Wikipedia:Reference desk/Archives/Humanities/2009 November 6

Welcome to the Wikipedia Humanities Reference Desk Archives
The page you are currently viewing is an archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.

November 6

Picture of Thomas Pynchon

Go to this article and scroll down to the picture next to #2. Is that Thomas Pynchon, and if so, what's the source of that photo? zafiroblue05 | Talk 06:16, 6 November 2009 (UTC)

It is quite hard to compare pictures accross a person's life, and Pynchon is a noted recluse; not of the level of J. D. Salinger, but he's not exactly a "public" person by any means. He did several famous cameos in The Simpsons; and while it was his voice being used, his character in the show was shown with a bag over his head, playing on his famous avoidance of the media, somewhat ironically as he lends his voice to the most famous animated series, like, ever.

Most extant confirmed pictures of Pynchon are from his youth, like from High School yearbooks or his US Navy portrait. See this Google Image search. Since most of the confirmed pics of him are probably over 50 years old, its hard to say if that pic you cite is definately him. He didn't have a beard in his youth, for example, and while his youthful pictures seem to show a distinctive gap in his teeth when he smiles, any more recent dental work may obscure this in a modern picture. The picture you found actually seems to more closely resemble Gary Snyder, a friend of Pynchon, see this image and article, page down till you see the image caption. It looks a lot like the image you found. Indeed, searching google, I can't find anything more modern than 50 years ago. --Jayron32 06:56, 6 November 2009 (UTC)

Actually, looking closer, that picture you found is almost certainly Gary Snyder. See this Google image search for Gary Snyder. I would be 99% certain the pic in that article next to Pynchon's name is Snyder and not Pynchon. --Jayron32 06:59, 6 November 2009 (UTC)

Additionally, the person who wrote the article you cite obviously doesn't know what he is talking about, since he states that the Pynchon novel V. was the source for the V miniseries and subsequent sequels, which it obviously wasn't. --Jayron32 07:03, 6 November 2009 (UTC)

Yeah, the page also says that Lolita is a Russian novel and that Kafka (whose works were largely unknown until after his death, whereas the Nobel Prize is given only to living authors) should have won one. Clearly not a person to be trusted. Deor (talk) 14:27, 6 November 2009 (UTC)

It also makes no mention of Henry James being an American, seems to think Joseph Conrad was Ukrainian, spells Kerouac Karouac, and puts an apostrophe in Finnegans Wake. In short, it's not to be trusted. 209.251.196.62 (talk) 15:40, 6 November 2009 (UTC)

It also makes the uncorroborated and highly suspicious claim that Salman Rushdie didn't earn the Nobel Prize due to the controversy he stirred with the publication of The Satanic Verses. Actually, I always thought he earned more literary prizes due to the book. Quest09 (talk) 16:06, 6 November 2009 (UTC)

I think the link provided in the OP is a perfect example of why one should never take anything on the Internet at face value. And that one should always verify such information using one or more other sources (like for example this great ref desk right here). --Saddhiyama (talk) 16:27, 6 November 2009 (UTC)

. Comet Tuttle (talk) 21:02, 6 November 2009 (UTC)

Name that magazine article

I'm trying to remember the name of a certain article quite possibly from Esquire magazine. It's a critique of the sensationalist "Pedophile-hunt" TV shows such as "To Catch a predator". The most prominent feature was a picture of a like sting in a snowy woods with wolves standing in for the production crew. Can anybody name the article or at least the magazine? Thanks. Spade9 (talk) —Preceding undated comment added 07:05, 6 November 2009 (UTC).

cult religion

problems posed by cult religion inthe society —Preceding unsigned comment added by 41.204.168.3 (talk) 10:35, 6 November 2009 (UTC)

This seems like a homework assignment, which we do not do for you. Check out Cult and Cult (religious practice) to start. Falconus^{p t c} 12:18, 6 November 2009 (UTC)

Also, be more specific. Your question (?) is put very broadly, and it's hard to figure out exactly what it is you want to know. TomorrowTime (talk) 12:25, 6 November 2009 (UTC)

Sounds like a subject for a seminar work. Too broad to be answered I guess....But you can start with the Falun Gong and continue to scientology.--Gilisa (talk) 12:46, 6 November 2009 (UTC)

