Power Line reports that the and came to this conclusion:
Data: Combat hasn’t caused murder spike
Combat stress has not created a spike in murders by soldiers in the Fayetteville area, according to a search of records by The Fayetteville Observer.
Tracking killings reported in the newspaper before and after the Sept. 11, 2001, terrorist attacks showed that more soldiers were accused of murder in the six years before the attacks than in the six years since.
Twelve Fort Bragg soldiers have been accused of killing 13 people in the six-plus years since Sept. 11, 2001, according to Observer records. In the six years before the terrorist attacks, 16 Fort Bragg soldiers were accused of killing 18 people.
Those numbers came from a search of the Observer’s archives and may not be conclusive. Law enforcement agencies do not track killings by whether the accused was a soldier. The Observer examined its own records after a New York Times story published Jan. 13 indicated that homicides involving active-duty service members and new veterans rose 89 percent during the past six years.
Phil Carter’s supposition, that the article is based on Lexus/Nexus/Google research looks much more likely based on this. I may reach out to the paper in Columbia and ask if they’d consider doing the same thing.
I served as the chief of media relations for XVIII Airborne Corps and Ft Bragg, NC, for almost two years in the late 1980s. The Fayetteville Observer had one reporter whose only beat was Ft Bragg and matter related to it. He was a real journalist, too. He was probably as knowledgeable about military affairs in general as the typical graduate of the US Army Command and General Staff College. I haven’t see the Observer since I left there in 1990, but it was a good paper then and sounds like it still is.
I wrote the editor of the paper in Columbus and asked him to consider doing the same thing for Ft. Benning.
A.L.
A.L.
I see that you label this “real journalism’ and “digging”:
“Those numbers came from a search of the Observer’s archives and may not be conclusive. ”
But when the NYT “conducted a search of local news reports, examined police, court and military records and interviewed the defendants, their lawyers and families, the victims’ families and military and law enforcement officials.” they, of course, are not doing “actual journalism” but a mere nexus/lexus/google search.
A stake in the hear of the claim? Really? Because the numbers from one base (out of how many) show a decline, there must perforce be a national decline? Does this mean that Ron Paul will be president if he wins one county in Texas?
I’d also note that that NYT list of 121 includes many former soldiers who won’t be found on bases or in the immediate area of the bases.
I would submit that the statement that this article “put a stake into the heart of the one meaningful claim that the NYT could make in it’s article on killer veterans.” is a far more outlandish claim than any you accuse the NYT of making.
bq. I would submit that the statement that this article “put a stake into the heart of the one meaningful claim that the NYT could make in it’s article on killer veterans.” is a far more outlandish claim than any you accuse the NYT of making.
You mean like:
bq. But in a Blue America where veterans are objects of fear or pity, I guess it makes sense.
Gosh, when this guy gets on a roll, there’s no stopping him, is there?
nope, mark – I’m thinking you’re missing the point of this. If I’ve got a population of soldiers, tracking the behavior of that population over time is probably the best way there is to look at a population’s behavior.
If I was designing an research project, that’s how I’d design it.
A.L.
Alan, let me introduce you to two concepts: facts and opinions. The New York Times article is based in the world of facts, and if the facts don’t hold up there’s not much to discuss except why they screwed up.
Opinions are broader-based and subject to debate (note that contemptuous hand-waving is not debate).
A.L.
A.L., that might work if the sample were larger enough–and perhaps it is, I don’t know. But what you would do in theory is not what the Fayetteville paper is doing in practice. They are not tracking the population of soldiers, if those tracks lead beyond the small range of Fayetteville. Presumably, most soldiers who serve at Ft. Bragg are not from Fayetteville and, presumably, when they are discharged they do not stay in Fayetteville. Another, similar, difficulty with your method would be that many who serve in Iraq are reservists and guard and would be even less likely to counted in the population you are counting. For example, the 3 individuals who’s stories have been detailed in the articles so far, would not turn up using your method (if my memory is correct) as the killings they were involved with did not occur near the bases they had been stationed at. The Fayetteville’s paper’s method would not have found them, while the NYT’s method would (and did).
