US Attorneys: Document Dump, Etc.

by hilzoy

I have now skimmed more of the recently released DoJ documents than I'd care to admit to, and they are a puzzling lot. Boring, for one thing: there are approximately a million trillion versions of the Talking Points on US Attorneys, Paul McNulty's testimony, a compilation of relevant editorials, etc. I suspect that the document dump reflects a "ha ha, you wanted documents, we'll drown you in documents!!" approach -- it's hard to see how else to account for the fact that several very tedious email exchanges are printed out, in their entirety, about ten times. (Bear in mind that every single email -- even just "sounds great" -- quotes the entire previous exchange.)

Still, there are a few striking points. One is the virtual absence of the Attorney General. I seem to recall hearing somewhere (can't recall where) that he is not a fan of email, so the lack of emails from him is perhaps not surprising. But I've read through Part 8 of the documents posted so far -- easily over 1,000 pages -- and there are no memos from him, and only a very few mentions of him (as in: the Judge wants this, the Attorney General is upset about that, etc.) If I omit things like the million and one copies of the USAttorneys' resignation letters, which are addressed to Gonzales, and other pro forma things, I think I could count the mentions of him that I encountered on my fingers. And many of those come after the scandal has erupted, not while decisions are being taken.

This is really odd. The people writing the emails and letters are often quite senior people in the DoJ, and yet the man who runs the place is almost never mentioned at all, in a plan that might end up costing him his job. I can think of two reasons for this. First, correspondence in which he plays a more prominent role is being held back; second, he actually doesn't do much, and the DoJ is run largely without him.

Whether or not the first is true, I can't see how the second could not be. Even if we were to assume that there are also lots of documents that the DoJ is cleverly concealing, since they reveal Gonzales' secret perfidy, there are just too many decisions being taken with (apparently) no input from him. For the most part, people don't even mention needing to run things by him; it's as though he doesn't exist.

This doesn't surprise me much. Gonzales is the ur-Harriet Miers, and if people took the Attorney General's job as seriously as they take Supreme Court Justices, he would never have been confirmed. Still, it was pretty stunning to see how complete a nonentity he seems to be in the Department he allegedly runs.

Also: you can see Kyle Sampson not just lying, but drafting dishonest letters for his superiors to send to Harry Reid under their own name, in part 3.7 of the documents, pp. 48-9 and 53-5 (pdf). He tells Gonzales that "I am not aware of Karl Rove playing any role in the Attorney General's decision to appoint Griffin", and puts almost the same statement in his draft of a letter sent to Harry Reid under the signature of Richard Hertling, the Assistant Acting Attorney General. I honestly can't imagine why anyone would hire him after this, except for political reasons. And if someone does hire him to curry political favor, I can't imagine they'll trust him much.

Last bit of news: I see that the Senate has voted to repeal the change in the way US Attorneys are replaced -- the one that was snuck into the PATRIOT Act. Good for them.

Comments

[moe99]

Like Mr. Maguire and FDL, I've yet to see Mr. Doe on TPM.

[CaseyL]

So far the only emails they've found for the 18-day gap are a few complaining that a couple USAs didn't prosecute enough obscenity cases. That's supposed to support the claim that the USAs were fired for not doing their jobs well enough.

Now, I don't know about you guys, but if I'm evaluating a federal prosecutor, I'd look at the whole portfolio. Of all the federal crimes, 'obscenity' would be way down my list of priorities, after murder, kidnapping, organized crime, and corruption. If a USA had a good record on those, I doubt I'd use 'lack of emphasis on obscenity cases' as a reason to fire.

And it begs disbelief that there'd only be a handful of emails for the 2+ weeks right as the decisions were made to fire the USAs.

There's also the issue of WH using non-governmental email. A fair number of emails bearing private addresses have been found, which begs the question of how many other 'sensitive' conversations were carried about via undiscovered addresses.

[KCinDC]

Not just private e-mail addresses, CaseyL -- addresses from domains owned by the RNC. In this White House there is no dividing line at all between campaigning and governing. Yet in the 1990s the House took 140 hours of testimony about whether Clinton had misused the White House Christmas card list for political purposes, and people obsessed over what phone Gore had used to make fundraising calls.

[John Doe]

Er, the fact that any investigation of a federal agency or the executive branch - and this is technically both - produces hundreds of thousands

And you know this . . . how? I can see that number of documents being at issue if this were the Microsoft antitrust investigation, or even a typical SEC investigation. But we're just talking about 1) emails and memoes, sent between 2) a handful of top principals, relating to 3) one narrow issue. I don't think Kyle Sampson could have created "hundreds of thousands" of documents on this issue even if he had done absolutely nothing else for the past three years.

