Site Answers:



Jim Miller on Politics





Email Policies:  If you send me email about a post, I may quote it in part, or full, on the site.

I have different rules about identifying "public" and "private" emailers. For me, "public" emailers include all public officials and journalists, and all other people famous enough to be recognized by many readers.  That includes some of the more prominent bloggers.  Everyone else falls into the "private" category.

If you are a "public" emailer, I will generally include your name with the quotation, unless you specifically ask me not to.  So if Howard Dean wants to confess to me what he really meant by that scream, he should tell me if he does not want me to share his name.

If you are a "private" emailer, I will not include your name with the quotation, without your permission.
- 4:06 PM, 8 February 2004   [link]


Frequently Asked Questions:  
  • Why are the thoughts "pseudo-random"?  Just a programmer's joke.  Computer games and simulations use "random" variables all the time.  In almost all cases, they are not truly random because they are generated by an algorithm that produces the same sequence of numbers for a given starting number, or "seed".  So, when programmers are being precise, they call them pseudo-random numbers to remind themselves of the difference.

    So, when I say my thoughts are pseudo-random, I am saying that there is a pattern behind them, even though it may not be apparent.

  • What is your corrections policy?  It depends on when I notice the error and what kind of error it is.  If I notice an error in a post immediately, I will usually just correct it with no comment.  I do the same for typos or for re-phrasing that does not change the meaning.   If I see a comma misplaced, or a word misspelled, or a slightly better way to say something, I will make that change without a note.

    If I make an error in a post and do not notice it immediately, then I add a note at the bottom to show that the post has changed.  Typically, I mark these with either "Oops!" for minor mistakes, or "Correction" for major mistakes.  If I correct a post that is more than two days old, I usually add a pointer to it in a new post.

    My goal is to correct all errors promptly.  If you see any that I missed, you do me a favor by telling me about them.
- Last revised 5:16 AM, 27 April 2004   [link]


Disclaimer On Turnout Comparisons:  In comparing turnout for the presidential nominating contests, there are many traps for the unwary.  Let me start by making a general argument about how much turnout we should expect, and then go through some of the factors that will modify that.

If everything else was constant, turnout, measured in raw numbers, should rise almost everywhere with population growth.  (There are a few exceptions.  North Dakota, West Virginia, and the District of Columbia all lost population between 1980 and 2000.)  There are some complications.  An aging population will have even more voters than population growth alone would predict.  A population with a large proportion of immigrants, legal or illegal, will have fewer voters than population alone would predict.  But, even allowing for these, each set of nominating contests should produce records in raw numbers in almost all the contests—if, I repeat, everything else was constant.

If everything else was constant, turnout, measured as a percentage of voters, should stay roughly constant.  Turnout has been roughly constant in our general elections for the last two or three decades, which is about how long, coincidentally, that we have had our modern system of nominating presidential candidates.  One of the complications mentioned above applies to turnout percentages, too.  Since older votes are more likely to go to the polls, turnout should rise slightly in states with aging populations.  The presence of large numbers of immigrants should not matter, assuming they are not registered to vote — not always a safe assumption— but there is another complication.  Not all states are good about keeping their election rolls up to date.  When you see a turnout of 10 percent, or 20 percent, or 45 percent of the eligible voters, you should be aware that the denominator, total eligible voters, may be off by a large amount.

Now for some of the things that are not constant.  First, and this has almost certainly already occurred to you, nominating contests are very different.  Some are contested (the Democrats this year) and some are not (the Republicans this year).  Some are long and others are short.  Some of the contests draw many serious contenders and others draw just two.  In each of these pairs, we would expect the first kind to draw more voters.

Second, the states and the parties change the rules constantly.  In the 2000 contest in Washington state, both parties used both caucuses and primaries.  This year, both are using just caucuses.  Although I have compared the Democratic results in the two years, it was with a very large caveat.  And the states change them in other ways, as well.   They change the days the primaries or caucuses are held, which changes turnout.  They change the primaries from open, allowing independents to vote, to closed, excluding them.   They even use mixed systems in which independents are allowed to vote, but their votes do not determine the allocation of delegates.  (I think that was the system in California in 2000.)  All of these changes can affect turnout.

