Thursday, December 30, 2010

A Moral Argument for God

I recently listened to an interview with Francis Collins on Point of Inquiry, concerning his book The Language of God: A Scientist Presents Evidence for Belief.  Francis Collins is an icon of the "scientist who is also Christian" (along with Ken Miller).  He's a converted atheist, now evangelical Christian, but is pro-evolution, director of the NIH, and the former director of the Human Genome Project.  I was immediately struck by two things:

  1. he was converted to Christianity primarily by the arguments in Mere Christianity, by C.S.Lewis.  In this book, one of the main arguments centers around the Moral Law, it's universality and internal (to the human) nature of it.  I'll go into that more later.
  2. he is obviously a very smart guy, so his opinions (especially on evolution) need to be taken seriously (at least once).  So when he claims that the Moral Law cannot be the product of evolution, one had to at least not write that comment off immediately.

So, I went back and read Mere Christianity, which I had done in college some years back. I had an immediate visceral reaction against it, as he laid out these philosophical arguments that seemed much more like word games and bald assertions than anything approaching truth.  I admit that I am steeped in the methods of science, and find arguments that claim surety yet are not testable to be empty.  In reading Francis Collins' book, I have found that his arguments are essentially identical to Lewis', but couched in more modern scientific language.  (As a footnote, I was, pleasantly surprised to see that Lewis lumped evolution with gravitation in describing laws of nature.)

Moral Law and the Argument for God

The definition of Moral Law here is:

"denunciations of oppression, murder, treachery and falsehood; the same injunctions of kindness to the aged, the young, and the weak, of almsgiving and impartiality and honesty" [pg 24,  Collins quoting from Lewis)

The argument for God based on the Moral Law takes on three components.   The three components of the argument are the following:

  1. Moral Law is universal to all human cultures.
  2. Moral Law includes pure altruistic behavior (think Mother Theresa or Oskar Shindler here), which cannot be explained by evolution
  3. Moral Law is internal to humans.

(There is a fourth point, which Lewis ties specifically to Christianity, which is that we often choose not to obey this Moral Law.  This sets up the idea of free will, and the idea of sin.)

Why is altruism a problem for evolution?  Collins writes:

Agape, or selfless altruism, presents a major challenge for the evolutionist. It is quite frankly a scandal to reductionist reasoning. It cannot be accounted for by the drive of individual selfish genes to perpetuate themselves. Quite the contrary: it may lead humans to make sacrifices that lead to great personal suffering, injury, or death, without any evidence of benefit. And yet, if we carefully examine that inner voice we sometimes call conscience, the motivation to practice this kind of love exists within all of us, despite our frequent efforts to ignore it.

He outlines three common evolutionary arguments for the origin of Moral Law:

 

  1. One proposal is that repeated altruistic behavior of the individual is recognized as a positive attribute in mate selection. But this hypothesis is in direct conflict with observations in nonhuman primates that often reveal just the opposite—such as the practice of infanticide by a newly dominant male monkey, in order to clear the way for his own future offspring.
  2. Another argument is that there are indirect reciprocal benefits from altruism that have provided advantages to the practitioner over evolutionary time; but this explanation cannot account for human motivation to practice small acts of con- science that no one else knows about.
  3. A third argument is that altruistic behavior by members of a group provides benefits to the whole group. Examples are offered of ant colonies, where sterile workers toil incessantly to create an environment where their mothers can have more children. But this kind of "ant altruism" is readily explained in evolutionary terms by the fact that the genes motivating the sterile worker ants are exactlythe same ones that will be passed on by their mother to the siblings they are helping to create. [pg 27-28]

So, if you accept these points, then Collins writes:

If the Law of Human Nature cannot be explained away as cultural artifact or evolutionary by-product, then how can we account for its presence? There is truly something unusual going on here. To quote Lewis, “If there was a controlling power outside the universe, it could not show itself to us as one of the facts inside the universe—no more than the architect of a house could actually be a wall or staircase or fireplace in that house. The only way in which we could expect it to show itself would be inside ourselves as an influence or a command trying to get us to behave in a certain way. And that is just what we do find inside ourselves. Surely this ought to arouse our suspicions?”

Encountering this argument at age twenty-six, I was stunned by its logic. Here, hiding in my own heart as familiar as anything in daily experience, but now emerging for the first time as a clarifying principle, this Moral Law shone its bright white light into the recesses of my childish atheism, and de- manded a serious consideration of its origin. Was this God looking back at me? [pg 29]

An Analogy

When I read arguments of this sort, especially the last quote from Lewis, I am offended by the confidence of the language from a totally flimsy and untestable statement:  "The only way in which we could expect [a controlling power] to show itself would be inside ourselves as an influence or a command trying to get us to behave in a certain way" [emphasis mine].  Really?  How do you get that?  How can you test that claim?  How do you know what options are open for a controlling power?  Aren't you already assuming that there is something inside us, beyond natural law (i.e. dualism), in order for this to be true?

I'd like to propose an analogy.  I am not sure how far this will go, and where it will break down (as all analogies do), but I think it makes a point.  There is a universal law, which I will call the Eating Law.  According to this law we as humans have an internal voice telling us that we want to eat fatty foods.

