April 2007, Part 3

Jim Miller on Politics

Pseudo-Random Thoughts

How Serious Is Vote Fraud In The United States?  No one knows the answer to that question.  Nor do I know of any good estimates.  From the nature of the crime, we can be certain that most illegal votes go undetected, and hence unmeasured.

When illegal votes are detected, they are rarely prosecuted.  The New York Times made much of a recent report that found few convictions for vote fraud.
Five years after the Bush administration began a crackdown on voter fraud, the Justice Department has turned up virtually no evidence of any organized effort to skew federal elections, according to court records and interviews.

Although Republican activists have repeatedly said that fraud is so widespread that it has corrupted the political process and, possibly, cost the party election victories, about 120 people have been charged and 86 convicted as of last year.
(Note, by the way the "organized effort" phrase in the first paragraph; that's a key qualifier that I will come back to.)

In an editorial, the Times even called the idea that the problem was serious, a "fantasy".

But all that report may show is that the risks from illegal voting are small, something I have been arguing here for years.  (And perhaps that professionals at the Justice Department, most of them Democrats, are unwilling to treat this problem seriously.)

Others, notably John Fund of the Wall Street Journal, come to different conclusions.  Here, for example, is what he said in Stealing Elections.
Election fraud, whether it's phony voter registration, illegal absentee ballots, shady recounts or old-fashioned ballot-box stuffing, can be found in every part of the United States, although it is probably spreading because of the ever-so-tight divisions that have polarized the country and created so many close elections lately.  Although most fraud is found in urban areas, there are current scandals in rural South Dakota and Texas.  In recent years, Baltimore, Philadelphia, New Orleans and Milwaukee have all had election-related scandals.  Wisconsin officials convicted a New York heiress working for Al Gore of giving homeless people cigarettes if they rode in a van to the polls and voted.  The Miami Herald won a Pulitzer prize in 1999 for uncovering how "vote brokers" employed by candidate Xavier Suarez stole a mayoral election in Miami by tampering with 4,740 ballots.  Many were cast by homeless people who didn't live in the city and were paid $10 a piece and shuttled to the elections office in vans.  All of the absentee ballots were thrown out by a court four months later and Mr. Suarez's opponent Joe Carollo was installed as mayor. (p. 5)
So, which view is correct?  Are illegal votes minor problem, or a serious, and possibly growing problem?  Let me describe a few cases, and then let you come to your own conclusion.

Let's begin with one of the most famous, the 1960 presidential election.  Kennedy would probably not have won that election without fraudulent votes in Illinois and Texas.  (If you haven't seen the evidence, take a look at this Earl Mazo obituary for a sample.)

Or consider the 1998 gubernatorial election in Hawaii, where Republican Linda Lingle challenged the Democratic machine that had run the state for so long.  She lost narrowly, by just 5,000 votes, and the winner's margin may well have come from fraudlent votes.  (She won the next time around in 2002.)

Or consider the 2000 presidential election.  Everyone knows how close the election was in Florida, but few know just how many illegal votes were cast in that election, illegal votes mostly benefitting Al Gore.  The Miami Herald and the Palm Beach Post each found more than 5,000 likely votes by felons, a group that is heavily Democratic.  The New York Daily News found that tens of thousands of voters had voted in both New York and Florida, illegally.  This group, too, was heavily Democratic in registration.  If Al Gore had gained a victory in the Florida recounts, his margin would have come from illegal votes.  Unquestionably.

Or, consider the 2004 Washington gubernatorial election.  The two parties found, without much work, that more than a 1,000 felons had voted illegally.  National studies show that felons tend to vote Democratic, so these illegal felon votes, by themselves, almost certainly provided Governor Gregoire's slim plurality.  (And there were many other dubious votes counted in that election, most of them going to Gregoire.)

Finally, consider a general problem, voting by non-citizens.  Election laws in many states encourage non-citizens to vote.  How many do is not known, but the number of non-citizens in the country is so large that, even if just one percent of them vote, they have undoubtedly tipped some elections — to Democratic candidates, in almost every case.

Is having one presidential election decided by illegal votes, and another almost decided by them a serious matter?  Is having gubernatorial elections tipped by illegal votes a serious problem?  (And I haven't mentioned several recent senatorial elections that have probably been tipped.  Not if you are a certain kind of partisan Democrat, because the edge that comes from illegal votes almost always goes, in general elections, to the Democratic candidate.

Is that why the Times chooses to minimize this problem?  I think that's part of the reason that they do.