Third, and this may be the largest effect of all, the timing of the primaries changes the turnout.  Voters are more likely to vote in primaries that can change the results; recently that has meant the earlier primaries, but if you go back far enough you can find late primaries that determined the results.  At one time, believe it or not, states sometimes made their primaries later, thinking that would allow them to decide the nominee.  The enormous emphasis on Iowa and New Hampshire has probably increased turnout in those two states, and hurt it in states that voted later.

Fourth, in states that have open primaries, the two parties interact because both are trying to attract independents.  If one party has a more interesting fight than the other, then the independents will flock to that primary.  This year, with no contest on the Republican side in Wisconsin, we would have expected a larger turnout in the Democratic primary, from that alone.  (It was actually almost 20 percent lower than in 1988, when there were open contests in both parties.)  Some observers think that Bradley might have beaten Gore in New Hampshire in 2000 had not McCain drawn almost all the independents into the Republican primary.  This raised the turnout for one party and lowered it for the other.

Fifth, in some contests I compare Democratic and Republican primaries.  In these cases, one should remember to correct for the sizes of the parties in the states.  In Virginia, for example, one would expect a higher turnout on the Republican side simply because Virginia has more Republicans.  I try to mention that when it is important.

Finally, we should remember that, with a large number of contests, some should be high, even records, just from the chance.  The New Hampshire Democratic contest this year set a record for turnout this year, but with almost 20 contests, that's not surprising.  (The statistically inclined may be able to find a test for the odds on that, although I think simpler tests won't work because you can't assume independence.)
- 1:22 PM, 20 February 2004   [link]


Disclaimer On Global Warming:  My own opinions about theories of global warming are mixed.  It seems reasonably well established that the earth's climate warmed a degree or so in the last century.  It also seems well established that most of the warming came during the first part of the century when effects from manmade greenhouse gases would have been smaller.  I think it almost certain that, at some level, greenhouse gases would warm earth.  After those three points, everything seems less certain, at least to me.

Let me begin with the climate models themselves.  They are giant simulations, bearing about the same relationship to the climate as a video game about World War II has to the actual war.  In both, the programmers try to capture the essence, without going beyond what current computers, and our present knowledge of the climate (or the war), allow.  A few years ago, many of the models were so bad that they could not "postdict" past climates correctly.  If you plugged in the numbers for the 19th century, you got nothing like the climate then, but something much colder.  That's as discouraging as a World War II computer game in which Hitler always wins.

There is a real possibility that, however carefully these models are constructed, they can not in principle predict future climates, because climate is inherently unpredictable.  Mathematical theory shows that there are situations that are inherently unpredictable and there is some reason to believe that future climates may be among them.  You probably have heard one of the favorite examples, that a butterfly flapping its wings in the Amazon could trigger a storm in the United States.  No simulation can solve such problems, at least not very far into the future.

I have even more doubts about some of the people pushing the theories.  Their actions are often inconsistent with what they say they believe.  For example, if global warming is to be feared, then we should begin switching power from fossil to nuclear fuels.  Almost none of those who worry about global warming back nuclear power, although it is one of the best near term fixes.  The actions they do favor often turn out to be actions they would favor for other reasons, as well.  For example, most of those who worry about global warming would want us to switch to mass transit, whether or not global warming is a problem.  (Mass transit does less for the environment than many think, especially in most of the United States, where the population is so spread out.)

Nor is it entirely clear that, should it occur, global warming would be a bad thing, net, despite all the horrific predictions you may have seen.  Every serious student of the subject agrees that it may have some good effects, at least in some places.  During part of the Middle Ages, the climate, at least of Europe, was a little warmer than it is now.  From what we know of the period, the effects were mostly good, with longer growing seasons and easier times for the population.