  1. The Eating Law is universal to all human cultures.  We differ on the specifics, but we all have the voice telling us what we want to eat (i.e. pringles, Big Macs,  etc...)
  2. Eating Law includes pure gluttony as a behavior (think people who eat themselves to obesity and death) which cannot be explained by evolution.  how could behavior that reduces life expectancy, mating probability, and health possibly be selected for?  There are people who eat so much they can't even move!
  3. Eating Law is internal to humans.  We can look at the eating behavior, but we'll never be able to observe the actual urge to eat fatty foods.
  4. We often choose not to follow this law (i.e. we choose to eat salad instead of Big Macs)

This is not simply hunger, which we can see in other animals.  It is an urge to eat, even when you're not hungry, fatty foods.  Now, re-read Lewis:

“If there was a controlling power outside the universe, it could not show itself to us as one of the facts inside the universe—no more than the architect of a house could actually be a wall or staircase or fireplace in that house. The only way in which we could expect it to show itself would be inside ourselves as an influence or a command trying to get us to behave in a certain way. And that is just what we do find inside ourselves. Surely this ought to arouse our suspicions?”

Now, we have two such laws, the Eating Law and the Morality Law. Which one is the message from the creator?  How could we test this?  I could probably come up with more examples of universal tendencies which take the form of internal messages to humans, but I don't need to.  Coming up with just one is enough to show how this argument is completely empty.

I am perplexed that someone as smart as Francis Collins can't see this.  I am further perplexed that someone would be "shocked by the logic of this argument".

Wednesday, December 1, 2010

Are Science and Religion Incompatible?

I've recently found pointofinquiry.org, which has a lot of very interesting talks about science and religion, so it got me thinking again.  Some of the more interesting talks are:

First off, I need to point out that I am an atheist as well as a scientist.  Did my learning of science undermine my religion (I was raised Roman Catholic)?  Was I jaded about religion before learning science?  Does science promote atheism?  These are some of the questions one needs to consider with respect to these ideas.  Given my one data point, I can only state a correlation, not a cause-and-effect: that as I got older, and learned more science, I got less religious.  However, I can point out several observations I've made about the interplay between religion and science.  I think some of the hard-line atheists, in their own personal zeal, make claims about science that are overreaching and not entirely correct.  I also think that the accommodationist perspective, stating that science is perfectly compatible with religion is also not correct.  Here are my main thoughts.

Science is in conflict with some religions

If you believe that the world is 8000 years old, then you are wrong.  Almost all branches of science, from physics, chemistry, biology and geology, agree that the Earth is billions of years old.  There really is extremely little wiggle room there.  Of course, one can always say that God made the world appear billions of years old, but in fact it really is only 8000 years old.  If you do that, then you could just as easily state that the world was made yesterday.  Further, it is a challenge to think of a reason why this deception would be need to be done.  Finally, from a scientific point of view, it is content-free (CF): it makes no measurable predictions, does not suggest the next measurements to be done, nor is it testable in any way.  Science has demonstrated, historically, that CF statements have always turned out to be the equivalent of nothing.

If your religion depends on insisting that the world is 8000 years old, then science is in direct conflict with it.  Those religions that have put themselves in the cross-hairs of science, by making specific statements about the universe, are making the error of making the religion able to be falsified.  Historically speaking, this is exactly what has always happened with testable religious claims: science shows them to be wrong.  The religion then has two choices

  1. reject science (usually hypocritically at the same time reaping all the benefits of science, such as longer life, better health, and increased technology)
  2. call those statements metaphorical, and remove the testable part of the religion that was rejected by science

Most liberal religious faiths have tended to choose (2).  The Catholic Church, for instance, has a pro-evolutiuon stance and is in-line with science on many things (a departure from historical behavior).  In choice (2), the Genesis creation story is allegorical and not literal, and thus not in conflict with modern biology, especially evolution.  They merely state that God had a hand in the process, perhaps, for example, injecting souls into humans some million years ago.  The conservative religious faiths which believe in the literal Genesis story, with creation in 7 days about 8000 years ago,  try to replace evolution with creationism (in the form of, so-called, intelligent design).  They are just plain wrong.  Further, because they've stated that their religions depend on evolution being false, they have thus made it possible for science to demonstrate they are wrong.  They have then created the conflict between their religion and science.

Other religious claims that are testable, and shown to be false, include:

  • faith healing
  • intercessory prayer
  • special creation

Science is not in conflict with some other religions...but with a caveat

There are many scientists who are religious.  Some relatively famous modern ones include Ken Miller, a Catholic biologist from Brown University who testified in the Dover Evolution vs Intelligent Design trial, Francis Collins, the geneticist, and the physicist Freeman Dyson.  In each of these cases, we have to be careful in defining their belief.  Conservative Christians will often point to highly prominent scientists (such as Albert Einsten) who are religious as evidence of no conflict between science and religion.  However none of these people support the anti-science agenda of the conservative Christians, and most believe in a vague, impersonal "God", such as  "God" representing the mystery in the universe.  A far cry from the person-like entity described in the Bible, but by using the same word (God), these scientists have inadvertently played into the hands of people who would like to misuse the belief for their own purposes.