So, mark, if you think that sampling is a poor way to conduct research, it’s a strong inference that you believe the Lancet studies to be bogus.
Stand by for either silence or a blizzard of obfuscatory drivel.
Phil Smith,
a) I didn’t say that sampling is a poor way to conduct research (re-read my comment).
b) I said this particular group was not a useful sample for the reasons I stated. I.e., it did not contain the same elements as the other group to which it was being compared.
c) While I have heard of the Lancet study, I don’t know enough –or really anything — about it to have an opinion.
d) Sorry that you find this drivel.
I’ll give you this, mark, you’re consistent. Your arguments are peppered with assumption and presupposition.
Let’s try this in small words. The NYT made a factual, statistical claim that is not borne out by the evidence they presented, and is not in line with evidence that others have gathered and presented. It is up to the NYT to show that they are arguing in good faith, and they haven’t done that. Nor have you.
Phil Smith,
Please point to any assumptions or presuppositions in my arguments.
Please point to any factual or statistical claim made by the NYT that is not borne out the evidence they (presumably, the antecedent here is the NYT) presented.
Please point to any evidence others have gathered that is not in-line with those claims.
It is not up to the NYT — or me — to show they are arguing in good faith unless there is credible evidence that their facts are erroneous.
Up to this point, AL has not been arguing that the facts in the NYT were incorrect, but that they were not put in proper context. That is an opinion. A.L. has given reasons for his opinion. I have argued my opinion that the facts were put in proper context. There has not been any demonstration, however, that the facts were incorrect.
If small words work better for you, go ahead and use them.
Wait a minute. It’s been demonstrated – I forget by whom – that many of the “murders” counted by the NYT were DUI manslaughter, acquittals, and self-defense. Are you really going to stand by the statement that none of the facts of the NYT piece are in dispute? If so, my contempt for your intellect and/or honesty has been understated.
And for pete’s sake, the entire subject of this thread is the data collected by the Fayetteville paper that is inconsistent with the claims made by the NYT. I’ve just given you two recapitulations of facts that show that the NYT and you are arguing in bad faith. And while it’s not for me to speak for A.L., his position started with the position that the NYT didn’t contextualize the data they presented adequately (or at all, for that matter). It certainly didn’t stay there.
As for assumption, while it may be a valid one, you even label your presumptions as such – e.g., presumably discharged vets don’t stay near the base. That may be a valid supposition, but it’s still just presupposition.
With that, I have to go back to work. My apologies to the WOC crowd.
Phil,
“It’s been demonstrated – I forget by whom – that many of the “murders” counted by the NYT were DUI manslaughter, acquittals, and self-defense.” I’ll remind you who first pointed that out. The NYT. In the original article.
So the Fayetville paper “collects data” while the NYT “makes claims.”
Your two examples don’t stand up very well upon examination.
“That may be a valid supposition, but it’s still just presupposition.” That statement doesn’t make any sense to me. A valid supposition is still just a presupposition? Yes, you found an instance of where I made a valid supposition. If it’s valid, why would you have difficulty with my using it? I don’t get it.
mark –
It does not matter how the NYT derived the number 121 – how much or how little work they did to come up with it. The point is that in isolation, without a statistical framework of some kind, the number is meaningless.
The Fayetteville numbers may be limited to Fort Bragg, but at least they mean something. “The New York Times found X cases of Y” means nothing at all.
So if the New York Times “conducted a search of local news reports, examined police, court and military records and interviewed the defendants, etc etc etc” and then printed A MEANINGLESS NUMBER OF SINISTER IMPLICATION, is that journalism?
Well, it is of sorts. It’s old journalism; establishment liberalism preaching to a captive audience and acknowledging only such criticism as it chooses to print in its own pages. You can’t blame them for not doing a real study of PTSD in veterans, because that is a job that is far beyond the scope and expertise of any newspaper.
