answers to the complex problems which are beyond our comprehension. Well, the truth is, there are simple answers, they're just are not easy ones." If we state this dichotomy in even-handed scientific or engineering terms, we might say that when modeling real systems, L/C minds differ in their assessment of the optimum balance of precision versus complexity.

At first, it might seem as though I'm missing Howard's point. However, let's return to Ethel and Fred discussing the marital problems of Desi and Lucy. Ethel sees all the stumbling blocks in her complex model of marriage. " Desi is constantly surrounded by show girls and groupies, Lucy is asleep by the time Desi gets home, etc. Lucy needs to be more forgiving". Fred says, "All that complexity doesn't matter, Desi vowed to be faithful. I don't respect a man who doesn't keep his word. Don't expect me to socialize with him, if Lucy doesn't." With respect -Joel

Stu Denenberg said...

Joel said:
"The high divorce rate is another demonstration that the human brain is a faulty decision maker. "

I think this may not be a strong example as a valid counterargument might be that like other animals, we are not suited for marriage to one mate for a lifetime and this just expresses that reality. But what do I know --- I'm happily married and so it suits me just fine.

The other point that was quoted by Reagan:
"For many years now, you and I have been shushed like children and told there are no simple answers to the complex problems which are beyond our comprehension. Well, the truth is, there are simple answers, they're just are not easy ones."

The counter for that one is the quote (I don't know who):


"For every complex problem there is a simple, elegant solution that is wrong."

Perversely (but always respectfully),

Stu

joel said...

Stu said that the high divorse rate may not be a good example of faulty decision making. Interestingly, I got the same criticism years ago from a liberal activist who I used to collaborate with on the L/C Mind concept. He was on his second wife. His reasoning was that he made the correct decision at the time he first married and then made the correct decision many yers later when he divorced and married another woman. That may be so. However, that doesn't change the fact that most very short-lived marriages are a consequence of poor choices. Perhaps another indication that we are bad decision makers as a species is the fact that we hardly ever follow Benjamine Franklyn's advise to write decision factors down on paper. Our minds are notoriously inept at focusing attention on more than a couple of things at a time and yet we refuse to draw up a balance sheet. We do cost benefit ratios in the workplace, but never at home for personal decisions. However, if you put us in the field with a shotgun, we are great at predictive shooting of skeet, allowing for windage or throwing a stone at a moving target. We successfully intuit friend and foe. I think those things demonstrate the real survival value and function of our big brains. By comparison music, art, philosophy, etc. must be secondary consequences of no survival signficance. They are mere spandrels despite the high value we place on them in the modern world. With respect- Joel

joel said...

I think the talk went quite well. The L/C approach seemed to allow people who are normally quite contentiois and polarized to discuss their differences without a lot of acrimony. Thanks for your help. With respect -Joel

Howard Pattee said...

Joel said (going back a bit): “The L-Mind thinks he can better understand the problem by examining the personal details. The C-Mind takes a step backward to see global consequences.”
In my experience, the opposite is the case with the issue of torture. L-minds that I know see torture as a global ethical atrocity that has nothing to do with the personal status of individuals. The only arguments supporting torture that I have uncovered from C-minds are the individual cases, like the “ticking bomb” scenario or particularly vicious terrorists.
All the Republican candidates, except McCain, support torture. None of the Democratic candidates support torture. In one 2006 survey by the Scripps Center at Ohio University, 66% of Americans who identified themselves as strongly Republican supported torture against 24% of those who identified themselves as strongly Democratic.
Do you know any arguments for torture other than for particular individual cases? That is, do you know any L or C arguments FOR torture as a global ethical policy?

Ira Glickstein said...

Is approval of torture a distinguishing mark between L- and C-minds?

I don't think so.

From Wikipedia:
Opinion of noted liberal professor

************************
Following the September 11, 2001 attacks, [Alan] Dershowitz [noted liberal Harvard professor] published an essay in the San Francisco Chronicle entitled "Want to Torture? Get a Warrant," in which he advocates the issuance of warrants permitting the torture of terrorism suspects if there were an "absolute need to obtain immediate information in order to save lives coupled with probable cause that the suspect had such information and is unwilling to reveal it."
**********************************

Thus, the "ticking bomb" argument for torture under extraordinary circumstances comes from a noted L-mind.