Like Mr. Maguire and FDL, I've yet to see Mr. Doe on TPM.

What is this even supposed to mean? That Mr. Maguire and FDL haven't seen me "on TPM" either? Well, no doubt they haven't. But what exactly does it mean to be "on TPM"? Is there something that I am supposed to be doing there? What would that . . . oh, never mind -- if you can't even write a single sentence in a non-ambiguous way, what's the point of asking for clarification.

[Ugh]

The "email gap" has made the top story on cnn.com (of course it's almost 10pm eastern).

[Jay S]

CaseyL,
Your priorities aren't the same as Al's wrt obscenity.

[CaseyL]

JayS - Yeah; shades of Ashcroft, who cut anti-terrorist programs in order to focus on porn back in, oh, the summer of 2001.

But, to get back on track, the issue here is legal, not political. It's clear that USAs were fired, or not fired, on the basis of how well they obstructed investigation into Republican corruption and how well they played along with bogus charges against Democrats. Unless "serving at the pleasure of the President" means "obstructing justice at the pleasure of the President," those are not legitimate reasons to fire USAs.

[John Doe]

I assume you're paraphrasing Josh Marshall's words [http://www.talkingpointsmemo.com/archives/013159.php]: The president fired US Attorneys to stymie investigations of Republicans and punish US Attorneys who didn't harass Democrats with bogus voter fraud prosecutions. In the former instance, the evidence remains circumstantial. But in the latter the evidence is clear, overwhelming and undeniable.

At least Marshall is honest enough to admit that the evidence (what little there is) is circumstantial in the first instance. But Marshall goes on to say:

What he does not say -- but what we know directly from the accounts of the players involved -- is that these were cases in which Republican operatives and activists complained to the White House and Republican members of Congress that certain US Attorneys weren't convening grand juries or issuing indictments against Democrats, even though these were cases where all the available evidence suggests there was no wrongdoing prosecuted.

All the available evidence indicates no wrongdoing? That's quite an incredible overstatement.

As the New York Times recently pointed out, there were about 3,000 fraudulent voter registrations that showed up in New Mexico alone, from Democratic activist groups. This is a state that Gore won by a mere 366 votes in 2000. Do you realize how simple it would be to have a handful of activists go around to different precincts and vote dozens of times under the identities provided in false registrations? When the leadership of the free world is at stake, there are certainly some people -- on either side -- that would be willing to do something like this.

Granted, we don't know that this happened -- but given that ballots are secret and there's no ID requirement in New Mexico, there's no way we could ever find out if this happened. All we do know is that someone perfectly set the stage up for some serious voter fraud, in a key swing state.

If it had been Republican activist groups that had submitted thousands of fraudulent registrations, then, far from pretending that it was unethical for a Democratic President to want the situation investigated, you guys would all be screaming for heads to roll.

Also, on another note, several of the Assistant US Attorneys who worked for Iglesias signed a letter attesting to his incompetence:

The letter address the recent controversy, describing Iglesias as an absentee boss who was more interested in travel than in running the office.

It said he "abdicated his responsibility as United States Attorney, turning over virtually every important decision to his subordinates."

The letter also said that Iglesias' "lack of leadership" resulted in a decline in the quality of work produced by his office and that the reputation of the office had suffered during his tenure.

[Steve]

As the New York Times recently pointed out, there were about 3,000 fraudulent voter registrations that showed up in New Mexico alone, from Democratic activist groups.

The NYT didn't say there were 3000 fraudulent registrations. One thing that gets annoying in these discussions is when people change a word to something that isn't true, in order to make their case stronger, and you have to waste time correcting it.

Granted, we don't know that this happened -- but given that ballots are secret and there's no ID requirement in New Mexico, there's no way we could ever find out if this happened.

If you think about it, it's pretty easy to find out whether anyone actually voted under a false registration. Is there a signature in the book next to the name of the fictitious person? Did someone mark off the fictitious person as having voted?

If it had been Republican activist groups that had submitted thousands of fraudulent registrations, then, far from pretending that it was unethical for a Democratic President to want the situation investigated, you guys would all be screaming for heads to roll.

There is nothing unethical about wanting the situation investigated. Could you please show me someone who has said it was unethical to want the situation investigated? The thing is, the situation was investigated, and the prosecutor determined there was no basis to bring charges. What's improper is to insist, as some Republicans did, that charges be brought regardless of what the official investigation shows.

It's natural that Republicans always think Democrats are guilty of stuff, and Democrats always think Republicans are guilty of stuff. But in America, no one should ever be prosecuted merely because some partisan politician thinks they should be. That's why threats against someone's job for not indicting enough members of the opposing party are so pernicious.