The problems are going to depend greatly on the specific cult and the specific society. You have not provided either of those pieces of information to us, so we will be unable to be much help I am afraid. Googlemeister (talk) 14:36, 6 November 2009 (UTC)

For example the definition of cult can also be a political thing, for example in the case of Falun Gong, according to the United States State Department, in China, whether or not a group is classified as a cult depends on the Chinese authorities and is "based on no discernible criteria other than the Government’s desire to maintain control." [1] --HappyInGeneral (talk) 14:41, 6 November 2009 (UTC)

Cult suicide (Jim Jones, Order of the Solar Temple); cult murder (Charles Manson's followers); cult biological and chemical terrorist attacks (Aum Shinrikyo, Rajneesh movement), prostitution (Flirty fishing), odd financial dealings (International Society for Krishna Consciousness. From the opposite side, moral panic. Rmhermen (talk) 15:06, 6 November 2009 (UTC)

insufficient problem posed question context answer difficult Dmcq (talk) 20:47, 6 November 2009 (UTC)

On the plus side, cults do tend to increase the sales of athletic shoes and Kool-Aid. :-) StuRat (talk) 21:28, 6 November 2009 (UTC)

US military deployment policy

Due to the latest massacre at Fort Hood, we're forced to look again at the US policy of deploying people to fight against those of their same religion and/or ethnicity. Has the military ever had, or at least considered, a policy of transferring/deploying such people into other regions, such as Japan or South Korea, where there will be no question of mixed loyalties ? An exception might be made for linguists, as a shortage of those who speak the native language but are not of that culture is likely to exist. Perhaps they could work from the US via teleconferencing or, if necessary to be deployed in the field, they could be unarmed (basically like Iraqi or Afghan nationals who work as translators).

So, I would like any info or links on such a policy discussion. Thanks. StuRat (talk) 14:35, 6 November 2009 (UTC)

I know they certainly had such a policy in WW2, where those of Japanese ancestry were only permitted to fight against the Nazis and the Italians (see 442nd Infantry Regiment (United States)). Now it would probably be labeled as racism or something. Googlemeister (talk) 14:38, 6 November 2009 (UTC)

Regarding the Fort Hood incident, I think a combat stress psycologist would be exactly the kind of specialty that would be required to deploy, just like a linguist. Rmhermen (talk) 14:59, 6 November 2009 (UTC)

Yes, I'm sure they need combat stress psychologists, but there's no reason to think they can't find a combat stress psychologist who isn't of the same religion and/or ethnicity as the people we are fighting, thus creating the problem of divided loyalties. StuRat (talk) 20:37, 6 November 2009 (UTC)

The U.S. military has not allowed soldiers to choose which wars or which enemies they would fight. Edison (talk) 16:11, 6 November 2009 (UTC)

I'm not suggesting that they should have the choice, the military should refuse to assign them to locations where divided loyalties are likely to become an issue. StuRat (talk) 20:41, 6 November 2009 (UTC)

Seriously, why should it make any difference. Assuming the U.S. military has done sufficient background checks, why would it matter about the religion or ethnicity of its soldiers. The U.S. has no native ethnicity (unless you count Native Americans, but they really haven't ever been ethnically "american" either, just that they were native to the land that later BECAME the U.S.); all Americans are essentially imigrants. German-Americans most certainly fought against Germany in WWII; seeing as they are the largest single ethnicity in America it would have been impractical to ban them from fighting on that front. The whole "banning the Japanese Americans from fighting on the Japanese front" was part of the same Xenophobic racism that caused the creation of Japanese American internment camps. The fact remains that there was never an effort to create a German-American internment camp because Germans don't look any different than other white people, Japanese do. It is nothing besides a coincidence that the nutjob who shot all those people at Fort Hood was a muslim or of muslim decent. General John Abizaid, former commander of CENTCOM was of arabic decent; I have no idea what his actual religion was, but no one doubted HIS ability to lead in Iraq and Afghanistan. --Jayron32 17:13, 6 November 2009 (UTC)

I found that John Abizaid's father and grandparents were Lebanese Christian and he has appeared on a Christian talk show as a Middle East expert, but I didn't find any definitive on his religion. Rmhermen (talk) 18:56, 6 November 2009 (UTC)

