January 2005, Part 2

Jim Miller on Politics

Pseudo-Random Thoughts

Living In A Danger Zone:  Your chances of dieing in a natural disaster are low.
Yes, it's a forbidding world out there. But we can take some comfort if we remember that natural disasters are rare and the chance that any one of us will perish in a tidal wave, hurricane, earthquake or other natural calamity is very, very low -- even though studies conducted over the past 25 years consistently have found that most people believe the odds are very, very high.

In fact, statistically speaking, a person is more likely to die by falling from a tall building, slipping in the bathtub or being legally executed than to perish in an earthquake, flood or "cataclysmic storm" such as a hurricane, according to the latest estimates by the National Safety Council derived from 2001 data from the National Center for Health Statistics and the U.S. Census Bureau.
And suicide is a far more common cause of death than all of those combined — especially for white males, which bothers me a little from time to time.  Car accidents are next, exactly as you would expect, and pedestrian deaths are more common than I would have guessed.

But you can increase your odds of dieing in natural disaster by seeking out danger; mountain climbers do that routinely — and often pay the price.  But people also increase their chances of dieing by living in places with beautiful views.
La Conchita, Calif., Jan. 13 - Ernie Garcia, 78, who has lived here in the shadow of an unstable sea cliff since 1988, ignored warnings to evacuate before Monday's fatal mudslide and has refused to leave ever since.
I have never found a general rule for controlling how much risk people should be allowed to accept for themselves that satisfies me.  I am inclined to let adults such as Garcia take whatever chances they want to take.  If he wants to live underneath a potential mud slide, that's fine with me, as long no children visit him.  But even that libertarian position doesn't work entirely because his death in a slide would almost certainly impose heavy costs on the taxpayers.  If he posted a bond to cover such costs, then I suppose that would solve that particular problem.

But that solution doesn't always work, as we have learned from mountain climbers.  When they need rescue, we don't ask if they can afford to pay for the very expensive (and sometimes very dangerous) efforts to rescue them, though we do sometimes try to recover some of the money later.  But as long as we do that, some climbers will take risks that they wouldn't otherwise.

A few of these problems have what I think are simple solutions.  For example, we shouldn't give federal aid to those who choose to build expensive houses on beaches that everyone knows will be hit by hurricanes in the next few decades.

But many such problems do not.  Consider, for example, a young man who refuses to stop drinking and driving.  Even if he does it in a way that does not endanger others, the chances that he will cripple or kill himself are high.  In either case, the public loses.  You can think of many similar examples.  What, for instance, should we do about a man who risks AIDS by reckless behavior?  The best answer I have come up with over the years is that we should try to discourage such people and then, however grumpily, take care of them when they don't take our good advice and suffer the consequences.

That's my position, but I would be glad to change it if someone showed me a better one.
- 8:01 AM, 16 January 2005   [link]

How Many Felons Voted Illegally In Washington's Gubernatorial Election?  I don't know, and, as far as I can tell, neither does anyone else.  But there are some numbers from the Florida 2000 election that suggest that illegal votes from felons may have provided Christine Gregoire's 129 vote margin.

First, the basic data from John Fund's Stealing Elections.

Both the Miami Herald and the Palm Beach Post found that, if anything, county officials were too permissive in whom they allowed to vote, and that this largely benefited Al Gore.   An analysis by the Post found that 5,600 people whose names matched the names of convicted felons who should have been disqualified were permitted to cast their ballots.  "These illegal voters almost certainly influenced the down-to-the-wire presidential election," the Post reported.   "It's likely they benefited Democratic candidate Al Gore.  Of the likely felons identified by the Post, 68 percent were registered Democrats."

The Miami Herald found a similar number with a similar search.  Both newspapers endorsed Gore, and I think it fair to say that the Palm Beach Post is one of the most Democratic newspapers in the country.

There were this many felons on the rolls in spite of a 1998 Florida law requiring county election boards to remove felons from the rolls.  (The bill was sponsored by two Democratic legislators and signed by Jeb Bush's Democratic predecessor, Lawton Chiles.)  Some of the county boards had tried to follow the law; others, mostly Democratic, had not.

First, a mildly technical point on the search:  Matching names (and birth dates) this way would provide a few false positives, citizens who were unlucky enough to share the name and birth date with a felon, but would miss felons who changed their names after they were convicted of a felony.  Most likely, more than 5,600 felons voted in Florida in 2000, perhaps thousands more.

Next, how many votes net did these 5,600 felons give Al Gore?  Unfortunately, the Post gives us only the number of Democrats, so we can't use the simplest method to estimate that number, giving Gore all the Democrats, half the independents, and none of the Republicans.  We can use that simple procedure to set bounds.  If we assume that all of the remaining 32 percent are Republicans, then we could estimate Gore's net gain at 2,016 votes.   If we assume that all the remaining 32 percent are independents, then using the same simple method, we would estimate that Gore gained 3808 votes from these 5,600 felons.  No doubt the actual number is somewhere between those two estimates.  We would probably not be too far wrong if we were to say that Gore gained nearly 3,000 illegal votes from these 5,600.  (And who knows how many from felons not found in the computer search.)

At this point it would be tempting to simply take that 3,000 and adjust for Washington's smaller population, about 3/8 of Florida's, giving us a net gain for Gregoire of about 1,000 illegal votes.  Tempting, but wrong, because it does not take into consideration important differences between Florida and Washington.  Florida's crime rate is somewhat higher than Washington's, so there are, most likely, proportionately more felons in Florida.

More important, the two states have different laws on the restoration of voting rights to felons, Florida making it much more difficult than Washington.  As I recall, Florida felons actually have to petition the governor to have their voting rights restored.  Washington gives felons several ways to restore their voting rights:

When are felons eligible to vote again?
All persons convicted of a felony who have met the conditions of their sentences may have their voting rights restored a number of ways.
  • Upon completion of their sentences, the Department of Corrections may restore the persons' rights through the court of their original convictions.
  • These persons may return to the court of conviction and petition the judge to restore their rights via court order.
  • They may receive a pardon from the Governor.
  • If their crime was committed before July 1, 1984 they may contact the Indeterminate Sentence Review Board and ask for restoration of their rights.