[John Doe]

If you read the NY Times article, Iglesias didn't want to prosecute because the one woman in question hadn't confessed that she intended to commit fraud. But if that's the standard, then voter fraud would never get prosecuted.

If you think about it, it's pretty easy to find out whether anyone actually voted under a false registration. Is there a signature in the book next to the name of the fictitious person? Did someone mark off the fictitious person as having voted?

If those things occurred, Democrats would argue strenuously that this proves the voter registration wasn't fictitious after all. And how do you prove otherwise? Look for a record of the person's ID? "Racist!" say the Democrats (really, they do say this). Check the person's claimed address? "But lots of people move," say the Democrats. And so on.

[Matt Weiner]

He said that when the F.B.I. interviewed her, she did not make any clear admission of guilt. And under federal election law, Mr. Iglesisas said, prosecutors would have had to prove that the woman, who had been fired for other reasons, had falsified the applications with the intent of influencing the election. Mr. Iglesias said “it appeared she was just doing it for the money.”

He said he ran the question of whether to prosecute her by Craig Donsanto, director of the Justice Department’s election crime branch. He said Mr. Donsanto said he did not “have a strong feeling either way.”

If you have any suggestions as to how to prove intent beyond a reasonable doubt in such a case, do let us know.

Your hypothetical generalizations about Democrats are also not appreciated.

[moe99]

80% of the political corruption cases this administration has brought are/were against Democrats. The 7 USAttys who got into trouble were either: a)prosecuting Republicans; or b) refusing to prosecute Democrats where there was no case. In Seattle, where I have practiced law for more than 25 years, the payback will have enormous repercussions with the moderate Republicans. McKay was told by DOJ in August of 2006 that he was good to go for his nomination for US Dist Judge. That went down in flames less than 4 months later. His brother was a chief fundraiser for the Bushes since 1988. I talked with him on Friday and I don't have to tell you what his opinion of the current administration is: it's not printable here.

These guys really aren't even up to the standard of a Mayberry Mafia because they're so frikkin venal and cruel.

[John Doe]

Your hypothetical generalizations about Democrats are also not appreciated.

What hypothetical generalizations about Democrats? Democrats do say that it's racist to look for voter IDs -- that's virtually the only thing they say when the question of voter ID comes up. If the fact comes up that voters don't live at their supposed addresses, Democrats do say that people move, (this sort of thing was constantly said in Ohio in 2004 when something like 35,000 voter registration mailing were returned as non-deliverable).

80% of the political corruption cases this administration has brought are/were against Democrats.

Steve -- notice that your own side has a serious problem with putting out false memes that never disappear, no matter how many times they are refuted. It's been proven beyond a reasonable doubt that the Cragan/Shields study is about as reliable as reading entrails. Look at the pathetically inadequate defense that these professors made of their debunked study; notice that they don't deny that it was based on Google searches (not on any kind of random sampling); note that, to them, the primary indicator of the study's reliability is that it appeared in an electronic non-peer-reviewed journal that is an offshoot of the Daily Kos; note that the professors admit that they "can't even get data" on these questions, but had to rely on searches of news stories. That whole study is just an exercise in intellectual dishonesty, and anyone who cites it is intellectually dishonest as well.

[moe99]

I went to the site that Mr. Doe left a link for and >surprise< there was no evidence of the authors of the study doing a google search to gin up the numbers as Mr. D implies.

In fact the authors are well respected academics. "Donald C. Shields, Ph.D., is Professor Emeritus, Department of Communication, University of Missouri – St. Louis. John F. Cragan, Ph.D., is Professor Emeritus, Department of Communication, Illinois State University. Their interest in transparent government began with the publication of their book, Government Surveillance of U.S. Citizens: Issues and Answers (Minneapolis, MN: Campus Press, 1971)."

And here is part of Drs. Shields and Cragan had to say in response to a Swift Boat attack of their study in the Philadelphia Enquirer:

"…In a little over a 1,000 words, Mr. Smerconish attacked our integrity and the accuracy of our data. He lamented that the study was not published, while failing to mention that we informed him during a phone interview that the study was going to be published by the electronic media journal, Epluribus Media (http://www.epluribusmedia.org/columns/2007/20070212_political_profiling.html). This article contains the complete raw data tables, naming each of the 375 politicians classified by their states, their political affiliations, the type of investigation, and the day a news story appeared locally describing the event. Because there is no national register of these investigations, it took more than 400 hours to find and organize this data (our results were fact-checked by multiple editors of Epuribus Media before they published it).
…As well, Smerconish failed to report that he knew the study had been presented at a peer-reviewed panel in Boston, MA, in November 2005, and that an earlier report of the data received honors as a top-three peer-reviewed paper at the annual meeting of the Southern States Communication Association, held in Baton Rouge, LA in April 2005…."