I agree with your point generally, but there were camps for German Americans, though on a much smaller scale to the Japanese ones. --Sean 17:35, 6 November 2009 (UTC)

I do not believe the claim that German American citizens were rounded up and interned like the Japanese American citizens, since I know a great many many German American citizens who lived through WW2 in the US without any problem. Certainly German citizens were properly interned. German American citizens were routinely drafted and fought against the German army in Europe. The article cited is poorly referenced. One ref is to a website of a group which claims such internments took place, and another is to a defunct inoperative reference. Edison (talk) 20:26, 7 November 2009 (UTC)

"It is nothing besides a coincidence that the nutjob who shot all those people at Fort Hood was a muslim or of muslim decent"—not necessarily true. Much is yet to be learned about this incident, but one aspect that has come out in the reports is that he sympathized strongly with the idea that Muslims were being persecuted by the United States in its War on Terror, and that he himself felt he was being harassed because of his religious beliefs. That does not support the idea that Muslims in general are untrustworthy, but it does indicate that his religions/ethnic affiliations may have played a part in all of this other than "coincidence." --Mr.98 (talk) 18:50, 6 November 2009 (UTC)

His personal experiences because of his religion may be being used by himself to justify his own actions, but to state that because this muslim went nuts and killed all those people that somehow all muslims must be suspect and under more strict scrutiny in the armed forces is pretty much bullshit. Any after-the-fact justification offered by this asshole for killing all of those people should not itself be taken as an indictment against people of the same religion as him. Using the facts of a person's life to "justify" an horrific act doesn't mean that people with similar experience are more likelyt to commit a similar act. Post hoc ergo propter hoc is still a fallacy here. --Jayron32 19:14, 6 November 2009 (UTC)

"It is nothing besides a coincidence that the nutjob who shot all those people at Fort Hood was a muslim or of muslim decent" — you do not know this and should not be asserting this. "To state that because this muslim went nuts and killed all those people that somehow all muslims must be suspect and under more strict scrutiny in the armed forces" — only you have proposed this. Why such hyperbole? Comet Tuttle (talk) 20:59, 6 November 2009 (UTC)

IT was the OP that stated that "Due to the latest massacre at Fort Hood, we're forced to look again at the US policy of deploying people to fight against those of their same religion and/or ethnicity." In other words, because a muslim committed the massacre at Fort Hood, the U.S. cannot trust muslims to fight for them in muslim lands. If the OP didn't mean that, he shouldn't have said that. --Jayron32 01:11, 7 November 2009 (UTC)

I didn't say that. I don't believe you will even find the word "Muslim" anywhere in my initial post. As I already stated, if we were fighting in Tibet, then I'd be worried about soldiers of Tibetan descent and/or the Buddhist religion, due to the same conflicting loyalties. StuRat (talk) 04:28, 7 November 2009 (UTC)

Well, I think we agree on the main points. I do think one could make an evidence-based argument that being a member of a certain religion raises one's risk factor in regards to being treasonous (in the same way that being a compulsive gambler makes one a better target for potential espionage recruitment), but one case does not a trend make. --Mr.98 (talk) 19:29, 6 November 2009 (UTC)

I have no doubt that 99% of soldiers remain loyal to the military despite being of the same religion or ethnicity as the people we are fighting, but that doesn't mean we need to accept the increased risk posed by the other 1%. I believe 5 British soldiers were just killed in Afghanistan by a member of the Afghan military they had taken in as a liason: [2]. Also, I recall that during the second Gulf War there was a similar incident where a Muslim solider tossed a grenade into a tent and killed some soldiers in Kuwait. So, it certainly looks like a trend to me, not a single isolated incident. StuRat (talk) 20:48, 6 November 2009 (UTC)

The question is not whether multiple muslims in the U.S. military have committed acts of atrocity. After all, multiple christians have done so as well. The question is whether or not the proprotion of atrocities committed by muslims is out of proportion to their overall representation in the military. A collection of anecdotes about muslim soldiers going nuts and killing their comrades is not a statistical study. For all we know, muslims may commit LESS such acts of horrific violence in the U.S. military than their non-muslim counterparts, we just don't know absent any sort of actual controlled study or analysis of actual data. Regardless of the claim, multiple isolated incidents does not a trend make. I could just as easily cherry pick a series of atrocities committed by men in the military who are shorter than 5 foot 6 inches tall, and make some equally outrageous claim about how short people show a trend towards violence. Cherry picking anecdotes is a bad way to support a hypothesis. --Jayron32 21:02, 6 November 2009 (UTC)