I would guess that the first is the most important.  A casual search of the Department of Corrections site found no information on their policy on the restoration of voting rights.  I have heard both that they do it routinely, and that they make it conditional on restitution, a condition not met, at least immediately, by many released felons.  I plan to write to the Department of Corrections asking for a description of their policy; meanwhile, if any of you happen to know the answer to this question, I would appreciate hearing from you.

We also should not assume that Washington's counties are as bad at clearing felons from their rolls as counties in Florida.  I have many criticisms to make of the King County elections office, but they seem to do far better than, for example, the election board in Florida's Broward County.  And there are other differences that could affect my estimate of Gregoire's gain from illegal felon votes.  The two states have a different ethnic mix, so it is plausible that felons in Washington state are not as heavily Democratic as Florida felons.

So I don't think that Gregoire received a net gain of 1,000 illegal votes from felons.   But, even allowing for all the factors that I have mentioned (and a few I haven't), I do think it entirely possible that Christine Gregoire received a net gain of 130 illegal votes from felons.   (There's a small irony in that, given her long service as Attorney General.)  And Democratic strategists may agree with that, given the party's resistance to letting Republicans do the same kind of search done by the Palm Beach Post and the Miami Herald.

Cross posted at Sound Politics.

Two Follow Ups:  After I posted my discussion of the accuracy of Washington state's manual recount, I saw two issues raised in comments over at Sound Politics that deserve follow ups.

First, some wondered about the existence of systems that would print paper ballots for voters.   As far as I know, no such systems exist, but I sketched some ideas for that kind of system in this post.  (I doubt that I was being original, but I haven't seen similar proposals elsewhere.)  The basic idea is to use machines to create paper ballots that can be read both by voters, and by separate vote counting machines with very high accuracy.  And I should add that I received some interesting suggestions from a partisan Democrat for improving on my sketch.  I am still thinking about the problem and would welcome more comments.

Second, I saw the claim, more than once, that an MIT/Caltech study had shown, contrary to what we would expect and most election officials believe, that hand counting of votes is more accurate than machine counting.  Here's the key quote from the study:

Have we made progress?  Do machine counts improve on hand counts?  At least in the comparison of optical scanning and paper, the answer is yes.  Historically, there is about a 1 percent difference between initial counts and recounts when ballots are tabulated by hand.   The discrepancy between initial counts and recounts falls to about .5 percent with the optically scanned ballots.

That's still way too inaccurate in my opinion, but it seems clear which method is better.   (Those who want to look at the study should also look at my criticism of one of their metrics, often misused by Democrats in the arguments over this last election.)

Cross posted at Sound Politics.
- 7:33 AM, 14 January 2005
Correction:  The link to the MIT/Caltech paper origninally pointed to Working Paper 1, rather than Working Paper 10.  I have corrected it above.
-8:12 AM, 16 Janaury 2004   [link]

Michelle Malkin Gets Hate Mail:  Don't read her post with samples if you are easily offended.  I won't quote from them because I want to keep this a family friendly site.  (You can see some of the milder samples in her column on the subject.)

Here's Malkin's summary from the column:
Rabid liberal elitists expect and demand that "women of color" in public and political life adopt their left-wing political orthodoxy.  When we don't accept such tripe, their racist and sexist diatribes against us are unmatched.
If you read the email she has received, you will find it hard to disagree.  Far too many on the left believe that one's ideas should be determined by one's race or sex.  And they aren't at all kind to members of minorities or women who disagree with them.

(As I have mentioned before, since I started this site I have received only one piece of hate mail — and that one was more funny than anything else.  Part of the reason for the lack of hate mail (which is fine with me) may be my somewhat milder way of stating my arguments.  But I don't doubt that, if I were a minority woman instead of a white man, I would have received some of the same kinds of attacks that Malkin has.)
- 5:00 AM, 14 January 2005   [link]

Was The Manual Recount More Accurate Than The Machine Recount?   That's the claim made, again and again, by supporters of Washington state's "governor", Christine Gregoire.  (They also say, again and again, that Washington's laws provide for a manual recount — which is not in dispute.)  They make this claim in spite of the fact that nearly all authorities on counting ballots — including King County's election head, Dean Logan — believe that manual counts are less accurate than machine counts.  And just plain common sense should tell you the same thing; we humans are better than our machines at many things, but tabulating millions of similar objects is not one of them.

To answer that question for certain, we would have to have at least one more recount.  But we can get a tentative answer just by looking at the changes between the machine recount and the manual recount, and judging whether they are what we would expect.  What I am doing, as some of you may already have realized, is throwing away everything we know about the strange happenings in the manual recount — for the moment — and just looking at the final numbers for the counties.

Before going through the numbers, let me lay out the ideas behind my thinking.  Machines are more accurate in vote counts, but they miss some ballots that people can read.  People are likely to be biased in their counts, even if they do not intend to be.  Democrats are more likely to err in marking their ballots, since Democrats draw more support from illiterate and partially literate voters.

When I put all these factors, except for the bias, together, I get this generalized prediction for the changes between the machine recount and the manual recount.  In each county, I would expect small gains for each candidate, almost proportional to their votes in each county.  I say almost because I expect that Democrats will make a few more errors than Republicans, errors that machines miss, but people can correct.  Some small counties, of course, will show no changes at all.   There will be some statistical fluctuation as well, so I don't expect exact fits to the model.   Finally, because of the bias (whether conscious or not), I expect Republican counters to slightly favor their party and Democratic counters to slightly favor their party.

At this point, to follow my discussion, you will want to print the election results, or open them in a separate window or tab.  Let's take the easy counties first.  Twelve counties, Chelan, Columbia, Douglas, Ferry, Garfield, Kittitas, Klickittat, Lincoln, Pacific, Pend Oreille, San Juan, and Stevens, had gains of two votes or fewer for each candidate, so let's eliminate them from further discussion.  Skamania almost passes this test; Rossi gained three votes there and Gregoire one, so I'll set it aside, too.

There were nine more counties, Benton, Clallam, Clark, Gray's Harbor, Jefferson, Mason, Thurston, Whatcom, and Yakima, in which the gains for the candidates were both (relatively) small and close to proportional to the votes of each candidate.  (By relatively, I mean less than 1 in a 1,000, compared to the candidate's vote.)  Five other counties, Lewis, Island, Spokane, Walla Walla, and Whitman, had small gains that were not proportional to the candidates' votes.   Since two, Island and Whitman, showed bigger gains for Gregoire than we would expect, and the other three show larger gains for Rossi than we would expect, I believe the results in these five to be just the random variations I expected, and will set them aside, too.