[Steve]

It's been proven beyond a reasonable doubt that the Cragan/Shields study is about as reliable as reading entrails.

No, it's been asserted by you, repeatedly, in vitriolic terms, which falls a long way short of "proof beyond a reasonable doubt."

Explain to me, if you will, how the methodology of using a Google search could possibly explain a 7:1 discrepancy between Democrats and Republicans? Does the media only write about indictments of Democrats? (That would sure contradict the conventional wisdom from your side.) Is it only investigations of Democrats that get leaked to the media? (That would raise a whole new set of issues.) What, exactly, is your explanation for how this ostensibly neutral method of research yields such skewed results?

It's hardly enough to say that this methodology doesn't find every single instance of investigation; of course it doesn't. The question is, is there any reason whatsoever to believe that the investigations which aren't picked up by this method are primarily those aimed at Republicans, and in such a degree as to balance out the 7:1 ratio among those investigations that ARE picked up?

Do you want to see a laughably flawed methodology? Take a look at your own link to the "Stubborn Facts" blog. The guy tries to minimize the number of states where a skewed ratio exists by means of a two-step process:

1) First, eliminating any State where the ratio of Republican investigations to Democratic investigations is greater than 1, because obviously there's no problem there;

2) Second, looking at only the remaining states, eliminating any State where the ratio of R to D investigations is greater than the ratio of Rs to Ds in the state legislature, because one assumes that elected officials SHOULD be investigated in proportion to their numbers in the state legislature.

And having done this, he's left with 24 states, from which he triumphantly concludes that even using the study's own data, "fewer than half the states show a propensity toward 'profiling' of Democrats for political prosecution by the U.S. Department of Justice." That's right - he proclaims the study debunked because, even if there is a systematic bias towards investigations of Democrats, it's only a problem in 24 of the 50 states!

But putting that aside, can you see the very, very serious problem with the method he uses to get the number down to 24 states - a problem so glaring it makes your citation to this post absolutely laughable? I'll assign that question as homework.

[John Doe]

I went to the site that Mr. Doe left a link for and >surprise< there was no evidence of the authors of the study doing a google search to gin up the numbers as Mr. D implies.

Sorry; I wasn't aware that you were ignorant of the previous thread on this topic. In that thread, I copied part of the relevant story by Michael Smerconish, who actually called up the study's authors. This is what he found out:

I asked Shields about his methodology. How did he arrive at the conclusion that the Bush administration targets Democrats?

It seems the research done by Shields, or Cragan, or both consisted of simply hitting the Search button on Google.

Specifically, their work consisted of a tabulation of the number of times the words or phrases federal grand jury or public corruption AND elected produced a result, along with "a census search of extant press releases available at each U.S. Attorney's Office official Web site." The names of Democrats come up far more often than those of Republicans.

Don't even bother disputing that this story is accurate -- Cragan and Shields specifically attempted to respond to Smerconish, and they never denied that their methodology consisted of Google searches.

Steve: Explain to me, if you will, how the methodology of using a Google search could possibly explain a 7:1 discrepancy between Democrats and Republicans? Does the media only write about indictments of Democrats? (That would sure contradict the conventional wisdom from your side.) Is it only investigations of Democrats that get leaked to the media? (That would raise a whole new set of issues.) What, exactly, is your explanation for how this ostensibly neutral method of research yields such skewed results?

I've already addressed this once, in that earlier thread. For the information of those of you who have never used Google, many newspaper articles never make it to Google, and particularly small-town newspapers. Also, even those articles that do appear online are only there for a few weeks. Good luck trying to find the average newspaper article from 2001 or 2002 still online somewhere.

So what does this mean? If Cragan and Shields go out and do a search on Google, any news story that they find from the 2001-03 timeframe that is still online is probably from one of a few major cities -- where, as it happens, local Democrats are usually in charge.

So what does this mean? It means that if you're just counting up news stories that appear on Google (which is a moronic methodology under any circumstance), the news stories you find are more likely to involve local Democratic officials in those big cities.

So what does this mean? It means that this is the equivalent of cherry-picking the data. It's the equivalent of measuring global warming by taking the temperature at noon in July and August. It's complete and utter crap.

Also, Steve, for your information, I wasn't citing to <>a href="http://stubbornfacts.us/politics/partisanship/more_debunking_of_the_political_profiling_study#comment-5071">that Stubborn Facts post itself. I was citing to the specific comment where Cragan and Shields wrote in to "defend" their study in hilariously inadequate terms (i.e., bragging that it got published by an offshoot of Daily Kos; any real academic would consider this a joke).