Absent the "ticking bomb" I don't think anyone, L- or C-mind would favor torture. That word brings to mind the Spanish Inquisition and Hollywood horror movies and so on.

The key is (paraphrasing Bill Clinton on the meaning of the word "is") "It depends upon what the meaning of 'torture' is."

I think the definition you find in "international law" is way too broad and inclusive.

I (fortunately) have no personal experience in this arena but I have read that things like long periods of annoying noises, cold cells, false information ("your confederate has confessed and implicated you"), "good cop - bad cop", and so on are effective in obtaining valid information from reluctant detainees.

Of course, harsh physical and mental treatment also yields false information, when the detainee tells his inquisitors what he thinks they want to hear.

When it comes to harsh treatment (short of torture) I think the L-mind thinks of how bad it would be to have that done to him or her, and "soft-heartedly" rejects any type of harsh treatment of detainees.

The C-mind is less squeamish and asks the biq questions. Does harsh treatment sometimes yield truthful, actionable intelligence? If those who are experienced in those areas say the evidence is positive, and if the C-mind thinks they are really professionals and experts and not just sadists or political operatives, etc., we tend to give them the benefit of the doubt.

Here is where I draw the line: The rogue soldiers at Abu Graibe prison were definitely not professionals or experts. They performed their harsh treatment just for the fun of it, and to take out their frustrations on people who happened to "look like" those who were blowing up their comrades (and some of the prisoners at Abu Graibe were the ones who were blowing up Americans.)

Those rogue soldiers were properly turned in by other soldiers, were tried and convicted and are in jail now. They accomplished absolutely nothing for us and hurt the US image very badly. (A C-mind would probably have made the punishment of the rogue soldiers worse than an L-mind.)

A friend of mine with battlefield experience in WWII told me that terrible things were done by US soldiers during the immediate aftermath of fierce battles. I don't reacall any well-known cases where those rogue soldiers were turned in or punished. If they were (and I hope they were) it never made it to newpapers or radio or even popular books.

Thus, the fact that rogue US soldiers in Vietnam and Iraq have been publicly punished is a great advance in military justice. I doubt any army or military or terrorist force has ever exposed and punished rogue soldiers as effectively as the US has in recent decades. (We still have along way to go, and I favor stricter regulations. The endemic presence of digital video and the Blogosphere are welcome means for exposing those who would dishonor our country by their rogue actions.)

Of course, someone will bring up "waterboarding". Unless you have personally experienced it, I don't think you have any basis for an opinion as to whether or not it is a justified technique that should be used in exceptional cases.

It is clear that, when properly done, it leaves no physical damage. It may leave severe mental damage, however.

I believe some Justice Department lawyer(s) volunteered for the treatment before they approved very strict rules as to its use. I am not about to volunteer to be waterboarded. However, as a C-mind, I am not so squeamish as to reject it out of hand, when authorized by competent officials and executed by competent personnel.

Ira

Howard Pattee said...

As Ira should know, one anecdotal case should not be taken as convincing statistics!

I repeat the statistical evidence:
All the Republican candidates, except McCain, support torture. None of the Democratic candidates support torture.

In one 2006 survey by the Scripps Center at Ohio University, 66% of Americans who identified themselves as strongly Republican supported torture against 24% of those who identified themselves as strongly Democratic.

Maybe you have better statistics that show otherwise.

Ira Glickstein said...

The only "anecdotal" evidence in my Comment is my friend's WWII battlefield experience. He told me, and I believe him, terrible things were done to German prisoners of war by some American soldiers with no punishment. He claimed to be an eye-witness. Do you really think there are more rogue soldiers under Bush than Roosevelt?