There were three small counties, Asotin, Franklin, and Okanogan, that showed, relative to the total number of votes, large gains for both candidates.  The gains were close to proportional, so I see this as evidence of careless vote counting in the first recount, rather than something sinister.

Five counties, Adams, Cowlitz, Grant, Skagit, and Wahkiakum, managed to lose votes between the machine count and the manual count.*  That shows that something was wrong with either the machine recount or the manual recount in those counties.  (I think in some cases, the counties have explanations for their errors.)  Rossi lost 14 votes, net, in these counties, and Gregoire lost 24 votes.

That leaves four counties, King, Kitsap, Pierce, and Snohomish.  All produced results that are, on their face, suspicious.  Kitsap's gains for both candidates were too large, though they were close to proportional.  Snohomish county, perhaps because it uses electronic voting machines, did not make that error, but it did produce an improbably large gain for Gregoire in a county that Rossi carried.**  Pierce County produced too large gains for both candidates, and a suspiciously large gain for Gregoire, again in a county that Rossi won.  If I were a paid Republican investigator, I would look hard at the manual recount in Pierce.

Finally, we come to King County, saving the best for last.  Let me start with what I consider the most remarkable point of all.  In December, after the court fight, King County added, and I think I have this number right, 566 ballots from those that had been rejected because they did not have the signatures in the computer file.  Between them, Rossi and Gregoire gained just 537 votes between the machine recount and the manual recount.  No doubt a few of those 566 ballots had no votes for governor, but the difference seems too large to me, given that about 2.5 percent of the state's voters did not make a choice for governor.  And it leaves no room at all for the votes that a human counter can see, but a machine can not.   It seems certain that King County lost some votes between the machine count and the manual recount.

Let me remind you of how we expect a manual count to be different from a machine count.  We expect small gains for each candidate, roughly proportional to their vote totals.  Pierce County produced implausibly large gains in the manual count; King County produced an equally implausible loss, after we allow for the 566 additional ballots.  How did that happen?  I am not sure though Stefan has certainly given us some cases to look at.  King County's result is remarkable; they were trying as hard as possible — so they say — to count every vote in the manual recount, and they ended up with fewer votes than in the machine recount.

(Some Republicans will find the disproportionate gain for Gregoire in the manual recount suspicious.   I am not certain it should be, since it appears to have come entirely, net, from the set aside absentee ballots, so the same argument that applied to the additional Snohomish ballots** might apply here, too.)

So, was the manual recount more accurate than the machine recount?  Probably not, given the results in Pierce and King counties.  That said, I think the important problems are not in the count, the machine recount, or the manual recount, but in the registration and the voting before the counts.

Cross posted at Sound Politics.

(*All of the counties that lost votes between the machine recount and the manual recount use the E&SS optical scan systems.  Those systems may be a little too good at counting every vote.   For more on other effects of particular voting systems, see this Timothy Goddard post.

**Since Snohomish counts election day votes electronically, these gains come only from absentee ballots, which may explain why they tilted to Gregoire.  There was a late surge to Rossi, according to the polls.  Since those who vote with absentee ballots often vote early, I would expect to see less of the surge there.  To put it another way; if we counted only the election day ballots, Rossi would probably have won easily, but would have lost handily if we counted only absentee ballots.)
- 4:18 AM, 13 January 2005   [link]

Worth Reading:  Norman Podhoretz's long discussion of the George W. Bush's strategy in the war on terror.  Podhoretz explains why he expects Bush to continue on the course he set after 9/11, describes the many opponents of Bush's strategy, (isolationists on the left and right, superhawks, liberal internationalists, and realists), and outlines the strategy the opponents share — doing everything possible to turn Americans against the war with terrorism by painting it as a failure.

And what will happen if Bush appears to be succeeding in Iraq?  Will Bush get credit from the realists?  Here's Podhoretz's prediction:
I think not.  I think they will do unto a success in Iraq what they did when Hamid Karzai was sworn in as the president of Afghanistan this past December.  In a powerful report on how the press chose to cover that story, Peter H. Wehner of the White House Office of Strategic Initiatives reminds us of what the realists always said about Afghanistan: that it "was too backward; too fractious; too medieval and religiously fanatical; and too ungovernable to ever move toward democracy."  Yet only three years after the war to liberate Afghanistan from the horrific Taliban regime, "a free election was held and a civilized, modern, pro-American president was sworn in."  Wehner then describes how the press treated what he calls "this momentous event":
The New York Times carried the story on page A8.  The Washington Post carried the story on page A13.  USA Today had the briefest mention possible on page A5.  The Los Angeles Times carried the story on page A3.
But merely burying the story was not good enough for the news pages of the Wall Street Journal (whose point of view is much closer to that of the New York Times and the Washington Post than to the conservative position of the Journal's own editorial page).  The paper's coverage, carried in the "What's News" column, consisted entirely of a one-sentence mention that "Karzai was sworn in as Afghanistan's president," immediately followed by this: "Taliban rebels attacked a military base near the Pakistani borders, killing four soldiers. U.S. troops killed two assailants."  And the Los Angeles Times went the Journal one better by taking the occasion to dwell on how much opium is still being produced in Afghanistan.
As if it was news that opium was being produced in Afghanistan.

And, as I am sure you know, these newspapers often set the agenda for the major networks.  So, if you watch network television, expect to see many gloomy stories out of Iraq in the next few months — especially if Bush appears to be succeeding.
- 7:28 AM, 13 January 2005   [link]

Not All Mammals Were Oppressed During The Cretaceous:  Dinosaur predators ate mammals during their long period of dominance.  Now it seems that at least a few mammals returned the favor and enjoyed baby dinosaur for dinner.
When the dinosaurs ruled the world, the mammals hid in the shadows, daring to grow no bigger than shrew-like insectivores that hunted at night.  Or so we thought.

Two stunning new fossils from China have overturned this preconception.  Not only did large mammals live alongside their giant reptilian cousins, but some were big and bold enough to go dinosaur hunting.