[Dantheman]

moe99,

"the Philadelphia Enquirer"

While it's unlikely it will ever return to its glory days in the 1970's and early 1980's, and the current publisher seems determined to turn it into a slightly more respectable version of The Washington Times, please do not conflate the Philadelpha Inquirer with the National Enquirer.

[moe99]

Dantheman: my apologies for the 'e' instead of the 'i.'

Mr. Doe: two words: 'peer review'

The fact that Drs. Shields and Cragan did not explicitly deny the 'google search' accusation, does not ipso facto, mean they have admitted it.

[John Doe]

Peer review = meaningless, if the peers are just as biased (or sloppy) as Cragan and Shields are. Also we don't know what a "refereed panel" at the National Communication Association even consists of. Given that the NCA seems to have about 1200 programs at its national convention, and given that the paper acceptance rate is over 60%, I'm going to go out on a limb and suggest that their standards are pretty loose. (As if we didn't already know that.)

[John Doe]

Steve: Here's a City Pages article that offers a good example of why the Cragan/Shields methodology is crap:

But are their numbers reliable?

Consider the Minnesota examples in the report. Cragan and Shields identify seven former or current office holders who they say were investigated by the Department of Justice.

There's no disputing that assertion in four of the named instances. Former Minneapolis City Council members Brian Heron, Joe Biernat, and Dean Zimmermann were all convicted in federal court on charges related to public corruption, as was Loren Jennings, a former state representative. All are Democrats, except for Zimmermann, who is a Green.

But there are questions about the claims regarding the other three Democrats who made the professors' list — U.S. Senator Amy Klobuchar, former St. Paul Mayor Randy Kelly, and Minneapolis Mayor R.T. Rybak. None was ever prosecuted by the U.S. attorney. Further, from the publicly available information upon which the professors relied, there is no definitive indication that any of the three were under investigation.

The inference that Kelly was targeted arises from reports of a 2005 bribery probe focusing on a top Kelly aide. At the time, Kelly, who did not return calls for this story, stated that the investigation was being used to tar him politically. He also intimated that the leak came from a fellow Democrat.

The notion of Kelly as a victim of Bush administration political profiling seems far-fetched when you recall that Kelly — in an act that would doom his political career — crossed party lines to endorse Bush for re-election.

Such inconvenient details have created a curious situation. Loyal Democrats like Rybak now find themselves challenging findings that are being used—very effectively—as a cudgel against the Bush administration. Last week, Andrew Luger, an attorney for Rybak, fired off a letter to Cragan and Shields, asking that Rybak be removed from the list. "There was no investigation of Mayor Rybak. It's pretty straightforward," Luger says flatly.

In an email defending the inclusion of Rybak, Shields cited a press release the mayor sent in response to a Star Tribune story about a federal investigation at City Hall. The case resulted in the conviction of former council member Biernat on the charge of accepting free plumbing work, but nothing at Biernat's trial or in subsequent news stories implicated the mayor in any wrongdoing.

So what led to the inference that the mayor was being investigated by the feds?

In the press release, Rybak said: "We have no information from the U.S. Attorney's Office about the purpose or scope of the investigation. We will fully cooperate with the U.S. Attorney's Office if they request our assistance."

It's the kind of boilerplate copy any politician would issue, but to Shields, it was enough to lump Rybak in with the Democrats who had been targeted by the feds.

While it's possible Rybak was investigated, it's highly doubtful. Tom Heffelfinger was the U.S. attorney in Minnesota until his resignation last year. Under federal law, Heffelfinger says, he cannot comment on investigations that did not result in charges. But it doesn't take a scholar to read between the lines: "Rybak should be pissed off," Heffelfinger says. "That's irresponsible for someone to publish something like that." [Pissed off at Cragan and Shields, that is.]

And Rybak isn't the only Minnesotan to land on the list for questionable reasons. Cragan and Shields included Senator Klobuchar because she called the FBI on one of her campaign workers. That worker, whom Klobuchar promptly fired, had viewed an unaired political advertisement by using a password to gain access to the rival campaign's consultant website. But does Klobuchar's reporting of this relatively minor incident really signify that she was "investigated" by the U.S. attorney — and was thus a victim of "political profiling"?

Like Heffelfinger, David Lillehaug thinks it is next to impossible for outsiders to quantify non-prosecuted "investigations" at the Department of Justice. Lillehaug, a Democrat who served as Minnesota's U.S. attorney during the Clinton administration, says that U.S. attorneys should be judged on the basis of verifiable records—in other words, whom they prosecute.