I found the link to the 2006 survey by the Scripps Center you refer to, with the exact set of questions: SURVEY: "SHOH34"

On their website, they describe their survey as follows [I have emphasized their typographical errors]:

*******************************
DESCRIPTION: Respondents were asked about perforamcne of president, attitudes toward Supreme Curt, whether they aproved or disapproved of spanking and of torture of suspected errorists. They also were asked about use of books, maazines and the Internet to getinformation on scioence, health and business. Respondents also ee asked if they had ever had food poisoning and on the portrayal of JEsus in "The DaVinci Code."
******************************

That is an extraordinary number of typos -- nine in a single paragraph. Right on on the front page of their website! The academics who did this survey may be "educated beyond their intelligence". If they did not correct simple errors on their website, do you trust them to word their questions fairly or analyze the data objectively? I do not.

Note also that they relate attitudes on spanking of children with "torture". See below for the exact wording of their questions where they equate "suspected terrorists" with Geneva Convention "prisoners of war". It appears this is a group of psychologists who don't like calling our efforts to curb terrorism a "war" but insist on giving terrorists all protections of prisoners of war. They probably consider any type of spanking child abuse. They clearly have an "agenda"!

Here are the two questions related to this topic that they asked. [Note the typos.]

*****************************
Question/VAR 11:

As you may know, Congress recently passed legislation giving President Bush authority to interpret the Geneva Conventions when ordering interrogation of suspected terrorists. The Gneva Conventions are international treaties tht ban the use of torture on prisoners of war. Do you thunk Congress did the right thing or the wrong thing is giving the President Bush this authority?*

Right thing 38%
Wrong thing 52%
Don't know 9%
Other 1%

Question/VAR 12:

Do you think the United States is sometimes justified in using torture to get information from a suspected terrorist, or is torture never justified?*

Sometimes justified 38%
Nevr justified 52%
Don't know 8%
Pthr 2%
*******************************

Would you buy a used car from these guys?

Ira

joel said...

Howard said:

Do you know any arguments for torture other than for particular individual cases? That is, do you know any L or C arguments FOR torture as a global ethical policy?

Joel responds: First let me apologize for the barely understandable spelling of my last few comments. I'm a two finger typist and haven't been able to use my spell checker recently. (The consciousness researcher Daniel Dennett has some interesting ideas about why we can't see our own typos.)

Howard's counter example is great. I think we can use it to refine the proposition. Remember that we started this discussion with Stu's notion that C-Minds are more judgmental (or "judgmenty" as he mimicked Steven Colbert). That is to say that C-Minds are more likely to make moral judgments concerning the behavior of other people near to them. I proposed that C-Minds make judgments concerning their personal relationships based upon generalization to all of society. Fred Murtz may wish he could go bowling with his good buddy Ricky Ricardo, but he holds both himself and Ricky to a higher standard.

We also need to recognize that C-Minds may argue in a L-Mind fashion and vice-versa in order to convince the other party. This is not necessarily the way they reason to themselves. A few months ago, I made a presentation concerning the nature of emotions. I asked my audience if they would personally torture a kidnapper who had hidden their grand-child in a place where they would soon die. The response was unanimously "yes" including the many L-Minds that I know within the group. As I distanced the proximity of the relationship, more and more of the L-Minds dropped out of the "yes" column.


I think that you have heard L-Minds argue against torture the following highly debatable generalizations:

The Constitution prohibits it. The Geneva Convention prohibits it. We protect out troops from an enemy who would feel justified in using the same tactic. The information obtained by torture is unreliable.

Similarly, C-Minds may argue that there is nothing which protects foreign combatants in the constitution. The Geneva Convention applies to non-signatories, but not to those who don't follow the rule of wearing uniforms or other designation. (It's worth reading the Convention, but prepare to be very, very bored.) Another general argument for torture would be that justifiable homicide in many states extends to protecting the life of others.

However, Howard is absolutely right that my defining principle separating L-Minds and C-Minds is faulty. I'm in the process of revision. With respect -Joel

joel said...