Named Repenomamus giganticus and Repenomamus robustus, the sturdily built mammals lived in China about 130 million years ago, around 65 million years before we thought their kind inherited the Earth.  At 1 metre long, R. giganticus was big enough to hunt small dinosaurs, and a newly discovered fossil of its smaller cousin, R. robustus, died with its belly full of young dinosaur.
Any political point?  None that I can think of, but this should provide some new metaphors for those who need them.
- 6:41 AM, 13 January 2005   [link]

How Close Are The Democrats And The "Mainstream" Media In Washington?   They sleep together.
Mona Lee Locke, Washington state's outgoing first lady, will return to her television-news roots when she joins KIRO-TV in April, the station announced yesterday.
That's Dan Rather's station, for those not familiar with this area.  And KIRO wasn't the only station that offered her a job.

As I recall, Mona Lee Locke kept her job at KIRO while her husband was King County executive, in spite of the possible conflicts of interest.
- 3:18 PM, 12 January 2005   [link]

An Example Of Distributed Vote Fraud:  While doing the previous post, I ran across this very clear example of what I have begun to call distributed vote fraud.  The New York Daily News found last August that 46,000 people were registered to vote in both Florida and New York.   Of those 46,000, 68 percent were Democrats, 12 percent were Republicans, and 16 percent were neither.  (And no, the newspaper does not explain why that doesn't add up to 100 percent.)

No Democratic party leader organized these voters — at least as far as I know.  Instead, they decided, by themselves, perhaps after talking with friends, to vote twice.  And, just as I argued below, far more Democrats than Republicans decided, on their own, to commit vote fraud.

Anyone want to guess how many fraudulent votes Al Gore received in Florida in the 2000 election from this group?

Cross posted at Sound Politics.
- 2:43 PM, 12 January 2005   [link]

Are Democrats More Likely To Commit Vote Fraud Than Republicans?  Nearly all informed political observers would say yes.  In his book on vote fraud, John Fund is apologetic about mentioning that, because he wants to make a general argument.  Here's how he begins the discussion:

A note about partisanship: Since Democrats figure prominently in the vast majority of examples of election fraud described in this book, some readers will jump to the conclusion that this is a one-sided attack on a single party.  I do not believe Republicans are inherently more virtuous or honest than anyone else in politics, and I myself often vote Libertarian or independent.

He then notes that Republicans have had less chance to commit vote fraud because they controlled fewer "local and administrative offices".  (Though Republicans have, as recently as the 1980s, sometimes used intimidation tactics that are certainly unethical, though perhaps not illegal.)   Fund then makes a more general argument:

In their book, Dirty Little Secrets, Larry Sabato and co-author Glenn Simpson of the Wall Street Journal noted another factor in why Republican election fraud is less common.   Republican base voters are middle-class and not easily induced to commit fraud, while "the pool of people who appear to be available and more vulnerable to an invitation to participate in vote fraud tend to lean Democratic."  Some liberal activists that Sabato and Simpson interviewed even partly justified fraudulent electoral behavior on the grounds that because the poor and dispossessed have so little political clout, "extraordinary measures (for example, stretching the absentee ballot or registration rules) are required to compensate."  Paul Herrnson, director of the Center for American Politics at the University of Maryland, agrees that "most incidents of wide-scale vote fraud reportedly occur in inner cities, which are largely populated by minority groups."

Democrats are far more skilled at encouraging poor people — who need money — to participate in shady vote-buying schemes.  "I had no choice.  I was hungry that day," Thomas Felder told the Miami Herald in explaining why he illegally voted in the Suarez-Carollo mayoral election.  "You wanted money, you were told who to vote for."  A former Democratic congressman gave me this explanation of why voting irregularities more often crop up in his party's back yard.  "When many Republicans lose an election, they go back into what they call the private sector.  When many Democrats lose an election, they lose power and money.  They need to eat, and people will do an awful lot in order to eat."

(Sabato is a Democrat; I don't know about Simpson or Herrnson.)

So Democrats are more likely to commit vote fraud because more of their adherents are poor enough to be bribable, because some activists will cross the line to help the poor, and because many Democratic politicians have no good alternative to public office.

These points are, as I said at the beginning, not something most informed observers would quarrel with.  But I think, before going farther, that I should make it clear how far the argument goes.  That more Democrats commit vote fraud than Republicans does not mean that most Democrats commit vote fraud.  I am sure that very few Democrats commit vote fraud in fact — but even fewer Republicans.

Some readers will prefer direct evidence to the conclusions of experts, however well informed.  I have that, too.  I do not know of a single major Republican vote fraud scandal in the last ten years.  But it is easy to find Democratic scandals, in Baltimore, Philadelphia, New Orleans, Milwaukee, and New Orleans.  You may recall one example from Milwaukee; in 2000, a Democratic socialite was caught exchanging cigarettes for the votes of the homeless.  (She's still an honored member of the Democratic party, by the way.)

(The 1998 Miami vote fraud scandal during the 1998 mayoralty race between Xavier Suarez and Joe Carollo is hard to classify by party.  Carrollo is and was a Republican.  At that time Suarez, who received many fraudulent votes, was an independent, though he has belonged to both parties.)

Democrats are the guilty in most of the smaller vote scandals, too.  I have started collecting these as they are reported.  Here are examples of vote fraud, or charges of vote fraud in San Francisco, East Chicago, Passaic, New Jersey, Orlando, Florida, South Dakota, New York and Florida, Cleveland, Colorado, Florida, Minnesota, New Mexico, Ohio, and Pennsylvania., Nevada, New Mexico, Orange County, California, and Kentucky.  Of these cases, only the Nevada case was purely Republican.  (A former employee of a Republican firm charged that some Democratic registrations collected by the firm had been discarded.  I don't know whether the case has gone to trial.)  The Kentucky case appears to have been bipartisan, though I did not make that clear in my post.

When vote fraud is detected, those caught are nearly always Democrats.  Either Democrats are much less skillful at fraud, which seems implausible to me, or they commit far more of it, just as the experts say.

Some Democratic leaders have tolerated vote fraud fairly openly, including Bill Clinton, when he was governor of Arkansas.   (Some believe that the rush by the Clinton administration to get new citizens in 2000 was part of an effort to stuff the ballot box, something I did not mention in the post.  It was certainly true that, under pressure for Clinton and Al Gore, the INS skipped many checks on would-be citizens in 2000.)