What is comes down to, as you look at religious scientists (which represent a very small minority), there are two types:

  1. scientists that are in fields far away from the "big-picture" questions (like, say, material science) and can thus maintain two opposite viewpoints at the same time.  People are very good at compartmentalizing their thinking.
  2. scientists who have scientifically untestable religious beliefs.  this includes the "God=universe" belief, or some vague spiritual belief.  Ken Miller falls into this category too...when asked whether Jesus had a Y chromosome (which could test for the virgin birth, for example), he says that he just doesn't have any data on that.  Even though some of his tests are, in theory, testable they aren't testable right now and perhaps, practically, never.

But shouldn't science stay out of religious issues?  Don't they speak about different things?

Stephen J. Gould (the great) used to refer to "non-overlapping magisteria" when speaking about science and religion.  Freeman Dyson refers to science and religion as "two windows that people look through, trying to understand the big universe outside, trying to understand why we are here."  I have already touched on part of the problem with this: where religion is testable, it has been shown to be false. All that is left are the untestable parts.  To me, this content-free religion is the same as not believing in anything.  Perhaps it gives someone comfort to believe in a vague, impersonal spirit "out there", but I dislike using the word "God" when referring to the mystery and wonder of the universe itself. Using this word communicates something very different to different people, and implies something that I believe is unwarranted.

Richard Dawkins, in one of the point of inquiry interviews, shows that this non-overlapping magisteria is really a farce.  He puts it this way...imagine if, at some point, we uncover archeological evidence with some molecular biology that gave you evidence for, say, the virgin birth of Jesus.  Do you think that the Christian religions in the world would say, "Oh no!  Non-overlaping magisteria!  We can't use science to speak about religious issues!".  Of course not!  They'd be screaming it from every church.  It's only because there is no evidence for any religious claims that we posit the non-overlap of religion and science.

So, in a nut-shell, where religion and science meet, religion either directly agrees with science (in which case it is redundant) or it is wrong.  Most of the testable religious claims have been shown to be wrong.  Where they don't meet, religion is content-free.  If someone decides to find comfort in that, then that is their issue, and science can't really speak to it.  I just don't find any value in it.

Religious vs Scientific Thinking

At its core, however, religious thinking and scientific thinking are nearly opposites.  Religious thinking relies on anecdotes, statements from authority, faith without evidence, and mistakes of causation from correlation.  Each of these mistakes arise from common human failings of reason and perception, for which the scientific methods have been developed to avoid.  To do proper science one does not rely on anecdotes or authority, other than nature as the final arbiter.  Evidence is everything, and certain knowledge is never achieved in science (as opposed to religion).  So it is no wonder that the more scientific you are, the less appeal religion tends to have.  But it also makes you less susceptible to the various guises of pseudoscience and less susceptible to being hoodwinked by cranks and quacks.

It is my opinion, then, that the best remedy for religion is simply to teach as much science as possible and let the religion problem work itself out as a result.  I don't think that the in-your-face strategies of the new atheists, such as Dawkins, are particularly effective at reducing religion.  It has a purpose for rallying the troops, bringing closeted atheists out, so I wouldn't dissuade him from this approach if he feels that that is the primary outcome he wants.  I also don't think we should compromise and say there is no conflict between science and religion, which leads to pandering for political reasons with the truth as a casualty.  There are fundamental conflicts between many religions and science, and those should be pointed out even if it is uncomfortable for those believers.  Science isn't in the business of making people feel comfortable with their beliefs.  Thats the role of religion.

Sunday, November 21, 2010

Is -x^2 positive or negative?

So, is -x2 a positive, negative, or undefined quantity for real-valued x?  Ask any physicist or mathematician and they will say that it is a negative number for real valued x making things like: exp(-x2) between 0 and 1.  That is why it came as a BIG surprise to me that computer scientists don't think that, and a program like Excel will interpret:

=-5^2

as positive 25!  After taking quite a while debugging a student problem calculating the normal distribution in Excel, it got me on a quest (and an argument with a colleague) to figure out who else thought this way.  I checked Matlab, Mathematica, Python, and Google as well as a calculator on the computer.  All interpreted -5^2 (properly) as -25.  To do otherwise, I believe, is perverse for any application that is doing mathematical applications.  I was directed to this page, which outlines many languages.  Pretty much just Excel, COBOL, Chipmunk BASIC and a few small scripting languages take the "unary minus" approach, which makes "unary minus" have precedence over exponentiation.

I am not sure why anyone would consider this a good idea, for working with actual math equations.  Of course one could add parentheses, but which is clearer:

y=exp(-x^2)

or

y=exp(-(x^2))

The second is obviously not ambiguous, but less clear.  Anyway, that is the entire reason why we have order of operations, so we don't have to do:

5+(3*4)-(2*3)+(2*(3^3))

So, Excel, come into at least the 20th century and figure out that exponentiation trumps "minus", whatever you want to call it.

 

Glenn Beck on Science

The following audio from a Glenn Beck show is illustrative of many things. Although he demonstrates incredible ignonance of evolution in this clip, he makes some good points...they just aren't the points he thinks he's making.

His main points are the following:

  1. "I don't know why it is unreasonable to say this...I'm not God, so I don't know how God creates...I don't think we came from monkeys...I think that's ridiculous...I haven't seen a half-monkey-half-person yet...did evolution just stop?"
  2. "They have to force [the idea of evolution] down your throat...when anybody has to force it, it's a problem...you didn't have to force that the world is round...[...]...you don't have to force the truth..you just keep adding evidence and evidence until it becomes self-evident."