Thanks Ira for that excellent illustration of how statistical evidence can be misleading or downright phoney. The bias of question makers is so often a factor in such results. We see how easily even someone as intelligent as Howard can be fooled, since most of us are are unwilling or unable to research the nature of the statistical study (especially if it's a meta-study). I'm much more in favor of so-called "anecdotal" evidence. Most discoveries in physics, chemistry and engineering have been initiated by anecdotal evidence not statistical evidence. The odd case that doesn't quite fit the prediction, results in the search for a cause. Anecdotal evidence is not final proof, but it must be explained none-the-less. Statistical evidence wears the guise of truth and is more likely to be accepted uncritically. With respect -Joel

Ira Glickstein said...

Marvelous example, Joel!

"...I asked my audience if they would personally torture a kidnapper who had hidden their grand-child in a place where they would soon die. The response was unanimously "yes" including the many L-Minds that I know within the group."

I think that simple example fatally punctures the "no torture, never, ever" position.

I'd appreciate it if one or more of the L-minds on the Blog would comment. Assume your granddaughter was taken by a gunman. The police and media are alerted and you and hundreds of volunteers are searching for her. You spot a car and driver matching the description and notice a shovel with fresh dirt plus some of your granddaughters clothes in the car. You and other volunteers detain him and he tells you the girl is buried in a box with only hours of oxygen.

What should you do?

1) Turn him over to the police? (They can question him but cannot legally apply harsh treatment.)

2) Reason with him? (Tell him searchers will eventually find her, perhaps in time, perhaps not. If she suffocates he will be punished for murder. If he reveals her location and she survives, it will be less serious for him.)

3) If he does not cooperate, take matters into your own hands and hurt him until he takes you to the burial spot. If he remains uncooperative, poke out an eye and give him an ultimatim regarding the other eye.

This is a personalized version of the "ticking bomb" that Alan Dershowitz forthrightly accepted as a reason the law should allow a court to give permission to torture in extraordinary cases.

I hasten to add that I would not permit use of torture except in a "ticking bomb" situation.

On the other hand, I would certainly allow experienced, professional intelligence officers, with the close oversight of superior officers, to apply harsh treatment to terrorists captured on the battlefield or red-handed in the midst of a plot. The harsh treatment would include cold cells, annoying noises, "truth serum" type drugs, threats, and so on. Also, in extreme cases where the information could save many lives, waterboarding.

As for the excuse that harsh treatment sometimes yields false information, in most cases the information can be checked. Going back to joel's example, if the kidnapper provides the true location you will find your buried granddaughter. In the case of a terrorist, if he gives true information, and you raid a "safe house", you will find confirmatory evidence, etc.

Ira Glickstein

Howard Pattee said...

I thought we had already agreed that C-minds tend to be tougher-minded and more committed (stubborn) than L-minds, so it is only reasonable that they would tend to support more severe enhanced interrogation techniques than L-minds. (Ira now apparently disagrees that this is a L-C difference.)
In any case, my point was meant as a counterexample to Joel’s generalization. In this case it is C-minds that have supported their view by individual cases while the L-minds support their view by global and moral arguments. I did not argue that either type of behavior should be absolutely prohibited. In fact, I would not be surprised (although we will never know) if in an actual potentially disastrous crisis both L- and C-minds would act about the same way — Hillary might act as tough as Rudy!

Howard Pattee said...

You guys are hard to keep up with. I find I have to go back before my last comment. I am not over Thanksgiving yet. I hope you all had a happy Thanksgiving and have recovered.

Joel said, “I'm much more in favor of so-called "anecdotal" evidence. Most discoveries in physics, chemistry and engineering have been initiated by anecdotal evidence not statistical evidence.”

That is certainly the case in the hard sciences. On the other hand, when we try to make distinctions about the behavior of POPULATIONS, like L and C minds, these are general statements that can only be verified statistically, otherwise there is no basis for their generality.
Also, by definition (Webster’s New World Collegiate Dictionary), an anecdotal statement is one that is “unverified by controlled experiment” and should be excluded from the controlled statistics whether or not it fits the generalization. Anecdotal events do indeed initiate entirely new theories, as has occurred in many sciences, but they still need controlled experimental verification to be accepted.

Howard Pattee said...

And while I’m at it, before eating my turkey soup, I have to respond once more. I also appreciate Ira’s skeptical research, but typos and bad questions are not necessarily evidence of bad data.