"I'm glad they [Shields and Cragan] are looking at this, but it's a pretty complicated enterprise," Lillehaug says. "Identifying targets is very difficult. It would be more useful to look only at prosecutions."

Unfortunately, Cragan and Shields did not separate their tally of prosecuted public corruption cases from the more nebulous tally of officials who were "under investigation."

"I wish I had," Cragan says. "If we could redo it, we would."

Cragan and Shields are obviously partisan hacks. And unsurprisingly, they were bending over backwards to include as many Democrats as possible on the list, even though that led to spurious results in 3 out of 7 cases that City Pages investigated. Why should I trust that Cragan and Shields also went out of their way to include spurious listings of Republican local officials?

[Steve]

It means that if you're just counting up news stories that appear on Google (which is a moronic methodology under any circumstance), the news stories you find are more likely to involve local Democratic officials in those big cities.

This appears to me to be an assertion that you could test, but have chosen not to. You could, for example, devise a similar search to determine how many elected officials show up on Google during a particular time frame, period, putting the issue of an investigation aside. If a Google search generally returns 7 times as many references to elected Democratic officials as to elected Republicans, or something close to that ratio, I'd say the debunking would be settled at that point.

A lot of people seem to be invested in debunking this study. Have any of them attempted something like this?

(By the way, the answer to the homework assignement is that if you judge whether Dems are being targeted by referring to the ratio of Ds to Rs in the state legislature, then it makes no sense to start by eliminating every state where more Ds than Rs got investigated. You could have a state which is overwhelmingly controlled by Republicans, but somehow, just as many Democrats as Republicans get investigated; under the so-called methodology of this "debunking," this would get written off as a state where Dems obviously didn't get unfairly targeted.)

[John Doe]

This appears to me to be an assertion that you could test, but have chosen not to. You could, for example, devise a similar search to determine how many elected officials show up on Google during a particular time frame, period, putting the issue of an investigation aside. If a Google search generally returns 7 times as many references to elected Democratic officials as to elected Republicans, or something close to that ratio, I'd say the debunking would be settled at that point.

That's ridiculous. There are many millions of stories that turn up on any Google search for the relevant terms (6.6 million results for Democratic city officials, for example). Many of the links are obviously not relevant. Why on earth should I spend thousands of hours combing through these results just in order to produce a definitive disproof of a methodology that is so obviously bogus on its face? (Besides, as we now know for sure, Cragan and Shields were cherry-picking the results, even to the extent of including Democrats who were never investigated at all.)

[J Thomas]

"Like Mr. Maguire and FDL, I've yet to see Mr. Doe on TPM."

What is this even supposed to mean? That Mr. Maguire and FDL haven't seen me "on TPM" either? Well, no doubt they haven't.

It was a different argument. Maguire kept making obviously flawed arguments about a political topic, and people kept refuting him using facts and reasoning they found on the FDL blog. Then they started saying that since FDL had the experts, why wouldn't he go there and argue with them? Why keep arguing in the bush leagues on that topic? And his answer was basicly that they treated him like a troll there. They didn't take his arguments seriously so he didn't want to go back.

Using a similar reasoning, the question is why you keep pestering people here about this topic, harassing us with your quarter-truths and urban legends when you could go to a place where the other side already has their facts together. Why don't you go to TPM and present your case there?

I thought it was particularly funny that they'd say that, "John Doe", because they have absolutely no reason to think that you haven't been doing just that under another assumed name. For all we know you could have a program that switches words around some and reposts the same things all over under different names. You could be posting everywhere your particular urban legends show up, and they wouldn't know it was you at all.

[J Thomas]

Here's a City Pages article that offers a good example of why the Cragan/Shields methodology is crap:

....

Cragan and Shields are obviously partisan hacks.

Interesting. You read this thing and you suppose it means that Cragan and Shields are partisan hacks, when it is just as easy to suppose instead that Mike Mosedale (the author of the City Pages smear) is a partisan hack.

The natural conclusion from that is that you are a partisan hack, and your claims can be ignored on that basis.

[Anarch]

The natural conclusion from that is that you are a partisan hack, and your claims can be ignored on that basis.

For doubleplus laughs, this point has already been made to him. Any bets as to whether it will stick this time?

[John Doe]

harassing us with your quarter-truths and urban legends

Funny that you can't demonstrate that anything I've said is an urban legend. But it's interesting to know that you feel "harassed" at the mere sight of a link to an outside source of information (such as the City Pages article) that might upset your predetermined beliefs.

[moe99]

Mr. Doe, this is finally descending into the level of "I am rubber, you are glue."