...not to mention the fact that they may, through better intercultural understanding, keep more people from being killed. If 5% of your Shai-Hulud worshippers go nuts and kill an average of 20 friendlies, but the other 95% manage to keep the local Fremen quite, you will still come out ahead. --Stephan Schulz (talk) 21:06, 6 November 2009 (UTC)

You seem to think I'm saying that Muslims are more likely to go nuts, per se. No, I'm saying that Muslims deployed to fight other Muslims (or about to be deployed, as in the Fort Hood incident) are more likely to switch sides than Christians. Similarly, if we were for some reason about to attack Tibet, I'd want to avoid sending any Buddhist soldiers there. Now for the statistics part: I can't recall any recent incidents of a Christian of European descent in the US military "going nuts" and killing a dozen people. I'm sure there are far more of them in the US military (probably at least 10 times as many), so I'd expect far more incidents, if it was just as likely to occur in either group. StuRat (talk) 21:22, 6 November 2009 (UTC)

No, I'm not assuming that. I'm not even assuming that they are more likely to "switch sides" if they are deployed to Iraq or elsewhere. I'm say that even if you grant such an assumption (for the sake of the argument) you may still come out ahead if you keep them in the force if you do a deeper analysis. I also think generalizing from a set of one is pointless, or each of Lynddie England and Charles Graner would imply that you cannot send any troops without massively violating human rights. --Stephan Schulz (talk) 21:32, 6 November 2009 (UTC)

Apparently you didn't notice my indentation. I was responding to Jayron, not to you. StuRat (talk) 21:39, 6 November 2009 (UTC)

Well, I did, but I assumed you just missed a colon. --Stephan Schulz (talk) 21:45, 6 November 2009 (UTC)

Now, as to your point, there are certain jobs, such as a linguist, or more broadly, a "cultural emissary", where having a background in the culture is desirable. However, for the majority of jobs where there is no advantage to be of the same culture as the enemy, why risk it ? In this case, why does one need to be a Muslim to be a good combat stress psychologist ? StuRat (talk) 21:39, 6 November 2009 (UTC)

Everyone you send abroad is a "cultural emissary", and you want all those people to be aware of local culture because that might be the difference between the locals becoming friends, neutrals, or enemies. We very much do not want to make this a conflict between "the west" and "the Muslims", but rather keep it a conflict between a free, tolerant and enlightened society and a group of backward fanatics (ok, we may need to work on that enlightened aspect on our side....). --Stephan Schulz (talk) 00:53, 7 November 2009 (UTC)

I don't think it's a good idea for all of our soldiers to have cultural contacts with the natives. Most soldiers, working in support positions, should be kept on base. This would include a "combat stress psychologist". Only those who need to have contact with the natives should, and they should be well trained in how not to offend them. If you have people who are going to get all bent out of shape if they ever see the bottom of your shoes, then we need to limit contact to only those who know this and all the other little traps out there. And, considering the risk of soldiers being captured and later beheaded on the internet, they better stay on base for their own safety, anyway. StuRat (talk) 04:23, 7 November 2009 (UTC)

Back to the original question, does the US military not have certain jobs that require background checks? And do those background checks not sometimes include whether they hold other citizenships, or have in the past? For example, might they stop a naturalized Chinese-American from working on certain military technologies the Chinese government might well be out to acquire through espionage? Sure this is under a different name, but it seems the same logic used regarding the Japanese-Americans in World War II, keep them away from things where divided loyalties could do a lot of damage. TastyCakes (talk) 20:53, 6 November 2009 (UTC)

I have no answer, but wanted to point out your example isn't a good one because your naturalized Chinese-American used to, ostensibly, be loyal to the government of China, whereas the WW2 Japanese-American case relied on racial ancestry to determine one's loyalty. Comet Tuttle (talk) 20:56, 6 November 2009 (UTC)

How about an American born Chinese-American that holds Chinese citizenship as well, or maintains close ties with China? I'm sure it would be done on a case by case basis rather than the blanket decisions of World War II, but doesn't such "discrimination" sound possible? TastyCakes (talk) 21:00, 6 November 2009 (UTC)