(Did you see just how sneaky that "organized effort" qualifier is?  Note that the Times is not saying that there are few illegal votes or that they have not changed election results, just that there is no vast left wing conspiracy.  And I am not sure they are right even on that point, given the problems that ACORN has in following election laws, in so many places.)
- 4:31 PM, 24 April 2007   [link]

Provocative:  Niall Ferguson makes an argument that will be best understood by those who with a little knowledge of statistics.
It was predictable.  Cho Seung-Hui was a taciturn, moody loner.  Four of his professors expressed concerns about the content of his work or classroom conduct.  After complaints by two female students, the campus police and a college counsellor tried to have him committed to a mental institution.  But a doctor didn't agree with the judge that he presented a danger to others.   And guns are easy to buy in America (though banned on Virginia campuses).  As a result 33 people are dead.

Journalists' efforts to explain the Virginia Tech massacre perfectly illustrate one of the central points of an idiosyncratically brilliant new book by Nassim Nicholas Taleb, The Black Swan: The Impact of the Highly Improbable (Penguin/Allen Lane).  Having been completely caught out by some random event, we human beings are wonderfully good at retrospectively predicting it.  In reality, however, Cho was what Taleb calls a "Black Swan".

Why a black swan?  Taleb's starting point is what philosophers call the problem of induction.   Suppose you have spent all your life in the northern hemisphere and have only ever seen white swans.   You might very well conclude (inductively) that all swans are white.  But take a trip to Australia, where swans are black, and your theory will collapse.  A "Black Swan" is therefore anything that seems to us, on the basis of our limited experience, to be impossible.
In other words, we put too much weight on our own, limited experience when we judge what is impossible, or even highly unlikely.  And we often use, whether we realize it or not, a normal distribution in estimating what is likely, and what is not.  But the normal distribution is not a good description of some kinds of events, earthquakes, for example.

(Here's a link to The Black Swan.   I'll have to pick it up some time, I suppose, and add it to my must-read stack.  By the way, the author is the manager of a hedge fund, so serious investors may want to look at what he has to say about risks.)
- 10:52 AM, 24 April 2007   [link]

But Not In The Southwest:  In the first round of the French presidential elections, the conservative candidate Nicolas Sarkozy led Socialist Ségolène Royal by more than five percent of the popular vote.
With more than 99 percent of the vote counted, Mr. Sarkozy was leading with about 30.7 percent of the vote, and Ms. Royal had about 25.2 percent.  Françla;ois Bayrou, the centrist who vowed to plot a new, conciliatory way of governing, came in a distant third with about 18.4 percent.
But Sarkozy did not lead Royal everywhere.  As you can see from this map, Royal led Sarkozy in Gascony and Brittany, and in the southwest of France generally.  Why did she do better in the southwest of France?  I haven't seen an explanation, but I would guess that she did better in poorer regions.  (Gascony has been famous for centuries for the pride of its inhabitants — and infamous for its poor soil.)

Royal's promises to give everyone lots of everything would probably have more appeal in poorer regions, if French politics are anything like ours.

(If you have done the arithmetic, you will have noticed that other candidates drew about 25 percent of the vote.  There were nine other candidates; none of them were serious choices, so one of every four French voters chose to express their feelings in this first round, rather than to choose a leader.  That is, by the way, much lower than in the first round of the last French presidential election.)
- 8:15 AM, 23 April 2007   [link]

You Can't Make This Stuff Up:  Unless you have a better imagination than I do.  Here's the latest advice from eco-activist and singer Sheryl Crow.
Singer Sheryl Crow has said a ban on using too much toilet paper should be introduced to help the environment.

Crow has suggested using "only one square per restroom visit, except, of course, on those pesky occasions where two to three could be required".
Should Crow just Shut Up & Sing?   Perhaps.  If she doesn't want to be greeted by gales of laughter wherever she goes.

(It is only fair to add that a cause is not necessarily wrong because one of its supporters says foolish things.  But we do have one more reason not to take the advice we get from pop stars seriously.)
- 7:08 AM, 23 April 2007
More:  Sheryl Crow now says she was joking, but then she would say that, wouldn't she?   Whether or not she actually was.  For the record, I did consider that possibility when I wrote the post and looked carefully at what she wrote to see any hint of a joke.  I didn't spot any, but perhaps she and I have very different senses of humor.

Meanwhile, Marty Mazur had his own reaction to the story, a reaction which is intended as a joke — and succeeds.
- 3:45 PM, 26 April 2007   [link]

The Tyler Kent Case:  Yesterday, I went to the first day of the big Seattle book sale.  (As soon as I finish this post, I will be going to the second day of the sale.)  As usual, I found some bargains, and I found one book that I would have been happy to buy at full price, Joseph Persico's Roosevelt's Secret War.