Now, point (1) demonstrates a profound ignorance, willful or otherwise, of the claims of evolution.  First, no biologist says we came from monkeys...monkeys and people are just as "evolved" as each other, and that they share a common ancestor which would have both features of monkey and human.  We have many of these transitional fossils (for a very nice summary see the Transitional Fossil FAQ).  It's really such an elementary error, that he should be thrown off the air just for spreading such ignorance.

However, I would like to focus on point (2) which perhaps seems reasonable: that truth should be self evident, and not have to be forced.  In a way, one might think that science works this way because everything should be repeatable, and we don't take arguments from authority.  However, there is a problem.  Although it is true that the fact that the world is round did not have to be forced (in contrast to the bogus propaganda story about Columbus demonstrating it), one does need to actually look at the facts in order to be able to judge...truth can be self evident but it requires one to actually look at the evidence.  Take evolution, for example.  If you look at the genetic data showing the differences in base-pair counts between animals, this single number mapped out for all animals demonstrates a tree structure (http://en.wikipedia.org/wiki/Common_descent and http://www.talkorigins.org/faqs/comdesc/section1.html).  Now recall that Darwin did not know anything about genetics, so this is a prediction made 50+ years before the measurements.  When you actually look at the evidence for evolution it is absolutely self evident.  The problem is several-fold:

  1. people refuse to even look at the evidence, choosing instead to willfully remain ignorant thinking they understand it.
  2. There is a certain amount of technical information necessary to understand any science, and that takes work.  If you don't want to do that work then you'll remain ignorant of it.  As an extreme example, quantum mechanics is definitely true, but it neither self evident nor intuitive (and I don't see conservative talking heads railing against it)

What's the solution?  I'm not sure, but challenging people to learn is a start.

 

Saturday, October 16, 2010

A walk down memory lane

I just found a page on "How to Find a Formula for a Set of Numbers".  It's a cool little procedure for taking a series, like:

2, 8, 9, 11, 20

and producing a polynomial to give you the next ones in the series, like:

n3- 17/2 n2+ 49/2 n - 15

where n is the term number, starting from n=1.  Try it out!  Anyway, it was a method I learned in high school math league, and thought it was so cool I wrote a BASIC program on the old TRS-80 computers to do it.  I had forgotten how to do it, and it was fun to see it again.  I particularly liked the comment on the page:

"""If someone gives you the sequence, say, "1, 4, 9, 16", you could run them through the above process and get the answer that the person is probably looking for: the rule is n2 so the next value is 25. But you could also invent any number as the next number in the sequence, say 42, and come up with a rule for "1, 4, 9, 16, 42". Feel free to work it out. It comes out to:

 

17/24 n4 - 85/12 n3 + 619/24 n2 - 425/12 n + 17
and the next term is then 121.

So if you want to be obnoxious, the next time you are given a quiz of "find the next number in the series" problems, just pick any number you like and fill it in, and you'll be completely correct. You'll probably get a failing grade on the test, but you can enjoy the smug satisfaction of knowing you were right."""

I knew a kid who, because of a ridiculous fluke, had to redo some of his middle-school competency tests in high school.  So, when presented with a series like 2,4,6,8,... he did this on a test (and yes he did fail the test and have to redo it).  He was also shown a number of clocks, and asked what time does this show, and for all of the answers put "analog time".

Friday, October 8, 2010

Power UnBalance

I love watching infomercials, but always wonder how much the sellers are exaggerating.  Take this infomercial for the "Power Balance" bracelet, which is claimed to increase balance and coordination:

http://www.youtube.com/watch?v=A_Ow-ZGMy5o

Now, go to this link which shows you how it actually works:

http://www.youtube.com/watch?v=Piu75P8sxTo

Make sure to watch the whole thing, because they give away the "trick" near the middle.  It is useful to go back afterward and watch the first one, now that you know the trick.

The real question, then, is: what should you do if you know a friend is considering buying this, or worse, has already bought it?  When I showed these videos in my class, I was told that the football team had purchased them already.  When some of my students presented them with the evidence, their response was that they didn't care whether it worked or not.

Astounding!

Saturday, September 25, 2010

Multiple Model Comparisons Revisited

Introduction

In a previous post, I hinted at how to do multiple hypotheses testing, using the ψ-measure. It turns out to be much clearer just using the posterior probabilities. The ψ-measure has a nice intuitive feel for the two-hypothesis case, but becomes convoluted in the multiple hyptheses case. Further, when introducing the application of Bayes theorem for students, I have found it to be clearer to follow the following procedure. We first look at Bayes theorem directly, for N hypotheses:

NewImage.jpg

We then calculate the numerator only, for every possible hypothesis:

 

NewImage.jpg

 

calculate the sum of all of these values,

NewImage.jpg

and then normalize

NewImage.jpg

The Octopus, Again

 

From the Wikipedia article, we have the following data:, which gave us correct=12 out of N=14:

  NewImage.jpg

NewImage.jpg

NewImage.jpg

The hypotheses that we consider are the following:

H = “Octopus is psychic, and can predict future (sports) events with 90% accuracy” R = “Octopus makes random choices” Y = “chooses flags with big yellow stripes 90% of the time” G = “chooses Germany 90% of the time”

Notice that both models Y and G, give us correct=12 for N=14 (if the “choosing Germany” chooses Spain in the Netherlands match, because of the similarity). The prior for the psychic octopus is, again, the very generous p(H) = 1/100. The two other non-random models should be more likely, before any data, so I take them to be p(Y)=p(G)=1/20. The random model, being the most likely, has the rest of the prior probability, p(R)=0.89.