In my defense, here is the reference I used to the Scripps Survey SHOH34. Note: the last sentence is the only one relevant to my post. Ignore the rest. Here it is:

“Not surprisingly, the poll found that Democrats overwhelmingly opposed harsh treatment of detainees and Republicans overwhelmingly supported it.”

That conclusion is entirely consistent with the Republican and Democratic candidates views that was my primary evidence. I have no evidence or reason from Ira’s concern about typos or his sarcastic comments to doubt that sentence. In fact, I thought most of us would agree with it. Joel thinks I was easily fooled, which implies that the sentence was misleading or incorrect. Can anyone supply contrary statistical results from other polls?

From http://newspolls.org/story.php?story_id=59
HEADLINE: "Support for torture is linked to attitudes on spanking"
WRITER/REPORTER(s): Thomas Hargrove and Guido H. Stempel III
SOURCE: Scripps Howard News Service
DATE: October 31, 2006

BODY: Americans who say it's acceptable to spank children are twice as likely to support the use of torture on suspected terrorists than are people who oppose spanking, a new poll has found.

The survey of 1,031 adult residents of the United States conducted by Scripps Howard News Service and Ohio University found that a narrow majority (52 percent) believes that "using torture to get information from a suspected terrorist" is "never justified."

Thirty-eight percent of Americans say torture is "sometimes justified" and believe Congress acted correctly in giving Bush broad authority in how to interrogate detainees in the war on terrorism. Ten percent in both questions were undecided or gave other responses like, "It depends what the facts are."

Not surprisingly, the poll found that Democrats overwhelmingly opposed harsh treatment of detainees and Republicans overwhelmingly supported it.

joel said...

The cited poll seems to suffer from all of the evils that Ira pointed out in his posting Lies, Damned Lies and Statistics. The Scripps suggested headline "Support for torture is linked to attitudes on spanking" implies causation when there is no such evidence. More Republicans than Democrats support the phonics method of teaching reading. Would an appropriate headline be "Support for torture linked to support for phonics"? The poll also shows a preconceived expectation when it says "Not surprisingly the poll found...." Finally there is a total failure in the wording of the question. I'm in favor of "torturing" terrorists to obtain certain kinds of information in certain situations. However I'm not in favor of torturing "suspected" terrorists as the survey question states. The word "suspected" throws everything off. With respect -Joel

Ira Glickstein said...

Howard, I really appreciate your involvement in this thread, as evidenced by three Comments in a row while working through your Thanksgiving turkey leftovers. Fortunately, we had our turkey at a friend's place and so have no leftovers.

You wrote:

*****************************
I thought we had already agreed that C-minds tend to be tougher-minded and more committed (stubborn) than L-minds, so it is only reasonable that they would tend to support more severe enhanced interrogation techniques than L-minds. (Ira now apparently disagrees that this is a L-C difference.)
********************************

Sorry I left a misunderstanding of my views. I agree there is a L-C-mind difference on "severe interrogation techniques" which I earlier refered to as "harsh treatment" (but NOT torture) of suspected terrorists. Clearly, L-minds are less likely to suppport harsh treatment.

The cause of the misunderstanding is the confusion of "torture" and "harsh treatment". You (and the Scripps folks) and, I believe, most L-minds do not distinguish between the two terms.

International Law prohibits harsh treatment of prisoners of war, usually defined as uniformed soldiers of an established state. Spies and other non-uniformed actors are not covered by the Geneva convention.

As I wrote earlier, I distinguish between torture and harsh treatment and between POWs and terrorists. For me, torture is permanent bodily damage and POWs are uniformed soldiers of an established state.

According to Joel, both L- and C-minds support torture in the "ticking bomb" scenario. Noted liberal Alan Dershowitz's essay supporting torture is evidence that an influential L-mind agrees with C-minds in this narrow case.

In your first Comment in this sub-thread, you used the word "torture": "All the Republican candidates, except McCain, support torture. None of the Democratic candidates support torture."