What they are doing is showing how easy it is (or rather, how easy it USED TO BE) to pretend to do something of the kind. Let’s not call it contempt of their readership; just the “low level of information” that makes the New York Times so useless.
mark, you seem to miss the problem that the Fayetteville Observer report highlights in that earlier NYT piece.
Reporter Arenschield wrote:
bq. Twelve Fort Bragg soldiers have been accused of killing 13 people in the six-plus years since Sept. 11, 2001, according to Observer records. In the six years before the terrorist attacks, 16 Fort Bragg soldiers were accused of killing 18 people. Those numbers came from a search of the Observer’s archives and may not be conclusive… The NYT found 121 cases in which veterans of Iraq and Afghanistan were charged with or convicted of a killing in the United States. Five of those cases involved Fort Bragg soldiers.
In other words, the NYT’s methodology missed at least seven Fort Bragg soldiers who committed homicides in FY02-FY07.
Their total number was 349 homicides committed by armed forces members and “new” vets, FY02-FY07, divided between 121 who had served in Afghanistan and Iraq, and 228 who had not.
What should that “349” actually be?
What should that “121” actually be?
What should that “228” actually be?
The NYT’s number for FY96-FY01 was 184. The Times didn’t “show its work,” so we can’t compare their count to the Fayetteville Observer’s tally of 16.
16 (FO) cases divided by 184 (NYT) cases gives a ratio of 0.087 for FY96-FY01.
12 (FO) cases divided by 349 (NYT) cases gives a ratio of 0.034 for FY02-FY07.
This suggests that there’s something wrong with that “184” number–it’s too low by a factor of 2 or so. Which would make sense: pre-2001, reporters might not have been as diligent about stating that the (alleged) offender was in the military or a “new” veteran.
More evidence that “War Torn” was a set of anecdotes with no statistical power. Over the copy that read
bq. The Times used the same methods to research homicides involving all active-duty military personnel and new veterans for the six years before and after the present wartime period began with the invasion of Afghanistan in 2001. This showed an 89% increase during the present wartime period, to 349 cases from 184, about three-quarters of which involved Iraq and Afghanistan war veterans.
an editor should have stamped “GIGO.”:http://en.wikipedia.org/wiki/Garbage_in,_garbage_out
Glen, I am arguing that the number is neither meaningless nor sinister. I argue that they provided the necessary statistical framework by comparing the numbers for the same group (people serving or recently served in the military) for six years before and for six years after/during major combat. They found a rise in the # of killings among that group. It was a small # to begin with, and it rose, according their account of their findings, by 89%. They said that Iraq/Afghan vets accounted for 75% of the rise. That’s really all there is too it.
From there they have gone on to speculate that the rise is partially attributable to combat-related PTSD. And then, the meat of the story so far, they have detailed the stories of 3 individuals.
What ever the faults of the NYT #s or their gathering methods, they both were certainly more informative than those of the Fayetville paper, which merely looked at killings in one very limited area.
I understand that your opinion is that the #s are taken out of context by the NYT and done so deliberately, specifically in order to make a specious claim. I just don’t think that opinion is backed up very well by any evidence or facts or logic.
I think that if you have a simple hypothesis such as “Combat-related PTSD will rise during war-time and will lead to an increase of its most severe manifestation,” a good way to test that is to compare the rate of incidence among the group of people most likely to be affected, i.e., people in the military, during wartime vs. during (relative) peacetime. I don’t see anything more sinister than that going on here, nor I have hear a very persuasive argument that there is.
Amac,
to it’s credit, the NYT article said that its method probably under-counted and so it’s no surprise that they missed many killings, homicides or otherwise. A more indepth spot-specific examination is bound to find more incidents. The exact number really isn’t the issue. If the NYT missed 50% of the occurrences, it’s still a small percentage of the whole.