[Matt Weiner]

Note also that John Doe ignored the quote from the New York Times that refuted his claim about why Iglesias didn't prosecute (the director of the DoJ election crime branch said he was neutral about whether he prosecuted or not!), and had no response to Steve's point that the NYT didn't claim that there were 3000 fraudulent registrations (the phrase was "3,000 faulty registrations, with some most likely resulting from mistakes"). If he can't be accurate about the very article he hyperlinked, why should we believe him on a single other thing?

I have no opinion about the Cragan and Shields study.

[John Doe]

the director of the DoJ election crime branch said he was neutral about whether he prosecuted or not!),

Um, that's supposed to be a point against me? What that really means is that Iglesias could have prosecuted the dang case. Word to the wise: Prosecutors are part of the executive branch, and the President gets the final say about what priorities they should place on prosecuting certain types of cases. That's why Clinton had a task force trying to investigate the possibility of prosecuting pro-life groups, for example. Setting investigative and prosecutorial priorities is what Presidents get to do.

the NYT didn't claim that there were 3000 fraudulent registrations (the phrase was "3,000 faulty registrations, with some most likely resulting from mistakes").

OK, fair enough, you caught me in a minor misstatement. Tje Times said, "3,000 faulty registrations," of which only some unknown number (could have been 3, for all we know) were "most likely" due to mistakes, with the rest possibly being due to fraud of some sort. So clearly, in a state that Gore won by 366 votes, there couldn't possibly have been an incentive for any funny business.

But on another note, how convenient that you have no opinion on the Cragan/Shields study. Why not? Are you aware of any other study that purports to measure the incidence of some activity over a 5 or 6-year period by seeing how many news stories or press releases pop up on Google? If you've ever seen any social science study at all, you'll be aware that something is amiss with such an approach, which is just unprecedented. And if you've actually taken a statistics course at any point, you'll be aware that studies based on cherry-picked data are not reliable. Why wouldn't you have an opinion about these most elementary matters?

[Steve]

There are many millions of stories that turn up on any Google search for the relevant terms (6.6 million results for Democratic city officials, for example).

Then you would include a date restriction, or limit the results in some other way. The point is that you assert Google is biased towards returning hits on Democrats, by a factor approximating 7:1, yet you're unwilling to suggest any methodology to actually test your theory. For someone who's so ready to criticize others' studies, that demonstrates a tragic neglect of the scientific method.

[gwangung]

But on another note, how convenient that you have no opinion on the Cragan/Shields study. Why not?

Because he'd like to study all of it? That's kinda reasonable.

Mr. Doe, stop being an idiot. You could very well be right (I repeat...you could very well be right), but you certainly are not approaching this with anything resembling proper methodology. Thowing around accusations of partisan hackery without knowing the study's methodology is the very antithesis of sound scientific practice---something you were excrociating the authors for.

That says far more about you than the authors.

[Matt Weiner]

As gwangung says, stop being an idiot. The Cragan-Shields study may be lousy or not; if it is lousy, it may still be the case that US Attorneys were prosecuting Democrats disproportionately; and if this is not the case, it doesn't diminish the evidence already presented that pressure was brought in specific cases.

Yes, it's a point against you that the director of the DoJ branch didn't tell Iglesias that he should prosecute the case. Are we supposed to think that the President set prosecuting election fraud as a priority for USAs and didn't tell the DoJ official in charge of election crime? No, we are not supposed to think that, unless we are idiots. So it's fair to conclude that, even given an emphasis on prosecuting alleged election fraud, the DoJ official thought this was a marginal case. Probably because absent a confession it was impossible to prove intent.

Incidentally, that letter you cited where one of the assistant USAs criticized Iglesias? One of the points of criticism was that Iglesias didn't speak out sooner about the improper pressure brought by Wilson and Domenici. Truly exculpatory of Republican misconduct!

Another point is that Iglesias is a Republican. Why he was supposed to have some incentive to cover up for the Democrats, I don't know.

[moe99]

We don't know if the search engine was google, or Lexis/Nexis (which is quite handy) or Westlaw. I have used all of them and each can produce reliable data, as long as cogent restrictions are used in the search engine. I have seen no factual basis yet for criticising the search other than unsubstantiated allegations of what engine was used.

[John Doe]

Then you would include a date restriction, or limit the results in some other way. The point is that you assert Google is biased towards returning hits on Democrats, by a factor approximating 7:1, yet you're unwilling to suggest any methodology to actually test your theory. For someone who's so ready to criticize others' studies, that demonstrates a tragic neglect of the scientific method.

Do you know how to do Boolean searches on Google restricted to news stories that appeared in 2002? It would be news to me if that were possible.

Just to be clear, though, I don't know that Google is "biased" 7-1 in favor of mentions of Democrats. It could be biased by some smaller magnitude, and the rest of any disparity could be because of actual differences in the level of corruption among local officials.