Now we calculate the numerators:

NewImage.jpg

Sum the values,

NewImage.jpg

and divide. achieving

NewImage.jpg

Thus, the two flag models went from being rare compared to random to being much more likely than random, and certainly much more likely than psychic. Bayes theorem, properly applied, is a quantitative embodiment of Carl Sagan’s famous quote “extraordinary claims require extraordinary evidence”. It is not just that the evidence must be extraordinary (like 999 correct out of 1000), but the evidence must be extraordinary to address all of the, somewhat rare but possible, hypotheses that would come up as much more likely given the initial result. The process of science is to perform experiments to address these alternative hypotheses.

Sunday, September 12, 2010

God and Hawking

From the book “The Grand Design” By STEPHEN HAWKING And LEONARD MLODINOW

....
Newton believed that our strangely habitable solar system did not "arise out of chaos by the mere laws of nature." Instead, he maintained that the order in the universe was "created by God at first and conserved by him to this Day in the same state and condition."

....

The press is pitching this book as a denial of God, claiming that Hawking has said that God does not exist. The media never seem to get the nuances of logical thinking, and its consequences.

What Hawking and Mlodinow are doing is a modernization of an approach used by Laplace (1749-1827) (http://en.wikipedia.org/wiki/Pierre-Simon_Laplace).  He worked on many things, including the dynamics of the solar system.  When Newton (http://en.wikipedia.org/wiki/Isaac_Newton) published his laws of dynamics 100 years earlier, he demonstrated that the speeds of the planets could be derived from a simple law of gravity.  In this way, Newton connected the Earthly things with the "Heavenly" things.  However, it was unclear to Newton whether the orbits of the planets would remain constant (as his religious philosophy would state), or if they would be unstable, change, and possibly fly apart given enough time.  He posited that one of the roles of God would be to nudge the planets, here and there, to keep their orbits stable.

Laplace, performing his calculations more precisely than his predecessors, was able to determine that the orbits would in fact be stable, without any extra tinkering.  Napoleon, when presented with the work of Laplace, asked him: "M. Laplace, they tell me you have written this large book on the system of the universe, and have never even mentioned its Creator."  Laplace replied, "I had no need of that hypothesis."

He did not say that there was no God (although that is what he believed), but that the concept of God was not necessary to explain the things that he was explaining using physics.  This included the formation of the solar system from a compressing ball of gas (due to gravity), which then forms the Sun in the center and the planets orbiting around.  This is essentially the model still in use today!

What Hawking is doing is basically the same thing, but with the origin of the universe.  Essentially the current model allows for the possibility of many universes to simultaneously exist and that, like a lottery winner, our universe supports life.  It may seem that the universe is "fine-tuned" to support human life, and that this would support the notion of an intelligent designer, Hawking is making the argument that a designer is not needed with our current understanding.  Like a lottery winner stating that the odds of winning are astronomical, and yet they won, and then reasoning that there was some design in this choice even when there wasn't.  As long as you have enough people playing (or enough universes) you'll eventually observe the unlikely, and that unlikely winner will feel singled out.  Hawking argues that the lottery winner (the life on Earth), is arguing the same way when it invokes a designer when it doesn't need to.  Hawking doesn't state "God doesn't exist", because that statement cannot be proven, but he simply states that it is an unnecessary hypothesis for the understanding of the origin of the universe.

Of course, *specific* Gods can be disproven.  For example, it is clear from many lines of evidence that the Earth is more the 6000 years old and that there never was a global flood.  However, you cannot disprove the notion of a God that creates the universe and is then hands-off, like deists commonly believe.  It is completely untestable.  It is also unnecessary, according to Hawking.  This doesn't make it wrong, it is just unnecessary in the same way that we don't need to invoke the divine when understanding how an apple falls from a tree.

Wednesday, September 1, 2010

Why pseudoscientists like the chi-square test (and why it shouldn't be taught)

In a prior post I outlined how orthodox statistics can lead to the either-or logical fallacies common in pseudoscience, like astrology and ufo-ology.

In this post I focus on the &chi2 test, it's pathologies, and why it is so useful for a pseudoscientist. The example is lifted from E. T. Jaynes' book "Probability Theory"

The two problems with &chi2 are:

  1. it violates your strong intuition in some simple cases
  2. it can lead to different results with the exact same data, binned in a different way


Both of these properties are useful to the pseudoscientist.