“pre-2001, reporters might not have been as diligent about stating that the (alleged) offender was in the military or a “new” veteran.” This possiblity was mentioned in the article, as I recall.
Also, I would argue that the article was not an attempt to prove that there is a rise of PTSD during war so much as that was the background for the heart of the article, which was the stories of individual vets who suffer from extreme PTSD.
I fully agree, and have said so elsewhere, that the paragraph explaining the 89% rise is terribly written and very confusing. I am surprised it hasn’t been corrected in on-line versions. It suggests the number should be about 260, not 121, as far as I can tell. But in either case, the number is low and suggests an increase.
bq. Alan… The New York Times article is based in the world of facts, and if the facts don’t hold up there’s not much to discuss except why they screwed up.
Sure, knock yourself out. In the end, however, it is your opinion of why these alleged discrepancies occurred that I and others have been hammering you on, and which cannot be supported by facts.
bq. Opinions are broader-based and subject to debate
Yet still can either be based upon or dissociated from facts, which is where your opinions lie:
bq. (note that contemptuous hand-waving is not debate).
Exactly my thoughts when reading many of your posts. In a nutshell.
If the Times’ numbers are garbage, then SO ARE YOURS….which raises the question:
If an “opinion” is based on garbage data, does that make the opinion garbage as well?
mark,
The ‘numbers’ paragraph in the NYT article was clumsily written, but it can be parsed; check back in the earlier threads. The wordsmithing issue is trivial, once worked out. The real problem is that the data set is no good.
Note that I’m not pro PTSD, nor am I suggesting cuts in treatment. But if you or I or the NYT is going to discuss an issue, our numbers should mean something. It’ disingenuous for the Times to discuss “89%” and “three-quarters” of anything when estimates of the numerator and the denominator are each subject to errors that are (variously) wild, erratic, systematic, random, and unknown. Somebody else (the Fayetteville Observer) looks at a subset of the data with much the same method that the NYT used, and they come up with a completely different number–that’s a warning flag right there.
How bad does bad have to get before we can agree that “that’s bad”?
It is precisely the limitation which makes it informative. The fewer limits, the more statistical analysis dissolves to nonsense. You could learn more from studying these few cases at Bragg than you can by ransacking the country for anecdotes.
Speaking of anecdotes, the most notorious murder to ever occur at Fort Bragg was Captain Jeffrey MacDonald’s murder of his family in the 70s. MacDonald was a doctor trained in civilian practice who had recently joined the Army; he was not a Vietnam veteran.
Marc,
I have one question for you.
Do you suppose the average population of Fort Bragg has been fairly constant in the six years prior to 9/11 as well as the six years post 9/11?
And if not, what factor might have caused the fluctuation?
AMac, I appreciate what you are saying but I think you are winding up a somewhat irrelevant thread. I don’t think it is disingenuous. It’s just an imperfect world. To expect peer-reviewed scientific journal standards in a newspaper article is impractical. It’s not an outlandish expectation that PTSD will rise within the military during a time of war. A cursory study seems to back that up. But the study is just background to the heart and soul of article which is about the individual stories of some of these guys. After all, they were searching for people, not for numbers. You could cut out all the graphs about the numbers and the merit of the article would remain.
Davebo, I don’t really have any idea. I’d guess that since 9/11, with more reservists and guards being called up, the average age in the military has gone up some. But I don’t know how many reservists or guards, if any, go through Ft. Bragg. I would also guess that in the months following 9/11 enlistment went up, and that there was a broader range of age of those enlisting (but only in one direction, towards older) than previously. But I have no idea if that phenomenon was sustained. I think I have heard that enlistment has gone down some lately. But again, I don’t know who goes through Ft. Bragg so I don’t if it would be affected by any age-related trends in the military at large.
Again,
What is the rate for homicides committed by military personnel for the years 1996-2001?
What is the rate for homicides committed by military personnel for the years 2002-2007?