In your more recent Comment you use the term "severe interrogation techniques" which I take to mean methods that do no permanent bodily harm. Your quote from the Scripps study also uses the later term. On the other hand, the Scripps headline uses the word "torture" as you did in your first Comment.

Bottom line: If torture is understood as permanent bodily harm both C- and L-minds support it only for the "ticking bomb".

C-minds are more likely to favor "harsh treatment" and "severe interrogation" of suspected terrorists. L-minds more likely to treat terrorists the same as prisoners of war who were wearing the uniform of an established nation.

As for the typo issue, that website page has been up for a year. Their failure to notice and correct these clear and simple errors leads me to doubt their ability to keep their statistical correlations straight.

Scripps plays fast and loose with the distinction between torture and harsh treatment, first (question 12) asking about Congressional authoriation to "interpret the Geneva Conventions when ordering interrogation of suspected terrorists" and in the next (question 13) using the word "torture". This is a well-known technique in "push-polls" which are designed to change respondents opinions and create false statistical results.

Ira Glickstein

Howard Pattee said...

Ira and Joel don’t trust that Ohio poll. So here is another. It reaches the same conclusion and about the same numbers (Margin of error in both was ± 3%). If you read the debates you will see a lot of waffling, but I would say the statement still holds that the Democrats’ attitude on torture are clearly distinguishable from the Republicans’ attitude.
I’m not convinced of the sincerity of any of these candidates. Believe me, both of them pay attention to these, and many other private polls, and that is what they are expressing to get votes.

Source: Pew Research Center for the People and the Press
Religion and Torture: A View from the Polls June 01, 2007 by John C. Green
Excerpt:
“The largest influence on views of torture was not religion, but political views:
Not surprisingly, party identification was strongly associated with views on torture. For example, 66.8 percent of Republicans held permissive views on torture, while 66.4 percent of Democrats had restrictive views. The independents were arranged in-between, but with a solid majority of the “pure” independents holding restrictive views. A similar pattern held for ideology: 59.0 percent of respondents who said they were “very conservative” reported permissive views, and 66.4 percent of those who said they were “very liberal” had the opposite position on torture. Here, too, a majority of moderates had restrictive views.”

For complete article see http://cfia.org/ArticlesAndReports/ArticlesDetail.aspx?id=6196&hId=3672
Relevant data to my point is at p. 25 of PDF file

Ira Glickstein said...

Thanks Howard for adding another poll on this topic.

The Pew poll asked if torture of suspected terrorists was justified "often", "sometimes", "rarely", "never", or "don't know". OK, sounds fair.

Then, inexplicably, they combined the first two categories and called them "permissive" and the second two and called them "restrictive" on torture.

Nope! The "rarely" group favors torture in some cases (e.g., "ticking bomb") and therefore belongs with the "permissive" set.

If you combine the "often", "sometimes", and "rarely" it turns out that about 64% of the total population favor torture in at least some cases. Only 32% of the total population says "never" while 3%"don't know".

If you divide them by race and religion, only 27% to 34% of white and black Protestants and white Catholics as well as persons not affilliated with any religion say "never" to torture.

The only defined group with substantially more distaste for torture (43%) are weekly worshipers who are not Protestant or Catholic (presumably Jews and Muslims and other small religions?)

The data in the document about Republicans and Democrats is based on the IMHO faulty combining of "rarely" with "never".

Do you have unequivocal quotes from all Democratic Presidential candidates (and Sen. McCain) that they would never, ever, ever torture a terrorist, even with a "ticking bomb"?

Do you have unequivocal quotes from all Republican Presidential candidates that they would favor torture absent a "ticking bomb"?

If not, I do not think it is fair to say "All the Republican candidates, except McCain, support torture. None of the Democratic candidates support torture."

Ira Glickstein

Howard Pattee said...