Finally, Democratic leaders behave as if they believe more Democrats commit vote fraud.   Nearly always, when the two parties split on election rules, the Republicans want more checks on fraud and the Democrats want less.  The infamous 1993 "Motor Voter" Act, which did so much to make fraud easier was opposed almost entirely by Republicans and had been vetoed by the first President Bush.  I don't say that all supporters of the legislation (including Washington's Maria Cantwell) even knew that it would make vote fraud easier, but some of them did.  Like the anonymous liberal activists, they see some fraud as a reasonable price for getting more representation for the victim groups they identify with.

It is telling, I think, that there is one group, military voters, for which Democrats tend to prefer tougher rules and the Republicans easier rules.  Military voters generally back Republicans, at least in recent years.  That Democratic leaders prefer rules that make cheating easier (for everyone except military voters) is understandable if they think they gain from the cheating, but hard to explain otherwise.

(This is the promised follow-up to an earlier post outlining distributed vote fraud.   In that earlier post, I left for another time this explanation for my belief that Democrats are more likely to commit vote fraud.)

Cross posted at Sound Politics.
- 12:26 PM, 12 January 2005   [link]

Nearly All Investigative Reporters Are Leftist Nuts:  That's the opinion of Jack Shafer, who covers the press for Slate.

I don't know that I've met more than four or five investigative journalists in my life who didn't wear their political biases on their flapping tongues.  Almost to a one, they're suspicious (paranoid?) about corporate power, dubious about the intentions of governments, and convinced that at this very moment a secret meeting is being held somewhere in which a hateful conspiracy against the masses is being hatched.  I won't provoke the investigative-journalist union by alleging that most of its members are Democrats or lefties, but aside from a few right-wing reporters sucking conservative teats inside the government, how many Republican investigative aces can you name?

Far from being a handicap, political bias appears to be a necessity for the investigative reporter.   On one level, you've got to admire Mapes for rejecting all the mounds of evidence assembled by hundreds of other reporters who tried and failed to conclusively prove that Bush got a special service deal.   For all Mapes' faults—and the panel documents her failings by the bushel—the panel still found that her colleagues "highly" regarded her. (One worries about the "lowly" regarded producers at CBS News.)

Shafer argues that these crackpots must be controlled by saner people in the organization, and that's where CBS failed.

If Shafer is right in this argument — and I have my doubts about parts of it — then CBS should pay for an outside inquiry on all the stories that "60 Minutes" has done in recent years, not just those done by the fired four.  If managers failed to control Mary Mapes, and all investigative journalists are like her, then CBS almost certainly failed with other producers as well.  This implies that CBS created — and I use created deliberately — and broadcast many other leftist fantasies, besides the bogus National Guard story.  There is some evidence that Mapes was less controlled than other producers, so I think the investigators should begin with her stories.

Which parts of Shafer's argument make me skeptical?  I accept his conclusion that nearly all investigative journalists are leftist nuts since Shafer has far more experience with those people than I do.  But I don't agree that being a leftist nut is a requirement for the job, or even desirable.  Two of the best investigative journalists I know, John Tierney of the New York Times and Byron York of the National Review, are neither nuts nor leftists.  (I'm not sure just what political views Tierney has, vaguely libertarian at a guess.)

Shafer's claim that nearly all investigative journalists are leftist nuts supports an argument that I have made here from time to time:  Much of the bias in our "mainstream" media comes from stories that are not covered.  Humans being imperfect, if investigative journalists were to look for for faults in leftist individuals or scandals in leftist organizations, they would find them.   I don't doubt, for example, that reporters for the two Seattle newspapers could find more faults with King County executive Ron Sims, or the Seattle city government — if they took the time to look.   And I am sure the same is true in many other urban areas.

Cross posted at Oh That Liberal Media.
- 5:34 AM, 12 January 2005   [link]

This Discovery  is so cool.
Astronomers are highly confident that they've taken the first photograph of a planet outside our solar system.
. . .
The planet candidate appears to orbit a failed star known as a brown dwarf.  The initial observations at ESO's Very Large Telescope could not determine whether the apparent planet was actually at the same distance as the brown dwarf or if it was a background object.  The Hubble observations show that the two indeed appear to be travelling together through the sky, suggesting they are gravitationally bound, as originally suspected.
Literally as well as figuratively, since a planet orbiting a brown dwarf is not going to be very warm, especially when it appears to be 30 percent farther away from the dwarf than Pluto is from our sun.
- 3:35 PM, 11 January 2005   [link]

If A Substantial Part Of A Nation's Population Is Anti-Semitic , then some of the nation's politicians will find a way to appeal to that part, however indirectly.  That's what the Conservatives are charging has happened in Britain, as Labour tries to regain the Muslim vote.
A senior minister was forced to deny last night that he had been anti-semitic in accusing Michael Howard and another Jewish MP of failing to stand up for British Muslims.

Mike O'Brien, the Energy Minister, issued a hasty statement after Mr Howard demanded an apology for remarks made by the minister in an emotionally charged appeal to Muslims to support Labour.
That a very large part of the Muslim population in Britain is anti-Semitic is not a secret, nor is the fact that Muslims have moved from their traditional home in the Labour party toward the Liberal Democrats, in order to punish Tony Blair for his efforts to liberate millions of Muslims from tyrants in Afghanistan and Iraq.  Whether O'Brien intentionally crossed a line is something I do not know.  But it is hard to avoid slipping into bigotry when you appeal to bigots.

(Last year, the departing Israeli envoy charged that anti-Semitism is common on the British left.
Israel's outgoing ambassador to Britain launched a stinging attack last night on Left-wing British politicians, accusing them of encouraging anti-Semitism.

Zvi Shtauber said he had been shocked during his three years in London by the anti-Semitic views of some Left-wing MPs and activists.

"All along we were afraid of the Right.  Now there is an unholy alliance between the Left and Islamic fundamentalists," Mr Shtauber told the Observer.
I am not an expert on British politics, but I think he is right in that observation, though he might want to add "some of" before "the Left".  And Americans should not sneer, because very much the same alliance is developing here.)
- 10:58 AM, 11 January 2005   [link]

The Newspaper Reactions  to the CBS report are interesting, to say the least.  The New York Times article is written from the point of view of the other CBS News employees.
"This should never have happened," said Leslie Moonves, the CBS chairman who announced the dismissals of four executives yesterday in the wake of an independent panel's report.  The panel found that the "60 Minutes" program that dealt with President Bush's service in the Texas Air National Guard was unfair and misleading after being rushed to broadcast without proper vetting.