Intuition and Chi-square: The Three-sided coin



In each of this case we will have some data, and two models to compare which try to explain the data. Intuition strongly favors one, and &chi2 favors the other. One of my favorite problems is the three-sided coin: where the coin can fall heads, tails, or on the edge. Imagine we have two models for a relatively thick coin:


  • Model A: pheads=ptails=0.499, pedge=0.002
  • Model B: pheads=ptails=pedge=1/3


And we have the following data:


  • N=29: nheads=14, ntails=14, nedge=1


Which model are you more confident in? Model A of course! If we use the &psi-measure for goodness of fit with these two models, as defined in my prior post, then we have (remember: smaller &psi means more confident in the fit, just like smaller &chi2):


E7E94805-9E10-451E-9B95-C8EB2BA875C7.jpg




7AFD29D5-C801-416E-82AC-F9B363147B22.jpg



with &psiB-&psiA=26.85 which makes model A more then 100 times more likely than model B (a &psi difference of 20 would be exactly 100 times). Perfectly reasonable. What about &chi2?


31844A12-1F3E-4D51-B2D0-EFAE00BE66A7.jpg



which makes model B slightly preferable to model A! Amazing! Where is this coming from? Apparently it is coming from the somewhat rare event of an edge-landing. If our data had been instead

  • N=29: nheads=15, ntails=14, nedge=0

then we'd have


  • &psiA=0.3
  • &psiB=51.2

and

  • &chi2A=0.093
  • &chi2B=14.55

where now both measures agree that model A is superior.




Why do pseudoscientists love the &chi2 test?
Answer 1: Because all they need to do is wait for that inevitable, somewhat rare but still possible, data point and &chi2 yields a pathologically high value


The &psi-measure and log-likelihood



To understand the other problem with the &chi2 test we need to understand what the &psi-measure is doing. As above, imagine we have a set of observations Oi. We define the total number of observed points and the relative frequency of each observation,


23229268-8820-42F8-BF8C-C984679DCB51.jpg



The maximum likelihood solution for the probabilities of observing Oi for each class, i, is just the relative frequency of each observation. This is the "just-so" solution, where we estimate the probability of seeing 14 heads in 29 flips as p=14/29. This "just-so" solution will have the closest match, and the highest likelihood (by definition). If we have a model which specifies a different set of probabilities for each class, then it's likelihood is simply


71272C22-E8B6-4497-82DF-FBDD89729630.jpg


The &psi measure can be rewritten as


B363F3D7-1964-421A-A8E2-B19020D0117A.jpg



So you can think of the &psi-measure as comparing a model with the "just-so" solution (which has maximum likelihood). Further, subtracting one value of &psi with another (for different models) performs the log-likelihood ratio between the models. A proper analysis should include prior information, which can be done almost as easily.

An almost equivalent problem



Imagine that we have a coin with 6 faces, and we are comparing the following models:


  • Model A: p = [0.499/2, 0.499/2, 0.499/2, 0.499/2, 0.002/2,0.002/2]
  • Model B: p = [1/6,1/6,1/6,1/6,1/6,1/6]


And we have the following data:


  • N=29: O=[7,7,7,7,0,1]


where I have listed the probabilities and the outcomes for each face. Notice that, grouping them together in pairs we retrieve the same as the first example. Thus when comparing the two models, with this equivalent problem, we should get the same value. Because the size of the problem changed, the individual &psi values will be different (larger) because there are more terms in the "just-so" solution. However, the difference between the models should be the same. The results are:


  • &psiA=11.35 (old value 8.34)
  • &psiB=38.2 (old value 35.19)

with &psiB-&psiA=26.85 (old value 26.85...the same!), and

  • &chi2A=32.6 (old value 15.33)
  • &chi2B=11.76 (old value 11.66)


The &chi2 for one of the models (Model A) has been inflated quite a lot relative to the other model. This means that, depending on how you bin the data, you can make whichever model that you are looking at more or less significantly different, without changing the data at all.




Why do pseudoscientists love the &chi2 test?
Answer 2: Because all they need to do is bin their data in different ways to affect the level of significance of their model over the model to which they are comparing


Still taught?



So, why is the &chi2 test still taught? I don't know. It has pathological behavior in simple systems, where somewhat rare events artificially inflate its value, and it can be easily used to prop up an unreasonable model simply by rearranging the data. Why not teach something, like the &psi-measure, which is grounded theoretically in the likelihood principle and does not have such pathological behavior? If you prefer to use the log-likelihood instead, then that would be fine (and equivalent).

I think it is about time to purge the &chi2 test from our textbooks, and replace it with something correct.

Tuesday, August 31, 2010

Orthodox Statistics Conducive to Pseudo-Science

I have just realized that the thought process used in orthodox statistics is conducive to pseudo-science. It adds, in my opinion, to the long list of reasons why Bayesian inference is demonstrably superior (also see here). Let me show with a couple of simple examples.

Astrology

From this skeptical analysis of some astrology data, listing the numbers of famous rich people in each sign, we see the use of the chi-squared goodness of fit test. The data are:

SignNumber of People
Aries 95
Taurus 104
Gemini 110
Cancer 80
Leo 84
Virgo 88
Libra 87
Scorpio 79
Sagittarius 84
Capricorn 92
Aquarius 91
Pisces 73
Total 1067


To apply the chi-squared test, we simply compare the above numbers to the expected numbers if completely random, which is 1067 people/12=88.9 people according to:

7660DA25-CEEC-4B31-A199-76CEA69E5015.jpg



where O are the observed data and E are the expected counts. Once we have the chi-square value and the degrees of freedom (11 in this case), we can look up in tables to get the p-value:


5785E00F-983A-4B86-AF40-5F141440DD5A.jpg



Normally, this might be the end of the story, given that there is not even close to a significant value (usual cut-off around p=0.05).