I don’t know how I got in the position of defending polls. I agree with Joel that polls should be avoided (even forbidden in political campaigns). As Ira has pointed out, polls can be misleading as well as misinterpreted. But just to be fair to polls, some of Ira’s arguments trying to discredit the polls strike me as faulty. Here’s what I think:
Case 1. Any characterization of L- and C-minds is necessarily about summary descriptions of population distributions. Ira is better at statistics than I am, so he knows that in a population distribution there are always individual cases (esp. the “tails” of the curve) that when described in detail do not agree with a correct statistical assertion about the population. In fact, in some cases no individual in the population will have the statistical property derived from the distribution. For example, the average height of 10 men might be 6 ft. but none of them are this height, nor can you refute this statistical result by pointing out that one of the men is only 4 ft. tall.
For the same reason, a vague statistical statement like “Liberals tend to opposed harsh treatment and Conservatives tend to supported it,” is not refuted when Ira points out (twice) that one maverick liberal, like Dershowitz, approves harsh treatment. In fact, Dershowitz could be a member of the statistical sample. The point is that in general, one apparent “bad fit” in no way invalidates the statistics, and it may be important for the result.
Case 2. Statistics are necessary when you do not have detailed information. Opinion poll questions usually do not contain much information and consequently they require tacit statistical thinking. To even think about a vague question like, “Do you condone torture?” a fair “yes” or “no” answer requires imagining a typical torture situation. As Ira and Joel have illustrated, whatever the answer, it is possible to elicit the opposite answer by defining a detailed special case of torture with much more detailed information. Such detailed situations are typically the tails of the normal distribution of situations. The point is that the individual “ticking bomb” cases, like Dershowitz and 4 ft. men, do not invalidate the statistical results based on a population distribution.
Case 3. Ira asked, “Do you have unequivocal quotes from all Democratic Presidential candidates (and Sen. McCain) that they would never, ever, ever torture a terrorist, even with a "ticking bomb"? This is a good example of petitio principii (begging the question) because the answer is obvious from the way the question is stated. Such an obvious “forced” answer contains no significant information and therefore neither supports nor refutes any general statistical conclusion.
Case 4. Ira said, “The cause of the misunderstanding is the confusion of "torture" and "harsh treatment". International Law prohibits harsh treatment of prisoners of war, usually defined as uniformed soldiers of an established state. Spies and other non-uniformed actors are not covered by the Geneva convention.” Ira says these polls do not make this and other distinctions. That is true because the purpose of these polls is to uncover a population’s moods and their moral, or emotional responses. Legal analyses are irrelevant. Pollsters are hired to find out the simple gut-level attitudes of average voters so candidates can better pitch their simplistic speeches to win votes.
Ira’s distinctions are important, but they are in the domain of the law. The lawyer’s job is entirely different from the pollster’s job. Cheney and Bush wanted the authority to torture prisoners without a trial, and it was the job of Yoo, Gonzales, and Addington to find the legal arguments to do so. This required many distinctions like Ira’s, and complex arguments that have (or will) end up in Supreme Court split decisions. Emotion responses and ethics have little to do with these legalities.
Pollsters also understand that the average voter doesn’t know what the Geneva Convention says, what a “non-uniformed actor” is, or how you legally define war ― and doesn’t care.
I said earlier, I would not be surprised (although we will never know) if both L- and C-minds would act about the same way in an actual potentially disastrous crisis. That is because in an actual situation the deciders presumably have much more detailed information to make an informed decision than is supplied by any simple poll-based statistics. Unfortunately, polls are used to get elected, but a president that uses a poll to make a decision is incompetent.

Ira Glickstein said...

We agree on most things.

I TOTALLY AGREE WITH THESE ITEMS (In Howard's words):

1) "... C-minds tend to be tougher-minded and more committed (stubborn) than L-minds, so it is only reasonable that they would tend to support more severe enhanced interrogation techniques than L-minds."

2) "Not surprisingly, the [Scripps]poll found that Democrats overwhelmingly opposed harsh treatment of detainees and Republicans overwhelmingly supported it."

3) "...As Ira has pointed out, polls can be misleading as well as misinterpreted."

4) "...In fact, I would not be surprised (although we will never know) if in an actual potentially disastrous crisis both L- and C-minds would act about the same way — Hillary might act as tough as Rudy!"