Note that the US military doesn't seem to worry about discrimination in other areas, like prohibiting women from serving in combat roles and discharging any "outed" homosexuals. In this context, discrimination for the goal of avoiding massacres seems acceptable. StuRat (talk) 21:10, 6 November 2009 (UTC)

I was trying to use the word in its most literal sense, not with the baggage it has acquired. As in having "discriminating taste". I don't think I object to this kind of discrimination in sensitive jobs either. TastyCakes (talk) 20:34, 7 November 2009 (UTC)

Yes, there are background checks for many military jobs. Yes, nationality and religion probably matter, as would your political ideologies and other personal factors. Yes, they'd be suspicious of Chinese-Americans applying to work on military technologies. (Cf. the Wen Ho Lee bruhaha). Whether these policies have accomplished much is hard to say—for things like espionage, a great deal of spies in recent years have just been motivated by money, not ideological or national loyalties. Extreme profiling has a negative effect as well (see e.g. Tsien Hsue-shen). --Mr.98 (talk) 21:30, 6 November 2009 (UTC)

It's of note that the day after the Fort Hood tragedy, another shooting spree has occurred, this time apparently by a fellow who ostensibly looks quite Caucasian. Do we feel the same way that we do about this Muslim fellow? There is a similar dearth of facts in both cases, but the directions our minds jump do tell us something about our prejudices. --Mr.98 (talk) 21:30, 6 November 2009 (UTC)

What incident is this ? Do you have a link ? Is it another shooting in the US military where a dozen people were killed and dozens more were wounded ? StuRat (talk) 21:42, 6 November 2009 (UTC)

It's on CNN—Orlando, FL, etc. Does it matter if it is in the military, per se? Does the efficacy of the attack matter? (He shot about half a dozen people.) The point is—there are nuts with guns out there in the world. --Mr.98 (talk) 22:12, 6 November 2009 (UTC)

Yes, it does matter if it's in the military, since this Q is about the problem of sending people to their ancestral homeland to fight against their own people. How does some random shooting outside of the military relate to this Q ? I don't think it makes sense to just say "all shootings are random, so there's no way they can be avoided". There are many ways to help, such as avoiding triggers, like soldiers with divided loyalties, or high school students who are ruthlessly bullied with no action taken against the bullies, or postal worker/veterans with post traumatic stress disorder being put under an abusive boss. Many of these shootings could be avoided with a little common sense. StuRat (talk) 23:34, 6 November 2009 (UTC)

I'm not saying, "all shootings are random," I'm saying, isn't it interesting that we are jumping to conclusions about the motivations of this particular shooting rather rapidly. --Mr.98 (talk) 13:42, 7 November 2009 (UTC)

He seems to have posted on the internet, praising the bravery and self-sacrifice of suicide bombers, so it looks like the initial assessment is correct. StuRat (talk) 14:01, 7 November 2009 (UTC)

Follow-up: I would very much like to be able to get some statistics on "soldiers who betray other soldiers" in the US military. Could I get a list of recent incidents and then maybe break it down by the religion and ethnicity of the perps ? StuRat (talk) 04:31, 7 November 2009 (UTC)

This WSJ article touches on the subject with recent attacks but is not comprehensive; it does say such killings were "far more common" in the Vietnam War, but cites no sources and names no numbers. (I think you may want to adjust for the claim that many soldiers killed their officers in Vietnam, as I think the motivations differ from what you're trying to get at.) Comet Tuttle (talk) 16:26, 7 November 2009 (UTC)

Thanks, that listed three incidents since 9-11:

1) The current Fort Hood incident, the murder of 12 soldiers, committed by Maj. Nidal Malik Hasan.

2) The murder of 5 soldiers, 6 months ago in Baghdad, by Sgt. John M. Russell.

3) The killing of 2 soldiers in Kuwait, with a grenade, in 2003, by Sgt. Hasan Akbar.