I don’t see where the NYT adjusts for the huge increase in active duty personnel post 9/11. At least 50k new active duty troops plus tens of thousands of reservists called to active duty. Pre 9/11, reservists who committed crimes and were not serving on active duty were counted as civilians.
Demonstrate that all of the 121 case homicides were atributable to combat-related PTSD. Oh, you can’t do that can you? So why all of the talk about the war and adverse consequences?
As a matter of fact, less than 30 of the NYT’s “cases” can be linked to a credible diagnosis of PTSD, to say nothing of combat-induced trauma. Once again, a defense lawyer’s plea at trial doesn’t count.
All of the “but isn’t PTSD worth talking about” hoohah is just a moving of the goalposts. The story is not about the ravages of PTSD, it’s about the ravages of Bush’s War and what it’s done to vets and to us.
Let’s review: To be a casualty of war, you have to have been in a war. REMFs don’t count. The NYT counts a lot of REMFs in order to pad their numbers.
Mark,
I was referring to age, but to the number of active duty on the base at any period in time.
And I’m just fleshing it out in my mind so bear with me. But wouldn’t you think a sizeable percentage serving from Fort Bragg in Iraq might otherwise have been back home at Bragg?
I seriously doubt any increase in enlistments would have offset the number deployed for a year to fifteen months at a time to Afghanistan and Iraq.
Do you get where I’m going here?
Oops, again!
I was (not) referring to age…
Davebo, sorry….I don’t know where I read age there. Maybe aver “age”? Sorry. I would guess that the total population of most bases in the last 6 years have gone down given that 150,000+ are serving abroad that weren’t serving abroad before 911. (plus the # in Kuwait and other places) Although, perhaps they are drawing from troops that would otherwise have been in Germany or S. Korea. I really don’t know. The NYT article states that the # of military personnel serving in the US is down from pre 911 levels. I have no way of knowing if that is true.
I would guess the increase in troop strength overall might be offset by the number of troops serving abroad during these wars. So, for any given base, I would guess it depends on how connected that particular base is to overseas deployment.
But, no, I don’t really see where you are going with this. Don’t forget that vets–that is guys who are no longer serving–are included in this count. It’s not just guys still serving post combat.
According to GlobalSecurity.org, “approximately 43,000 military and 8,000 civilian personnel “work”:http://www.globalsecurity.org/military/facility/fort-bragg.htm at Fort Bragg.” (Looks like this is a 2005 figure.)
12 homicides over 6 years calculates out to a rate of 28 offenders per 100,000 per year. Assuming that 43,000 is the denominator the Fayetteville Observer used, and that it has held steady. But are there other soldiers who are resident at Ft. Bragg for some period of time (e.g. for training or resupply), but who don’t count towards the figure of 43,000 who work there?
*Davebo at 25*
_And I’m just fleshing it out in my mind so bear with me. But wouldn’t you think a sizeable percentage serving from Fort Bragg in Iraq might otherwise have been back home at Bragg?_
I came in to say the same thing – I’d say the total population could easily drop by a quarter by the 6 years after, vs the 6 years previous. Going higher could be easily argued better if you had access to how deployments were handled. “My Wiki link says”:http://en.wikipedia.org/wiki/Fort_Bragg,_North_Carolina that the base has a high population density but only 29k, but doesn’t anything about the surrounding areas.
Real journalism would also have called out the 18 to 34 population group as skewed due to urban populations, and adjusted for city size < 50k, or recognized they weren't using the same set of data as the NYT (see below). *AMac at 19* _Somebody else (the Fayetteville Observer) looks at a subset of the data with much the same method that the NYT used, and they come up with a completely different number--that's a warning flag right there._ The Observer also only counted active duty it appears, not recently discharged - or as was also brought up, Reservists(as NYT did), which are less likely to be clustered around Fort Bragg or labeled as a Fort Bragg soldier.
I deeply enjoy the fact that the primary defender of the Times on this thread was able to write, “The NYT article states that ___. I have no way of knowing if that is true.”