All that said, I agree with those who say we should be cautious about the effects we may be having on the earth.  We should look for ways to both reduce emissions of greenhouse gases and to take them out of the atmosphere.  The latter is more practical than you may realize.  Right now, the continent of North America takes in more carbon dioxide than it produces.  It is, net, a sink, not a source.
The fact is that while the North American continent emits about 1.6 billion metric tons of carbon every year, North American carbon sinks actually absorb 1.7 billion tons of atmospheric carbon every year. North America is therefore a net consumer of carbon dioxide.
(If you are wondering how that could be, just think about our farms and forests with all those plants gobbling up carbon dioxide.  Iain Murray does not give his source for the numbers, but I am fairly sure it is a paper published in Science magazine, about 1998.)

We also should be looking for ways to reverse climate change, should it occur.  There are a number of measures we could use to both cool the earth or warm it should that be necessary.   Some would require more capacity to put materials in orbit than we now have, another reason for developing our space faring capability.  (We should be doing this even if there were no danger of man-caused global warming.  Investigators have found evidence of sharp shifts in climate, long before man had the ability to affect the climate or even before man appeared on earth.   This article has some examples, if you are not familiar with these past changes.)

Above all, I think we should do more to improve our scientific understanding of the problem, including the costs and benefits of any changes.  The nasty debates on the subject indicate to me, among other things, that the science is less settled than the advocates admit.
- 2:19 PM, 3 April 2004   [link]


Disclaimer On Distributed Vote Fraud:  Whenever I write a post on distributed vote fraud, I say, quite frankly, that I do not have good estimates of its extent.

Critics sometimes attack me for that admission, unfairly, I think.  The attack is unfair because I would love to do better estimates, but the cost is far beyond my means.  Even a preliminary investigation of the extent of distributed vote fraud, for just one state, would, at a guess, cost at least a million dollars.  The investigation would have to start by drawing a large sample of voters, not the 1,000 or so in the usual poll, but at least 10,000.  You need that many because you are trying to find the 1 vote in 100 or 1,000 that is illegitimate.  You would then want to investigate all 10,000 to see how many were not citizens.  I have no idea how much that would cost per voter, but it would not be cheap.   I do know that employers do not always do much investigation of prospective employees, just because it is too expensive.

And citizenship isn't the only thing that a full investigation would cover.  You would also want to look for forged signatures on absentee ballots.  Probably the best way to do that would be to hire signature experts and have them examine hundreds of signatures looking for miss-matches that passed the clerks.  Again I don't know what such experts charge, but I doubt that they work for the minimum wage.

Those two parts of the investigation would not cover all the possibilities, by any means, but I think they are enough to give you an idea of how expensive an investigation of distributed vote fraud would be.  And, if one of my critics happens to have a few million dollars to spare, I would be glad to talk to them about doing a preliminary study.

One thing that convinces me that the problem of distributed vote fraud is serious, and that it benefits Democrats, is the attitudes of party officials.  President George H. W. Bush vetoed the "Motor Voter" act, and was backed by most congressional Republicans.  When it passed again in 1993 and was signed by President Clinton, most of the backers were Democrats.  In state after state, Democrats have opposed such simple controls on vote fraud as requiring photo identification to register and vote.  Perhaps all these Democrats are wrong to think that there is an advantage for their party in what I call distributed vote fraud, but I doubt it.
- 7:54 AM, 9 December 2004   [link]


Voting Rights For Felons:  Two states, Maine and Vermont, put no restrictions at all on voting by felons.  If you are in prison in Maine or Vermont, you can request an absentee ballot* and vote from your cell.    Some states, Florida being the best known example, ban felons from voting for life unless they receive what amounts to a pardon from the governor.  Most states are in between; here's a graphic from the New York Times that will give you an idea of the limits in various states.  In general, I would say that states with higher crime rates are more likely to impose limits on voting by felons.

My own views are in between the extremes.  I think that we should forbid prisoners from voting, and that we should not, in general, forbid felons, once released, from ever voting.

I said, "ever" and "in general".  Now let me fill those in.  I think it is right to forbid felons from voting until they have completely finished their sentence, until they are no longer on probation and have paid any fines or restitution they may owe.  As long as they still owe something either to society or to their victims, they should not vote.  I also think that some crimes — first degree murder being the obvious example — are so severe that anyone convicted of them should never vote.  And I think that patterns of behavior that mark a person as a habitual criminal should also prevent them from ever voting.  For example, if a state has a "three strikes" law, it seems reasonable to me to include permanent loss of the franchise as one of the penalties.