If anyone knows of any other incidents, I'd like to know. Also, do we have info on what portion of the US military is Muslim ? Somehow I doubt if 2/3 of the US military is Muslim, to match the perps above, or if 14/19th is, to match the number murdered, but I'd like to know the actual numbers. StuRat (talk) 20:08, 7 November 2009 (UTC)

CNN was saying it was 3500. I'm not sure if they meant total with reserves (almost 3 million troops) or active (1.4 million), but either way it's a tiny fraction. I don't believe that includes translators and interpreters and stuff, who I think are classed as civilians on contracts. TastyCakes (talk) 20:24, 7 November 2009 (UTC)

This story says the Pentagon says 3500 but that some Muslims in the military claim there could be as many as 20,000. TastyCakes (talk) 20:29, 7 November 2009 (UTC)

OK, so we're approximately in the range of 0.1% - 1%. In that case, having two-thirds of the incidents coming from a group that small would seem to indicate that there is a problem, and it's not simply a coincidence. StuRat (talk) 23:47, 7 November 2009 (UTC)

A social scientist would say you can't tell anything from a sample size of three. -- Mwalcoff (talk) 06:01, 8 November 2009 (UTC)

In a case like this we obviously don't want to wait until we have a sample size of 1100. If my math is correct, if you randomly choose 3 people from a population which is only 1% Muslim, the chances of two of them being Muslim would be 3*(0.01)³ or 3 in a million. In other words, it's not just random. While I agree that you can't determine the specific risk that a Muslim deployed to fight against Muslims will turn on his own men, from this, it's enough of an indication that there may be a problem that further study and actions should be taken. StuRat (talk) 13:46, 8 November 2009 (UTC)

Your math is not correct. It's 3*(0.01)²(0.99) (or 3*(0.01)² for the chance of at least 2). Notice that you overstated the unlikelihood by a factor of 100. But bad math aside, this exercise is pretty absurd. Your "statistics" are the anecdotal reports in one newspaper article. As Mwalcoff said, three is not an appropriate sample to start drawing conclusions from. You may have heard it said before that two does not make a trend. Rckrone (talk) 18:58, 8 November 2009 (UTC)

Yes, I used 3 instead of 2 as the exponent, which now gives me a 3 in 10,000 chance of this occurring at random. Still extremely unlikely. As I said before, it would be nice from a mathematical perspective to have a sample size of 1100 massacres to work with, but as a practical matter it would be absurd to wait for that. In the real world life-and-death decisions must be made with only partial information. One might as well have argued "well, Pearl Harbor was only attacked once by the Japanese, so let's wait until they do it a few hundred more times so we can gather valid statistics that we can then use to determine if our defenses are working or not". As for the accuracy of the newspaper article, it is a well-respected paper that I'm confident attempted to do the best research possible to find all pertinent cases. It's possible they may have missed some, and I've asked for info from any other sources to double-check them. However, until I find any info that conflicts with the newspaper article, I'm going to use their data as the best available. StuRat (talk) 19:53, 8 November 2009 (UTC)

I'm not really sure what statistics has to do with Pearl Harbor. What I'm saying is that it's not a useful tool here either. I'm not saying that there's not information to be gleaned from these events, but that information is not going to come from applying statistics. As for the WSJ article, it never claims to be an exhaustive list just three examples. I'm not saying there's anything wrong with the WSJ, I'm saying there's something wrong with construing the article as something it isn't. Rckrone (talk) 20:03, 8 November 2009 (UTC)

It seems unlikely to me that any recent massacres in the US military would have escaped the notice of the WSJ, and it also seems unlikely that they would have chosen to omit any from the story. StuRat (talk) 20:13, 8 November 2009 (UTC)

I will say this again: I'm not accusing the WSJ of bad reporting. The article never says they are presenting a comprehensive list. In fact quite the opposite they present two specific prominent examples that serve as illustrations for how to categorize these incidents. The article does say that the death toll from incidents like this is believed to be "fewer than a dozen" and the cases they mention add up to 7 victims, so obviously there aren't more than a handful of other cases, but the potential max of 7 is quite far from 3. There's also the possibility of cases where soldiers attempted to kill someone. Rckrone (talk) 20:22, 8 November 2009 (UTC)

I'm not trying to say that it's impossible that religion played some role in some of the cases where soldiers have snapped. In fact there seems to be evidence that it did in both cases mentioned here. But the way you're trying to draw conclusions using statistics is not really meaningful. Rckrone (talk) 19:53, 8 November 2009 (UTC)