I'll have more to say about what I have learned from this book in later posts, but for today, I just want to mention one minor incident, the case of Tyler Kent, a code clerk in the American embassy in London in 1940.  As a code clerk, Kent was able to read the messages that passed between Roosevelt and Churchill — and he didn't like what he read.  Finally, his distaste drove him to disloyal actions:
Tyler Kent, as he brooded in the airless silence of the code room translating messages into the state department's Gray code, fretted that FDR was "secretly and unconstitutionally plotting with Churchill to sneak the United States into war."  He had developed a corollary obsession:  "All wars are inspired, fomented, and promoted by the great international bankers and banking combines which are largely controlled by the Jews."  He had, he later admitted, "anti-Semitic tendencies for many years."  Kent finally decided where his duty lay.  He had to gather evidence that he could place into the hands of the U.S. Senate and the American press to expose Roosevelt's duplicity and keep the United States out of war.  Roosevelt, Kent believed, had to be stopped, especially since, it was rumored, he might run for an unprecedented third term.  And so Kent began to steal and copy documents from the code room which he hid in his flat in a brown leather bag, a crate, and in the cupboard. (Persico, p. 23)
And Kent's copies eventually fell into the hands of our Axis enemies.

Kent's thefts were detected because material from the copies was sent in a German diplomatic code that the British had broken.  Kent was arrested, tried, and sentenced to a British jail.  He stayed there until the end of the war.  He was then deported to the United States where he lived until his death in 1988.

There are two aspects to this case that strike me as relevant to us now.  First, Kent committed a tactical error when he gave the copies to foreigners who shared his political views, rather than to American newspapers or politicians.  There were newspapers then — notably the Chicago Tribune — that hated President Roosevelt as much as the New York Times now hates President Bush.  If Kent had gone to them, or to the politicians who opposed Roosevelt's policy of resisting Hitler, he might not have been caught, and he might have changed history.

Second, although Kent held some absurd ideas about the causes of war and the guilt of the Jews, he was partly right.  Roosevelt was being duplicitous, was breaking our neutrality, was conspiring to help Britain as much as public opinion in America would allow.  (At that time, less than 3 percent of Americans wanted us to enter the war on Britain's side.)  And Roosevelt's administration was breaking American laws to provide that help, something I'll discuss at length in a later post.  Roosevelt's appointees were breaking American laws because they felt they felt they had to — and because Roosevelt had ordered them to break those laws.

Now what Roosevelt and his administration were doing was unquestionably duplicitous and illegal in part.  But was it wrong?  Given the threats from Hitler and imperial Japan, was Roosevelt right to ignore American laws?  Was he right to be duplicitous?  In my opinion, the terrible lessons of World War II shows that Roosevelt was right both in breaking the laws and in concealing much of what he was doing from the American public.
- 10:43 AM, 22 April 2007   [link]

Court Procedures:  Yesterday, they did hold a trial, and I was chosen for the jury — after several others were challenged — including, curiously, the same woman who was challenged on Monday.  And, again, she was challenged by the defense attorney.

On Monday, the presiding judge (I think that's the correct title), Michael J. Lambo, had a clerk draw all of our jury numbers.  The first six drawn went on the jury.  When a juror was challenged, the juror with the next number drawn replaced them.  Yesterday the judge picked the first six in the same way, but then chose replacements himself, picking those in front first — which is how I happened to get chosen.  (I am not sure what his name is.  the sign said "Pro Tem", that is "substitute".  When I asked him what his name was, I thought he said "Scott Greer", but I could not find a likely "Scott Greer" searching on either lawyers or judges.)

The first way of choosing replacement jurors is fairer, assuming you believe that randomness is the measure of fairness.

(Strangely, neither judge was introduced to us and I was unable to find either name at the Kirkland Municipal Court web site.)

On both Monday and Thursday, far more jurors were called to the courtroom than were needed for the trial.  (And others, I suspect, ignored the summons.)  The same thing happened last year in the King County court, and probably for the same reason.  It can be terribly inconvenient — for those running the courts — if the court runs out of jurors, but it is not at all inconvenient — for those running the courts — if there are many extra jurors.  And, given the pay, it doesn't cost the courts much to bring in extra jurors.  (As it turns out, I will get paid a little for my service this week; see the correction at the end of this post for the details.)