5) "...I said earlier, I would not be surprised (although we will never know) if both L- and C-minds would act about the same way in an actual potentially disastrous crisis. That is because in an actual situation the deciders presumably have much more detailed information to make an informed decision than is supplied by any simple poll-based statistics."

SO WHERE DO WE DISAGREE ?

I make a distinction between "torture" (permanent physical bodily damage) and "harsh treatment" or "enhanced interrogation".

Someone may say it is "torture" to listen to his young son learning to play his violin, but that is as far from what we mean by torture of terrorists as war with guns is from the "war" on poverty!

I was also suprised when Howard wrote: "...I agree with Joel that polls should be avoided (even forbidden in political campaigns)."

This is a free speech issue and I have to come down on the free speech side! If people agree to answer questions in a public area, or by phone or Internet, and if these opinions are tabulated and reported, I can see no Constitutional way to stop it.

It is true that "exit polls" have, when reported by mainstream media while an election was still in progress, distorted subsequent voting. I am pleased the media have voluntarily agreed to withhold results till the polls close. However, if some media chose to violate that agreement, I would not want the government to interfere.

I hate "push polls" used by organizations to change opinions. However, I see no Constitutional way to prohibit them.

I would support a law requiring the phone company and Internet providers and the media to identify the actual organizations or people conducting polls. That way, distorted questions and false tabulations could be traced back. Particular politicians and activists would be discredited. Perhaps the errant pollsters could then be prosecuted for libel or some other crime.

I also support "do not call" lists that would apply to political organizations as well as commercial interests.

However, I would not prohibit polls per se.

Ira Glickstein

Howard Pattee said...

I agree we can't stop polls, but they are clearly a trend that badly weakens the actual voting process. Popularity itself has become a perverse value in the US. Look at the celebs that dominate the news media (Why? Because they are famous!). If people come to believe the polls many will vote like the polls.

joel said...

Ira would not prohibit polls. Howard agrees. They both think polls are a bad thing for the public. I don't see how regulation of pollsters violates the First Amendment any more than truth in advertising does. However, I doubt that any regulation is politically possible.

Therefore I suggest that education may be the answer. Florida's Sunshine State Standards already require that math curriculums teach statistics. Rather than just mathematical manipulations, a study of statistics should contain a healthy dose of Ira's "Lies, Damned Lies and Statistics" or its equivalent. When the general population develops a healthy mistrust of statistics, the industry may self regulate to instill confidence again. With respect -Joel

Ira Glickstein said...

Polls should not be prohibited but they could be regulated in a manner similar to "truth in advertising" as Joel suggested. See Federal Trade Commission for info on "truth in advertising".

In general, I favor minimal government involvement unless and until someone, such as a competitor or consumer or activist organization brings an action against an orgnization they think has violated the standards. Then, the FTC should investigate and pursue the case if it has merit.

In the case of political polls, the actions would be initiated by opposition political organizations or activist groups or media organizations that judge a poll to be biased, such as a "push-poll", or tabulated in a misleading way, etc. Perhaps polling organizations should be required to also publish the raw data collected, including the data that they may have discarded, as well as the demographics of the polling subjects, etc., etc.

However, if an organization conducts a poll -- no matter how biased the questions or how tilted the selection of those to be polled -- and makes the exact questions and raw data available, I do not think the government should get involved, except to check that the data provided in the actual data.

Publicity is the best solution! For example, it just came out that CNN, in selecting the questions used for the recent Republican debate, picked quesitons from several people who were well-known Democratic activists, despite CNN's intention to make this a Republican debate for Republicans.

I understand that CNN appologized and eliminated those questions when rebroadcasting the debate. However, the publicity makes people aware of CNN's carelessnes (or bias - take your pick).

That type of publicity is, IMHO, sufficient correction for their error.

I do not want to see the government get into a position where they have to micromanage pollsters. Let the free market expose the (alleged) errors and bias and let the public decide.

As for educating our children as to the possible misuse of statistics, I agree with Joel that needs to be done. If there are any eucators out there, I offer my Blog series on "Lies, Damned Lies, and Statistics" to be used in your classrooms free of charge!

Ira Glickstein