"This is a rude awakening for CBS News," Mr. Moonves said, "and the CBS News culture has to change."

What exactly that will mean is still uncertain, though several staff members reported the morale in the department to be devastatingly low.  "We are all sad and miserable," said one CBS production staff member, who spoke on the condition of anonymity to protect against criticism from superiors at the network.
Yet even the New York Times is able to connect the very large dots and gently raise the question of bias.
Ms. Mapes, who lives in Texas, was also known inside CBS for her long-time aggressive coverage of President Bush, going back to his days as governor.  Though Mr. Moonves and other CBS executives yesterday pointed to the panel's exoneration of the network on charges of political bias against the president, not everyone agreed that it played no role at all.

"It sounds like you had a star reporter here who fell in love with a story," Mr. [Alex] Jones [the director of the Shorenstein Center on Press and Politics at Harvard University] said. "Her previous work had given her a reputation sufficient to bowl over everyone else.  It seems like it was a combination of competitive pressure, hubris and a little politics.  I think it's foolish to separate this entirely from politics, no matter what the report says.  All in all that's a witches' brew."
Just a little politics.

The Washington Post story has an entirely different point of view on the story, seeing it as another battle in a titanic war between Bush and the "mainstream" media.
President Bush was reelected, and Dan Rather wasn't.

That, in a nutshell, is the outcome of a bitter four-month struggle between the White House, which insisted there was no basis for the "60 Minutes" report casting doubt on the president's National Guard service, and a major network whose controversial anchor chose to give up his job before the release of the outside panel's report that sharply criticized him yesterday.
. . .
But beyond the magnanimity, Republican leaders were confident the report would add weight to their charge that the media have long been unfair and inaccurate in reporting about Bush.  After four years in which Bush's truthfulness has been questioned by the press on matters including economic projections and the threat posed by Iraq, the White House can point to a high-profile episode in which it was indisputably wronged by a major media outlet.
What makes this article bizarre is that it is coauthored by Dana Milbank, whose bias against Bush is so open as to be almost funny, and who has had to admit errors* more than once in his coverage of the White House.  The article is written as if Milbank is on a hillside watching a battle in a valley, but in fact he has been down there in the muddy trenches for years.

The two Seattle papers used the same AP story, but selected different parts of it to print.   The Seattle Times was straightforward in its selections; the Seattle PI gave almost half its space to the defense given by the discredited Mary Mapes.   It is not hard to see who the PI supports in this controversy — and why readers should be skeptical even of their AP stories.  Those who look at the versions in the two newspapers will probably be struck, as I was, to see that the PI does not give a single line to the White House reaction, not one.  Mary Mapes is not the only close minded journalist in the United States.

(*Since Milbank coauthored the piece, I expected errors and found one immediately.   Kurtz and Milbank write that "Bush has often asserted that he does not read newspapers or watch television", but that isn't quite what Bush said.  Instead, he said he didn't pay much attention to the editorials in the New York Times — which I can assure you is only sensible.)
- 9:40 AM, 11 January 2005   [link]

Let's Say You Want To Spread Some Nasty Gossip, and you want to preserve deniability.  An old, old trick is to ask a question which raises nasty suspicions.   The gossip does not say that some nasty thing is true; they ask for your opinion on the question.  Natalie Solent caught the very respectable BBC in exactly that trick.
The British Broadcasting Corporation, funded by the British taxpayer considers it an open question whether, ten days ago, between one hundred thousand and a quarter of a million people were at best deliberately not saved or at worst murdered by the United States Government.
Simply asking the question will make many people think that the answer is yes, as the gossip knows, and as the BBC knows.

An equally old trick is to say not that "such and such is true", but that "so and so is saying such and such is true".  The nasty gossip who uses this trick knows that the story, especially after it passes through a few people, will lose the "so and so is saying" part.   The gossip gets to spread a smear without ever telling a lie.  The very respectable New York Times used exactly that trick to spread the similar suspicions about the tsunami, using an anonymous post on the far left bulletin board, the Democratic Underground, to start the gossip.  To his credit, the administrator of the site, David Allen, wrote this letter in protest.
As the administrator of the Web site Democratic Underground, I am perplexed as to why you would consider an obscure posting on a busy Internet discussion forum to be worthy of an article ("Myths Run Wild in Blog Tsunami Debate," news article, Jan. 3).

Your article did not make clear that the message in question was posted by a completely anonymous individual, whose identity and political agenda are impossible to determine.  The article also did not mention that the posting appeared to be an innocent question from a person guilty of nothing more than ignorance.  Indeed, the posting's title (which you did not mention) was "One more dumb question regarding the earthquake in Asia."
David Allen thinks that the New York Times used the anonymous post to smear his bulletin board.  Knowing the New York Times better than he does, I am almost certain that they used it to smear the United States — just as the BBC used a similar rumor for the same purpose.

That such respected news organizations have the morals of nasty gossips says something about the profession of journalism in both countries, and it isn't something pleasant.

(Here's the original New York Times article if you want to pay to read it.)
- 3:32 PM, 10 January 2005   [link]

Error Or Non-Vote?  While looking at this David Postman column , I saw a link to a much cited working paper from the Caltech/MIT Voting Project on "errors" in voting with different voting systems.  Postman quotes the first part of this paragraph:
The central finding of this investigation is that manually counted paper ballots have the lowest average incidence of spoiled, uncounted, and unmarked ballots, followed closely by lever machines and optically scanned ballots.  Punchcard methods and systems using direct recording electronic devices (DREs) had significantly higher rates of spoiled, uncounted, and unmarked ballots than any of the other systems.  The difference in reliablities is approximately 1.5 percent of all ballots cast.
That is what the report says; that is not, however, what the data in the report says.

Nearly everyone will read that paragraph as saying that voters make more mistakes with punch cards and electronic voting machines, but the researchers have no data on voter mistakes.   Instead, they measure what they call "residual votes".
Our measure of reliablity is the fraction of total ballots cast for which no presidential preference was counted.  We call this the "residual vote."