Subset of the Data



So, if we only take the extreme values, say:

SignNumber of People
Gemini 110
Pisces 73
Total 183


then we calculate a different chi-squared, with 1 degree of freedom, and get


83387956-03D9-4DB0-810E-4BD2E7D836F1.jpg



Now this is pretty silly: of course, if you take the extreme values of 12 numbers, and pretend that they came from a 2-category situation, then it'll appear more significant. What about lumping 6 points together, say Capricorn to Gemini (the first part of the year) and the second part. In this case we aren't cherry picking, and the sums should be less significant than the individual data. We then have:

SignNumber of People
Capricorn-Gemini 565
Cancer-Sagittarius 502
Total 1067


And we expect 533.5 people in each category. Notice that we went from (the most extreme) 20 person difference from expected in about 100 to a 30 person difference in 500...closer to the expected. What do we get from our chi-squared test?


D9E69A77-09CD-43A1-88CF-256EB300D492.jpg



The test says that this is significantly different from random, more than the individual data! At least the goodness of fit measure, chi-squared value, went down to denote a closer fit to expected but the reduction in the number of data points changes the test quite a lot.

A different measure



E.T. Jaynes suggests in his book to use a different measure of goodness of fit, the &psi measure closely related to the log-likelihood


A210014E-5B3B-4DB4-AE14-63D0EB367E3D.jpg



Using this measure on the above examples, we get

  • All data: &psi = 28.9
  • Extreme data: &psi = 39.1
  • Lumped data: &psi = 8.1
which is completely in agreement with our intuition. The chi-squared test does not match our intuition, and seems to give significance to things that we know shouldn't be. But what about the test with the &psi-measure? How can we tell whether it is a significant difference? One could, in theory, give an arbitrary threshold but that would not be particularly useful, and would not be what a Bayesian would do. What a Bayesian would do is compare values of the goodness-of-fit measure to different models on the same data. It makes no sense, if you have only one model, to reject it by a statistical test...reject it in favor of what? If you have only one model, say Newton's Laws, and you have data that are extremely unlikely given that model, say the odd orbit of Mercury, you don't simply reject Newton's Laws until you have something else to put on the table. The either-or thinking of orthodox statistical tests is very similar to the either-or thinking of the pseudoscientist: either it is random, or it is due to some spiritual, metaphysical, astrological effect. You reject random, and thus you are forced to accept the only alternative put forward. I am not implying that all statisticians are supportive of pseudo-science, and they are often the first to say that you can only reject hypotheses not confirm them. However, since the method of using statistical tests does not stress the searching for alternatives, or better, the necessity for alternatives, it is conducive to these kinds of either-or logical fallacies. An example of a model comparison, from a Bayesian perspective, on a problem suffering from either-or fallacies can be found in the non-psychic octopus post I did earlier.

Friday, August 27, 2010

The Non-Psychic Octopus

Introduction


I saw in the newspaper an article about a supposedly psychic octopus, which predicts world cup matches by making a choice between two different foods labeled by the team flags. Paul the Octopus has an impressive record of 12 correct out of 14. Or is it impressive? How can we determine whether this performance is evidence for psychic behavior, or something else. A typical statistical analysis might start with the null hypothesis that the octopus was random, so was choosing the teams with probability p=0.5. The likelihood of getting 12 right in 14 is

217E3119-BC22-4A09-B534-D62BAA0BF53C.jpg


which is fantastically strong against the null! Even if you do the p-value test for the the correct data being more extreme, you get p-val=0.00646.

So, we reject the null, and the octopus must be psychic!...(or not)



Bayesian Analysis Against Random


Let's look at this another way, and perhaps we can gain some insight. It will be convenient to talk about odds, rather then probability, and further to use the log of the odds so that this becomes an arithmetic problem. The odds is defined as the ratio of the probability for a hypothesis, H, and the probability for the inverse, not H.


E5899678-9897-42AF-8CBB-5D687A02A731.jpg



We define the log-odds, or evidence as defined by E. T. Jayes,


BAAC0566-3B79-4B44-8F0D-C5046788B491.jpg



A few comments before we commence with more calculation. The prior evidence reflects our state of knowledge before we see the data. How likely is it that an octopus is psychic? Most reasonable people would say highly unlikely. Generous odds would be 100:1 against, although personally I'd probably put it at least a million to 1 against. Let's be generous. That gives us a prior evidence of


94E23FDD-36D3-4B6A-B72C-7C759FAAE1B1.jpg



If we had been naive, and set equal odds, then this evidence would be e=0. So we start with evidence e=-20 for a psychic octopus (which is strong evidence against it, because e<0), and then we observe the data. If we assume that a psychic octopus is right 90% of the time, and that the only alternative is a random octopus correct 50% of the time, then we have added evidence for each correct answer:


85F4C7A1-EFC7-4E8C-88B0-E98B846B1F97.jpg



Each incorrect answer gives:


D48A7B4A-112D-4BC6-A267-95B797A15F9C.jpg



The evidence gets pushed up from the prior with each correct answer, and down for each wrong answer. Notice how wrong answers are penalized more than right answers. This is because the psychic octopus is pretty good (p=0.9). We get a final (posterior) evidence for 12 correct and 2 wrong:


5BC4EE45-59BD-4A67-A2FF-779E7EC92145.jpg


which is about 2:1 odds against the psychic octopus.