I believe that the following is a valid problem for a statistician: "Given the assumption that an event has a 1% probability, but the measured results show that this event occurred in 2 out of 3 trials, what is the probability that this assumption was correct ?". I will cross-post to the Mathematics desk to confirm this. StuRat (talk) 20:03, 8 November 2009 (UTC)

Hmm , I can only wonder who will fight for the USA when the fight is against a "Christian country"? Only atheists, Jews, Muslims and Buddhists? That would make for a relatively small army indeed. Flamarande (talk) 14:49, 8 November 2009 (UTC)

In such a case it wouldn't be practical to keep all Christians out of the combat area, but you could still keep those of the same ethnicity away. In the case of the war against Serbia, for example, we could have kept the Serbs home. StuRat (talk) 17:33, 8 November 2009 (UTC)

(Edit conflict, reply to StuRat) This is not a place for debate, but your argument demonstrates a fallacy for which there must be a term that I don't know. It is possible after any event or series of events to find something seemingly unusual about it and use it to claim it can't be random. Bill Bryson talks about how creationists do this in his book A Short History of Nearly Everything. Creationists will say that the chances of the human genome developing randomly out of evolution is one in a zillion, so it must be intelligent design. But the key is they're doing this in hindsight. The chances of an NFL game ending with a score of 21-13 is 0.25%, based on the compilation of historical scores at www.pro-football-reference.com. So the chances of three games in a row ending with that score is, if my math is right, one in 64 million. But if that happens and you look back on it, can you conclude that it's some kind of miraculous occurrence? Of course not. Now if you predict ahead of time that three games will end in that score and it happens, that would be impressive. But in hindsight, no. -- Mwalcoff (talk) 15:01, 8 November 2009 (UTC)

I believe that's a type of cherry picking. In the football games, for example, they just choose three games that happen to have the same score, and considered that to be the sample size, rather than including all the games in the sample size. In the creationist case, they just consider the Earth to be the sample size (of one), rather than all the planets in the universe. If I had, for example, looked at every army in the world and only found this pattern in the military of Poland, then set my sample size to just the size of the Polish army, then this would indeed be a case of cherry picking. However, I didn't do that. If you wish, we could broaden the sample to include each military from majority non-Muslim nations which employ Muslims to fight against other Muslims, and I suspect the results would be similar. StuRat (talk) 17:27, 8 November 2009 (UTC)

I'm not talking about cherry picking. I'm talking about if you open the paper tomorrow and find all three games involving a given NFL division's teams ended in the same score, you can't call it a miracle, because you wouldn't be asking "what are the chance of that?" if it hadn't already happened. The license plate example comes from famous physicist Richard Feynman. Here's how Bryson quotes it:

• "'You know, the most amazing thing happened to me tonight,' he would say. `I saw a car with the license plate ARW 357. Can you imagine? Of all the millions of license plates in the state, what was the chance that I would see that particular one tonight? Amazing!' His point, of course, was that it is easy to make any banal situation seem extraordinary if you treat it as fateful."

Now had Feynman predicted he would see license plate ARW 357 and then saw it, that would be significant. But to look back on any occurrence after the fact and treat it as significant because it was unlikely to happen randomly is the fallacy I'm talking about. -- Mwalcoff (talk) 17:49, 8 November 2009 (UTC)

Well, in the football example there's two different things going on, just considering the odds of those three games having the same score (not considering all the other scores that didn't have that score), and also looking for the particular score of 21-13 rather than looking at any score that's repeated 3 times. I still call that "cherry picking", because, just like only picking the ripe cherries, we've only picked the scores that seem like they show a pattern, and ignored all the rest. If we go back to my original Q, what is "all the other data that I'm ignoring" which would tend to disprove my assertion that sending people in to fight against members of their own ethnicity and religion is risky ? StuRat (talk) 19:42, 8 November 2009 (UTC)

Actually, you're the one making the positive assertion. It isn't anyone elses position to disprove an unproven assertion. Its still just a hypothesis without sound data to back it up, and its an unproven hypothesis. You can't just make a hypothesis like "sending people in to fight against members of their own ethnicity and religion is risky" and then claim that it remains true until someone provides data to disprove it. If its true, then data will support it. It doesn't work the other way. One could make any random statement and say "its true unless you prove it false!" That isn't how it works!!! If it is true that it is riskier to send someone to fight against his historical ancestors, then where is the data which backs that up? And, remember, multiple anecdotes is not data... --Jayron32 20:32, 8 November 2009 (UTC)

are not data. --Tango (talk) 20:36, 8 November 2009 (UTC)