A ballot may show no presidential vote for one of three reasons.  Voters may choose more than one candidate — commonly called an over vote or a spoiled ballot.  They may mark their ballot in a way that is uncountable.  Or, they may have no preference.  The latter two possibilities produce under votes or blank ballots.  The residual vote is not a pure measure of voter or of machine failure, as it reflects to some extent no preference.  Consequently we prefer the term residual vote instead of error rate or uncounted vote.
To see why "residual votes" are not errors and are not even a good measure of reliability, let's look at the total votes for some offices in Washington state's last election.   The total number of ballots cast in Washington state, according to the latest recount was 2,883,499.  A little bit of arithmetic on the results from here, here, here, and here gives us this table.*

"Residual Votes" for Washington State Races in 2004

officeTotal VotesResidual Vote
secretary of state2,660,8177.7
insurance commissioner2,559,64211.2

Now if we were to take the idea of residual votes seriously, either as evidence of errors in voting or as measure of reliability, we would have to conclude that Washington's voting technology began to fail terribly as voters moved down the ballot.  Since that's absurd, we should recognize the obvious, for example, that the 11.2 percent of the voters who did not make a choice in the insurance commissioner's race did so deliberately.  But if they did, it seems unreasonable not to conclude that .8 percent of the voters did not make a choice in the presidential race, and so on down the ballot. So what we are seeing is not errors, but, for the most part, non-votes.

The authors of the working paper provide more evidence for that idea in their data on changes in the "residual vote" from one election to another.  In Table 3 of the paper, they show that "residual votes" fell between 1988 and 1992 from 2.5 percent of all votes to 2.0 percent of all votes, and they fell in every category of election technology.  We can conclude either that voters got smarter in those four years, that voting technology improved across the board, or that fewer voters skipped the presidential choice.  The third alternative seems far more likely to me than the first two.

Can we even use residual votes to compare voting systems, as the authors claim?  Probably not, because the different systems are used in places with different characteristics.   Paper ballots, for example, are used almost entirely in rural areas.  That they have a lower rate of residual votes than other methods may only show that people in rural areas are smarter than those in urban and suburban areas.  (Or, if you prefer, there are fewer illiterates in rural areas.)

Although I disagree with the authors, at least they are clear that they are not measuring error rates when they calculate residual votes.  But many have misunderstood them and have interpreted residual votes as error rates.  I am fairly sure that this mistaken claim from Washington state's "governor-elect", Christine Gregoire, comes from exactly that misunderstanding.
Yes, we know the voting machines aren't all they are cracked up to be.  Machine manufacturers admit that the average error rate for machines is 1.4 percent.
She's made this claim a number of times, often enough so that someone must have corrected her, or at least tried to correct her.  Does she just ignore the corrections?  Does she not understand them?  Either possibility is unpleasant, especially in a governor.

(*My calculations of the residual votes may not exactly match those in the study.  The on-line election results do not include write-in votes, and I suspect the count of ballots does not include those disqualified for one reason or another.  At a guess, if I included both, then the residual vote would rise slightly.  But I am sure that the great increase in residual vote as you move down ballot would still be there.)
- 2:33 PM, 10 January 2005   [link]

What CBS Left Out:  The network has finally released their report on the fake National Guard documents story, and I was struck by what was left out.  I have not, obviously, found time to read the report, but nothing in the article mentions any review of other stories done by the four people they just fired.
Asked to resign were Senior Vice President Betsy West, who supervised CBS News primetime programs; 60 Minutes Wednesday Executive Producer Josh Howard; and Howard's deputy, Senior Broadcast Producer Mary Murphy.  The producer of the piece, Mary Mapes, was terminated.
Mary Mapes, at least, has been at CBS for years, and came to the network with a reputation for biased work here in Seattle.  Shouldn't CBS look at her earlier stories?  I am not a big fan of the New York Times, but when they finally recognized that Jayson Blair was not an ideal reporter, they did look at all his previous stories and found that many were false or plagiarized.  CBS should do the same thing with stories done by these four.

(I did not see an apology to President Bush, either, though I seem to recall that Dan Rather did make a half apology earlier.)
- 8:45 AM, 10 January 2005
More:  It is now clear that CBS does not intend to look at other stories by their gang of four.  Given the failures shown by all of them (and Dan Rather) in this story, that is astonishing.  Suppose a store caught four of its employees stealing on a single day.   Wouldn't the store want to check for other thefts by the four?  Suppose they all worked for the same supervisor.  Wouldn't the store want to check the rest of the supervisor's record?

Think I am too harsh?  Mary Mapes still thinks her story is true.
I believe the segment presented to the American people facts they were free to accept or reject, and that as those facts were presented, there was nothing that was false or misleading.
And so does Dan Rather, whose half apology to President Bush was forced on him by management.  If Mapes and Rather are unable, even now, to grasp the facts in this not very complicated story, how can we trust any of their other stories?  And it is not that they were not warned — by other CBS employees before and after the story aired.

Let me be as clear about this as I can.  Their failures on this story, and their blockheaded refusal to look at the evidence after the story was challenged, bring into doubt every other story these two have done.  In this post, I argued that Dan Rather had quit the wrong job, that although he might be an acceptable news reader, he was hopeless as a journalist.  Everything I have read supports the idea that Mary Mapes is, to put it mildly, a misfit in her chosen profession, that she is unable to accept information that does not fit her preconceptions.  There are jobs that people with such problems can do, but reporting and editing are not among them.

This wasn't the first story Dan Rather and Mary Mapes got wrong.  CBS owes us an investigation of their entire careers — if, that is, CBS wants to be taken seriously as a news organization.
- 8:28 AM, 11 January 2005   [link]

Jobs For The Girls:  There was an underlying issue in Washington state's gubernatorial election that has received little attention.  Pro-abortion extremists saw an opportunity to find another job for one of their girls.  That Christine Gregoire had shown, as Washington's attorney general, no executive ability and no real understanding of public policy, did not matter to the pro-abortion extremists.  Nor did it matter that Washington state's governor can do little to change the state's policies on abortion.  Such practical questions do not matter to the extremists.