And there is one kind of crime where losing the right to vote is an especially appropriate penalty.   Maryland disenfranchises people convicted of vote buying or selling.   I would extend that to other kinds of vote fraud.

But, excluding vote fraud, I think that if a person commits a single minor felony and serves their sentence completely, then they should have their voting rights restored — along with all their other rights, including the right to carry a gun.  Many states exclude anyone with a felony conviction from some occupations.  I think that's a more serious problem for most felons than disenfranchisement.  And, if we can't trust a person with a gun, then we shouldn't trust them with a vote.

(*Students of voting would be happier if prisons had their own precincts, though one can certainly understand why towns that have prisons don't want the prisoners to vote locally.)
- 11:15 AM, 20 June 2005   [link]


Disclaimer On Errors By "Mainstream" Journalists:   That I have noted an error in a post does not mean that the rest of the newspaper or broadcast is free of errors, or even that I did not see or hear other errors.  I encounter so many errors by "mainstream" journalists that I could not correct them all, even if I worked at it full time.
- 1:30 PM, 12 September 2006   [link]


KUOW's Gang Of Four:  Every Friday morning at 10, four local journalists discuss current events (and, often, their own columns) on the Weekday program on our local NPR affiliate, KUOW.  Usually, the host is KUOW's Steve Scher, and the three guests are Knute Berger, who edited an alternative paper, the Seattle Weekly, for many years (and now has what appear to be a couple of part-time jobs, one of them at Crosscut), D. Parvaz of the Seattle PI, and Danny Westneat of the Seattle Times.  (For many years, Susan Paynter of the PI was in the gang; after she retired from the PI, her place was taken by Parvaz.)

From time to time, I listen to the program, though I usually do so, as I first mentioned here, as a pathologist.  I am interested in why our "mainstream" news organizations are failing, why they are losing readers, watchers, and listeners, why "mainstream" journalists are held in such disdain by the public.  And listening to Weekday often gives me some clues on those questions.

The typical Gang of Four show is partisan, frivolous, often innumerate, and almost always far too Seattle-centric.  They are not partisan in the sense of supporting the Democratic party — though they do — but partisan in the sense that they treat moderates and conservatives very differently than they treat leftists.  They are frivolous in the topics they choose to discuss; nothing delights them more than a Republican sex scandal, even if if involves no sex and no laws broken, but they are unwilling to touch on many far more important issues.  They are, like many "mainstream" journalists, uncomfortable with numbers; they would have trouble, for instance, discussing budgets for any of the local governments.  Seattle has a population of about 570,000, which is less than a third of King County's population.  (And there are large numbers of people living in the counties to the north and south of King, Snohomish and Pierce, respectively.)  When the gang discusses local issues, they almost always discuss them from the point of view of that reactionary city, Seattle.  Progressive exurbs, such as Monroe, Puyallup and Covington, might as well be on another planet, as far as the gang is concerned.
- 2:40 PM, 12 October 2007   [link]


Urban Imperialists:  Many "mainstream" journalists, and more than a few Democratic politicians, treat rural areas as conquered provinces, to be exploited by the people in urban areas.  That exploitation can take many forms, but the most common form is restrictions on the use of land.  In some cases, those restrictions are so severe as to amount to expropriation.

Even "mainstream" journalists who are not urban imperialists generally know little about rural areas, and the people who live in them.  And few of them care to know much about people who live outside of out central cities.  (Exception:  Some would be interested in poor people in rural areas, if those people happen to be black — as some are.)

Partly this is a matter of partisanship.  As our rural areas have become more Republican, our our Democratic politicians and our "mainstream" journalists have become less sympathetic to the people living in rural areas, and their problems.
- 12:50 PM, 3 February 2009   [link]


Where Is The Statue?  The current background shows a group crossing the Snake River at one of the fords used on the Oregon Trail.
- 4:31 PM, 8 February 2004   [link]