"Anecdotes" would refer to random bits of data someone thinks they remember, not facts, like the incidents listed. It may not be enough data for your liking, but, as I've said many times now, waiting until we have enough massacres so we have a "statistically reliable sample" isn't an option, so we must do the best we can with the data we have. StuRat (talk) 22:11, 8 November 2009 (UTC)

It doesn't matter how reliable the anecdote is, it is still an anecdote. To draw conclusions you need to know how many soldiers have fought against their own ethnic group, how many have fought against other ethnic groups, how many incidents that have been in the former group and how many incidents there have been in the latter group. We don't have those 4 pieces of information. --Tango (talk) 22:33, 8 November 2009 (UTC)

What exactly would qualify as "data" and not "anecdotes", under your definition ? StuRat (talk) 23:52, 8 November 2009 (UTC)

Something systematically collected from a representative sample and analyzed according to predetermined criteria would be a start. Otherwise, see Bible code. --Stephan Schulz (talk) 08:46, 9 November 2009 (UTC)

"Sampling" implies that you're only looking at a portion of the murders, which is sure to introduce sampling error. Why not look at every case in the given time-frame, as I propose ? StuRat (talk) 23:52, 9 November 2009 (UTC)

Please note the "representative". Of course your sample can be the whole population. But how do you get at "every case"? They may not all be published, or at least not be equally prominently published. That's exactly what causes (many) sampling errors. --Stephan Schulz (talk) 11:24, 10 November 2009 (UTC)

The problem of making a sample representative is completely eliminated if you use the entire population. Now, I'm sure that the US military keeps records of everyone convicted under the Uniform Code of Military Justice, so the problem just becomes one of access. A Freedom of Information Act request could certainly be made, and I see no justification for a denial. So, I'm asking if some news organization has done this and now has the data we need to do a correct analysis. StuRat (talk) 14:08, 10 November 2009 (UTC)

Follow-up: I'd like a list of all convictions (and also charges) of murder under the Uniform Code of Military Justice, where the victims were fellow soldiers, after 9-11-2001. I've tried to Google it, but had no luck. In addition to the three massacres listed previously, I've found the following murders, so far:

4) Calvin Hill accused of murdering Ashley Turner to cover up theft from her: [3].

5) Jonathan Law accused of murdering a fellow marine: [4].

6) Kyle Dayton accused of the murder of Carl J. Ware, Jr: [5].

7) Jonathan Campos accused of the murder of August Provost: [6].

8) Richard Smith and Mathew Kvapil accused of the murder of Christina E. Smith: [7]. Article also mentions the (murders ?) of Megan Touma and Holley Wimunc.

9) Cesar A. Laurean is accused of the murder of Maria Lauterbach.

9) Various murders of women: [8].

10) 11 murders at Fort Carson: [9].

11) George Smith accused of the murder of Kerryn O'Neill: [10].

12) Murder of Ciara Durkin: [11].

These all seem to be single murders or murder/suicides, but what I'm really looking for is multiple homicides. StuRat (talk) 23:52, 8 November 2009 (UTC)

....and, no offense, but there is your cherry-picking. If the data does not support your conclusion, change the criteria until they do. Next, look at religiously motivated crimes... ;-) --Stephan Schulz (talk) 08:43, 9 November 2009 (UTC)

Not really. Any investigator could likely tell you that the motives for massacres are typically far different than those for your average single murder. And even if you include all those single murder cases which I just listed, it still doesn't bring the numbers of US military murders by Muslims anywhere near to being in line with the Muslim population percentage in the US military. StuRat (talk) 23:44, 9 November 2009 (UTC)

That WSJ article I pointed to, btw, was the very flimsiest record you could have chosen for sample data. Reporters are lazy. You're correct that comprehensive data would be needed to try to draw any conclusions. Comet Tuttle (talk) 07:19, 12 November 2009 (UTC)

Not the flimsiest, that would be using a scandal rag as a source. Now, I'm still asking how to obtain the comprehensive data. Can't anybody help with this ? StuRat (talk) 16:06, 12 November 2009 (UTC)