But the relevant point is, Cragan and Shields are the ones who came forth with a startling claim, but no rational person who has any familiarity with social science would think that they have proven their case. They've admitted that they had "no data" outside of Google searches, and we now know that they were counting certain Democrats as having been "investigated" when no such investigation ever happened.

Thowing around accusations of partisan hackery without knowing the study's methodology is the very antithesis of sound scientific practice

Please. I know quite a bit about the methodology. Indeed, I'm the only person around here who has shown any curiosity about the methodology, and any ability to do additional research in that regard.

But this research is necessary only because the authors didn't even take the minimal step of describing their methodology in the "study" itself. Really, if you don't know how unusual and unprecedented this is, you have no idea what you're talking about. Every last social science article has a section describing the methodology being used -- from the econometric to the ethnographic. It's always up to the authors of a study to show their work.

And it's not "sound scientific practice" to blindly accept anything that is labeled a "study" and to try to shout down anyone who digs into what methodology was used.

[John Doe]

We don't know if the search engine was google

Yes, we do. We know this because, as I have repeatedly pointed out, a journalist asked them, and they said that they used Google. We also know this because when they responded to said journalist, they didn't deny having told him that they used Google.

Can we have a moratorium on saying that "we don't know" things that have been proven, if not to a 100% certainty, at least to a certainty that is far and beyond anything else that has been alleged about the entire U.S. Attorneys issue?

[John Doe]

The Cragan-Shields study may be lousy or not; if it is lousy, it may still be the case that US Attorneys were prosecuting Democrats disproportionately; and if this is not the case, it doesn't diminish the evidence already presented that pressure was brought in specific cases.

True enough. Their claims could be unsubstantiated but still true -- an unjustified true belief, as philosophers might say (but you knew that).
But since that's purely a theoretical possibility, people shouldn't be bandying about, as if it were a proven fact, the claim that Bush prosecutes Democrats at a 7-1 ratio. Why not direct some of your ire in the direction of these gullible fools, rather than just parachuting in to take potshots at me for being too insistent on seeing the evidence?

So it's fair to conclude that, even given an emphasis on prosecuting alleged election fraud, the DoJ official thought this was a marginal case. Probably because absent a confession it was impossible to prove intent.

OK, fair point. But Iglesias could still have been fired for poor management issues. You seem to be implying that either he was legitimately fired for failing to prosecute real voter fraud, or else he was fired for unlawful or unethical political reasons. This is a false dilemma.

[J Thomas]

But Iglesias could still have been fired for poor management issues. You seem to be implying that either he was legitimately fired for failing to prosecute real voter fraud, or else he was fired for unlawful or unethical political reasons. This is a false dilemma.

Sure, and it could have been both. But the preponderance of evidence so far indicates he was fired for insufficient loyalty to Bush.

Where this is leading is that over the next 2 years USAs are not going to be fired for disloyalty unless there's a very solid case for the cover story. The control is eroding away faster and faster.

Bush might possibly succeed in destroying the Republican party. No certainty. It's only a few years ago they were talking about destroying the Democratic party for good. But look at the gerrymandering. People talk like it's an evil trick that works for evil people. But it can backfire. I'll describe the ultimate exaggerated case to make the logic plain. Say your side has 1/3 the voters and the other side has 2/3, but you get to draw up the voting districts. And you draw them so that they get 100% of the votes in the districts they win, while you win each district by a small margin. So they win 1/3 of the districts with 100%, and you win 2/3 of the districts with tiny margins. You have 1/3 of the popular vote and 2/3 of the wins. Sounds good, right? But then what if something happens to make your party a tiny bit less popular than it was before. Instead of winning 2/3 of the seats by tiny margins, you lose all of them, 1/3 by 100% and 2/3 by tiny margins.

And right now, that's where the Republican Party is headed. They have no possible argument why a patriotic american should vote for them except the claim that no matter how bad they are, the democrats are even worse. This is not an argument that patriotic americans can stomach. If they can publish it well enough, the result will be that the Democratic Party will also become a third party.

Bush said he had political capital and he was going to spend it. That's the kind of rich guy Bush is, he spends his capital. He appears to be planning to spend the very last of his political capital around December 2008, leaving no inheritance at all for the next guy.

So I really don't understand at all why guys like you are still loyal to him. Does he pay you well? If you guys don't make a big show of reforming the GOP, cleaning up the corruption and all that, there may not be a GOP in a few years. Why would anybody contribute to the Republican Party if the party can't give out patronage? The libertarians are going to take your place, and you can join them but you'll be pretty far down the totem pole, at the far end of the trough.

So why are you so blindly loyal?