That extremism explains curiosities such as this fawning Deborah Solomon "interview" of Gregoire.  For those who don't read Solomon regularly, I should explain that she generally does sharp, even confrontational interviews, at least when she is interviewing someone who does not follow the party line.  (I amused to see Solomon describe Dino Rossi, a state senator who took the lead in finding a budget compromise in the last session, only as a "Republican businessman".)

And that same extremism explains this bizarre column from the Seattle PI's Susan Paynter, with this headline: "If you're pro-choice, pray for Gregoire".  To show just how dangerous Rossi would be, Paynter cites events in the nation's capital, North Carolina, and Ohio.  She believes that events such as these — which have no connection to Rossi — prove that he should not be governor
  • On inauguration day groups seeking to overturn Roe v. Wade plan a celebration near the White House focused on their hopes for anti-choice appointments to the U.S. Supreme Court.

  • The same groups ended the year by picketing 20 Planned Parenthood locations, claiming that Planned Parenthood's holiday card greeting, "Choice on Earth," "mocked the birth of Christ" and perversely celebrated killing in a sacred season.
Did Rossi have any connection at all to these exercises of free speech?  Even Paynter doesn't claim that.  But he does have pro-life beliefs and that's enough for her.

The logical outcome of Paynter's standards is, of course, a religious test for office, something forbidden by the Constitution.

I have long thought that we should choose the best candidate for a position, regardless of their sex.  In my youth I sometimes met reactionaries who believed that one sex was better suited for political office than the other.  It is odd to see so many with that same view in such "progressive" institutions as "mainstream" newspapers.

(Pro-abortion extremists often claim that women who are pro-life are not really women, just as black extremists often claim that moderate or conservative blacks are not really black.  It is especially strange in the abortion debate, since, as everyone knows who has studied the subject, the strongest supporters of the pro-life position are married women with children.

I should probably add, so no new readers are confused on the point, that like most voters I am a moderate on abortion, wanting more restrictions than we now have, but not an outright ban.)
- 8:13 AM, 10 January 2005   [link]

Think The New York Times May Be A Little Out Of Touch With The Average Person?   Then you'll find confirmation in the lead article from today's travel section, Why Is Everyone Going to Bhutan?
Two years ago, Penny George "couldn't have located Bhutan on a map."  But after hearing friends rave about their trip to the tiny Buddhist kingdom tucked in the Himalayas, Ms. George, president of a foundation that promotes holistic medicine, was hooked.  This fall, she and her husband made the long journey from their home in Minneapolis to Bhutan's sole airport, then spent seven days on a guided tour, trekking into virgin forests, tiptoeing into temples and passing through villages where men and women still go about in traditional dress.  "Bhutan has bubbled up in the collective consciousness," said Ms. George. "I just felt like I had to go."

Move over, Cambodia. Bhutan is the new must-see destination in southern Asia.
Read a little farther down and you will learn that "everyone" means 9,000 visitors to Bhutan last year (3,000 of them Americans), up from few thousand just a few years ago.  Those numbers explain why the checkers at the grocery stores haven't been telling me about their trips to Bhutan, even though "everyone" is going there.

(Could I have located Bhutan on a map before I read this article?  Yes, I knew Bhutan was located east of Nepal, between India and China.   But I couldn't have sketched it correctly, since I thought that Nepal and Bhutan bordered each other.   That's close, but not quite right, as you can see from the map accompanying this article from the CIA Factbook.)
- 10:04 AM, 9 January 2005   [link]

Rock, Scissors, Paper?  The Seattle Seahawks were eliminated yesterday by the St. Louis Rams.   This was the third Seahawks loss to the Rams this year, even though the Rams had a poorer record than the Seahawks.

That got me to wondering, as I have before, whether team strengths are always transitive.  Ordinarily, if team A defeats B and team B defeats C, we would expect team A to defeat C.  I am sure that's what usually happens, but I wonder if there aren't exceptions, that sometimes, just as rock beats scissors beats paper beats rock, we should expect team A to beat B, team B to beat C, and team C to beat A.  I can't think of a clear example, but it does seem to me that this would be most likely to happen in football.  Do any of you know of an example, or, even better, a statistical study that shows that this happens from time to time?
- 6:41 AM, 9 January 2005   [link]

There Is Something Wrong with a system that produces these results.
A West Seattle elementary school principal who has already cost Seattle Public Schools $90,000 to settle a sexual harassment suit will receive $132,000 in exchange for his resignation.

Under an agreement signed this week, Dan Barton, 49, will also get glowing letters of recommendation from top district officials.

Barton's exit deal brings the total amount of settlements given to district principals over the past five years to more than a half-million dollars.

Barton was given a written reprimand in 1999 for sexually harassing female teachers at his school, Gatewood Elementary.  Last October, he was put on paid administrative leave while district officials investigated new sexual harassment allegations.

The investigation concluded there was insufficient evidence to support the claims against Barton, according to district deputy general counsel Faye Chess-Prentice.
If Barton is guilty, then the school district is rewarding a guilty man — with more money than his victim(s) received.  If he is innocent, then Seattle is losing a fine principal — assuming the reports of progress at the school are accurate.

Seattle principals, as well as teachers, have the protection of tenure.  I am skeptical about tenure for teachers, especially college professors, and I am completely opposed to it for administrators.  High schools almost never give their football and basketball coaches tenure — because they value performance in sports.  They should do the same for academics.
- 5:39 AM, 9 January 2005   [link]

The Indian Ocean Tsunami Hit Florida:  Though not very hard.
The tsunami that ravaged countries all around the Indian Ocean also hit the eastern United States, though only the tide gauges noticed.

A tide gauge at Atlantic City recorded the passage of a "train" of waves, just under nine inches from crest to trough, 32 hours after the earthquake struck off Sumatra's west coast on Dec. 26, said Dr. Alexander B. Rabinovich of the Canadian Institute of Ocean Sciences in Sidney, British Columbia.   A gauge at Port Canaveral, Fla., recorded 13.4-inch waves 24 minutes later.
This is not just a curiosity.  Those who study tsunamis welcome the data from the Atlantic, where tsunamis are rarer, but still a threat.  (I have seen a claim that a giant landslide in the Canary Islands could set off a tsunami that would dwarf those that crossed the Indian Ocean on the 26th, and would destroy much of the American East coast.  I have no idea whether it is a real danger.)
- 5:00 AM, 9 January 2005   [link]