In this area, the jurors see a videotape explaining the rules that they have to follow.  I have now seen it three times and each time been struck by the unqualified statement in the tape that jurors could take notes.  I hadn't recalled seeing jurors taking notes in accounts of most other trials, but I was pleased to see it, because I have found that taking notes helps me focus on what I am hearing and seeing.  As it turns out, the videotape was wrong, and the judge ordered my notes confiscated after we delivered the verdict.  (If I understood the judge right, jurors can take notes — if the two sides agree to that before the trial.)

(As you probably noticed, I have not said anything about the case yesterday — and don't intend to.  It was a domestic violence case and I learned more about one troubled marriage than I ever wanted to.  And the trial, though required by law, as I understand it, was almost certainly the wrong remedy for the family's troubles.

After listening to the particulars of the case, I began to feel a bit like the man Mark Twain told about.  The man was tarred, feathered, and ridden out of town on a rail.  The man said, that, if it weren't for the honor, he would rather have walked.  That's how I felt, except that I would have been glad not to have the "honor" of being a juror for this particular trial.)
- 9:29 AM, 20 April 2007   [link]

Probabilistic Planning:  The uncertainty about jury duty lets me indulge in one of my favorite hobbies, probabilistic planning.  What I would like to know, for the next two days, is, first, whether I will be called in and, second, whether I will serve on the jury, if I am called in.  I have a little bit of data, enough so that I can make a guess for each question.

I have been told to be ready to report to this court three times (including once last year); only once did I actually have to show up.  So, I estimate that I have 1 chance in 3 of having to report each day.  So, the chance that I will not have to report either today or tomorrow is (2/3)2, or about 44 percent.

And, if I do report, what are my chances of serving?  There were 20 jurors in the pool Monday; 6 were chosen.  Assuming those numbers are typical, I have a 30 percent chance of serving if I am called in.

Can I go a step further and calculate my chances of serving this week?  Yes, and a better probabilist could give you the answer now.  But I have found that, for me, when it gets this complicated, that it is best to draw a "tree" of possible events, as a check on my calculations.  Maybe I will do that later today.

This is a toy problem, for your amusement.  But similar, though far more complex calculations are made in business all the time — and in war.  And I have found that probabilistic planning does, sometimes, help me plan my own life.
- 7:47 AM, 19 April 2007
Update:  Just checked the court's web page and learned that I will have to report today.  So, off I go.
- 7:58 AM, 19 April 2007   [link]

Seanet Outages:  If you were unable to get to my site early yesterday morning or early this morning, you are in good company.  I couldn't either.  All the more reason to speed up my switch to the new site, which I still hope to do this month, in spite of the jury duty interruption.

(On the whole, I have been satisfied with my internet provider, Seanet — or I would have switched long ago.  But Seanet no longer meets my needs, which have grown over the years.)
- 7:13 AM, 19 April 2007   [link]

Worth Reading:  As part of a three part series on critical thinking, Bill Whittle dispatches a bunch of conspiracy theories.  For example:
Recently, Rosie O'Donnell said on national television that she believes 9/11 was orchestrated by the US government.
. . .
"This is the first time in history that fire has ever melted steel," she said.  This is a statement of such pristine and perfect idiocy that it surely must be emblazoned in stone across the entrance to the Physics Imbecile wing of the Moron Museum of Natural History.
And would astonish every single person who has made steel, beginning with the first blacksmith.   (As far as I know, she has never corrected even that absurd claim.)

Whittle's essay is much longer than the average blog post, but it is worth your time — and if you have a friend who believes in one of these conspiracy theories, pass it on to them.
- 3:58 PM, 18 April 2007   [link]

Jack Shafer Recycles:  Slate's press critic thinks this 2004 column on the Pulitzer prizes is still good.   He began with this question:
Outside of a couple thousand journalists working at the top-tier newspapers that stand a chance of winning one, does anybody really care about the Pulitzer Prizes for journalism?
Which I can answer by recycling one of my posts.
For several years I have been saying that the Pulitzer prizes were best understood as reprimands.   The prizes, more often than not, were awarded for work that deserved criticism, not praise.  (And the same is true of the Nobel Peace Prize, which in recent years has often gone to those who have damaged the cause of peace.)  I was partly joking, since there are fields in which the prizes seem to go to people who merit prizes, rather than reprimands.  But only partly, because in some areas, the judges consistently select winners who deserve, not prizes, but reprimands.
(That's the Pulitzer prizes in journalism; I have no opinion about the prizes in other fields, where they may identify meritorious, rather than shameful, works.) So, yes, some of do care about the Pulitzers, because they help us identify particularly bad work, and particularly dishonest journalists.  (And should come with fines, rather than prizes.)

Are there some reprimands among this year's prizes?  Yes.  There were fourteen prizes awarded.  One, that going to Andrea Elliot of the New York Times, is certainly a reprimand.  From their descriptions, I would say that at least six others probably are reprimands, the prizes given to Walt Handelsman of Newsday, Oded Balilty of the Associated Press, Renée Byer of the Sacramento Bee, Charlie Savage of the Boston Globe, Kenneth Weiss, Usha Lee McFarling, and Rick Loomis of the Los Angeles Times, Cynthia Tucker of the Atlanta Journal-Constitution, the editorial writers at the New York Daily News, and Jonathan Gold of the LA Weekly.  There were several others that may be reprimands, but I couldn't be sure without more evidence.

Even that quick look shows that Pulitzers are useful.  If a journalist wins one, then we have good reason to be suspicious of his work.  And good reason to think that he has received a reprimand.

(Why an honest journalist would want to have the same prize won by, among others, Walter Duranty, is puzzling.  Perhaps one day a journalist will have enough integrity to refuse the prize until the Pulitzer committee revokes the one it gave to Duranty.  Perhaps.)
- 3:29 PM, 18 April 2007
More:  Jonathan Tobin shows why Andrea Elliot deserves a reprimand, not a prize.
- 10:14 AM, 24 April 2007   [link]

They Keep Us Hanging On:  On Monday, the court reporter drew six names for the jury.  After the voir dire, the questioning of the jurors by the prosecution and the defense attorney, one juror was dismissed.  The fourteen extra jurors (including me) were sent home at about 11:30 that morning.

Before we were sent home, we were told that there was no trial scheduled for Tuesday, but that we might be needed on Wednesday, Thursday, and Friday.  We were told we should check each evening after 5:00 PM to see whether we would be needed the next day, at the court web site, or at the "juror information line".  So I did, and here's what I found this evening:

NOTE: All jurors are required to call the juror information line at 425-587-3181 after 8:00 am on Wednesday April 18th for further instruction.

So maybe we'll be needed tomorrow, and maybe we won't.  As I said in this post, this uncertainty doesn't matter much to me, but it probably does to many other jurors.

Cross posted at Sound Politics.
- 7:13 PM, 17 April 2007
Update:  As of 8:35 AM, there were no juror directions on the line, just a message saying that we should call the line after 8:00 AM, so I got ready to drive to the court.  (It takes about 15 minutes to get there, and I had originally been told to report at 9:00 AM.)  Just before I walked out, I checked one last time — and learned that I did not have to report today, after all.  But I do have to go through the same thing tomorrow morning.
- 8:50 AM, 18 April 2007   [link]

On A More Pleasant Note, I have found a picture for the Canadian bloggers section.  That waterfront picture links you to a webcam in Halifax, Novia Scotia.  (And to more live pictures of the city.) As always, suggestions for live pictures to illustrate other sections are welcome.

Fun fact:  The residents of Halifax are called Haligonians, or so this disputed Wikipedia article says.  (I'm haven't checked to see what is being challenged, but I don't think it is that name.)
- 10:38 AM, 17 April 2007   [link]

Reasoning From Particular Cases:  The horrific murders at Virginia Tech yesterday have inspired the usual automatic reactions.  Those who thought that we have too loose gun control laws were certain that these murders demonstrated the truth of their case.  Those who thought that "gun-free" zones, such as Virginia Tech, make it easy for murderers thought the murders demonstrated the truth of their case.  And so on, from many others.

Most of those who draw general conclusions from this case are wrong; all of them may be wrong.  The unhappy fact is that we usually can not draw general lessons from individual cases.  To make this point clearer, let me turn to another famous case, the three Duke Lacrosse players accused of rape.  From the very beginning, many tried to draw general conclusions from the case, that whites oppress blacks, that district attorneys attack rich white kids when they can gain poltically, and so on.  I did not say much about the case because I did not know much about the case, and because I did not think that — whatever the facts were — the single case would allow us to generalize.

Suppose, for instance, that the lacrosse players had been guilty as charged.  Would this tell us much about rape in general?  No, because it would not change the fact that whites are far more likely to be raped by blacks than the other way around.  Nor does the fact that the accuser (who may be mentally imbalanced) was not telling the truth tell us much about accusations of rape.  Most, I am sure, are well-founded.  (Though I have seen claims that false accusations are more common with rape than with most other crimes.)

It's an unhappy fact, because individual stories can so compelling.  We want to draw general lessons from them because they have so much emotional power.  But only rarely can we do so, if we want to be honest in our reasoning.
- 10:10 PM, 17 April 2007   [link]