More to the Story



Most pseudoscience gets propagated by people who reason naively. They will say that there are two possibilities, say random and psychic, and they they must both be equally likely before the data. So, when rare data is found, they reject random and claim this is evidence for psychic phenomena. This line of reasoning is incorrect for two reasons:

  1. random and psychic are not equally probable a priori - random is much more likely in cases like this
  2. there are more possibilities


We already saw how point (1) can be handled by proper prior information. Point (2), with multiple hypotheses gets mathematically a bit trickier (there are more terms to carry around) and is thus messier, but conceptually is fairly straightforward.

We have two hypotheses so far:

H="Octopus sees the correct future 90% of the time, and is psychic"


R="Octopus chooses randomly."



Let me introduce two more hypotheses.

Y="Octopus chooses flags with big yellow stripes 90% of the time"


G="Octopus chooses Germany 90% of the time"



How would you choose the prior probabilities for these hypotheses? Personally, as I said before, I'd have p(H) way below p(R) by about a factor of a million, but being generous, let's put it about a factor of 100. What about p(Y) and p(G)? I'd say that these might be comparable to random or, if I knew something about the vision of octopi or how the person feeding the octopus might rig the food in the direction of his favorite team, I might even have p(Y)>p(R) or p(G)>p(R). Certainly p(Y)>p(H) and p(G)>p(H). So what happens with the data?

For hypothesis Y, there are N=14 games of which the octopus chooses 12 with bright yellow stripes (there is one where it chose Germany over Ghana and should have chosen Ghana which has a bigger strips, and another with Germany and Spain where Spain should have been chosen). For hypothesis G there are N=14 games and the octopus chooses 12 for Germany (2 teams are chosen that are not Germany, and one match where Germany wasn't a choice and it chose Spain, which has the closest flag). Thus, the data support both of these hypotheses exactly as much as the p=0.9 psychic hypothesis. Therefore, the evidence will push these hypotheses up by as much as the psychic, over the random, and will make the psychic octopus even less likely.

So, when you hear fantastic claims supported with a comparison to random, the two things you must do are:

  1. Ask yourself what the prior probability of the fantastic claim is. Even if a random explanation is very rare, it will probably still be favored against the fantastic claim.
  2. Ask yourself what other possibilities, even if unlikely, could explain the data. Since the fantastic claim is exceedingly unlikely, even somewhat unlikely explanations may be supported by the data more than the original fantastic claim.

Saturday, July 10, 2010

A new and content-free model

In his article "A New and Effective Climate Model", Stephen Wilde (guest posting on Anthony Watts' blog) states his dissatisfaction with current climate models, and proposes another "model" for climate which he hopes will improve the state of climate modeling in general. In the article he has items like:



  1. Solar surface turbulence increases causing an expansion of the Earth’s atmosphere.
  2. Resistance to outgoing longwave radiation reduces, energy is lost to space faster.
  3. The stratosphere cools. Possibly also the number of chemical reactions in the upper atmosphere increases due to the increased solar effects with faster destruction of ozone.
  4. The tropopause rises.
  5. etc...


This list continues for 26 points, not an equation in the mix. So why am I so hooked on equations? Take the first item, and call "Solar surface turbulence" T, and the size of the atmosphere, A. Saying T goes up, so A goes up, could be like:

A2673E80-F8DF-4075-9854-7A956B9CE989.jpg


which would predict a nice linear response. What about this:

F7B70424-1CA7-4D17-9A93-B57938360DA8.jpg


or this?

D11EAF1C-6B62-446A-90CD-A77E4C6D597E.jpg


Each of these is a translation of "when T goes up, A goes up", but they have radically different forms, and they have radically different effects. You can't build a proper scientific model in words alone. Words are not precise, and there are many different ways to translate them into something that is precise, that can actually make meaningful predictions.

A model of just words is not really a model, in the scientific sense. Lord Kelvin said it best:

"In physical science the first essential step in the direction of learning any subject is to find principles of numerical reckoning and practicable methods for measuring some quality connected with it. I often say that when you can measure what you are speaking about, and express it in numbers, you know something about it; but when you cannot measure it, when you cannot express it in numbers, your knowledge is of a meagre and unsatisfactory kind; it may be the beginning of knowledge, but you have scarcely in your thoughts advanced to the state of Science, whatever the matter may be."

Thursday, June 24, 2010

Pascal's Global Warming

I was just alerted to the following video supporting strong action for climate change:

ssp_temp_capture.png

The argument is a little dated, actually, most famously presented in the 1600's well before global warming was a concern. The video uses a rehash of Pascal's wager, replacing personal salvation with policy to deal with climate change. Essentially it states that, if the worst case scenario is very bad, then it is better to deal with that worst case scenario with the possibility of it being a waste of resources, because the alternative (not doing anything and being wrong) is much worse. In Pascal's case, he was arguing for a belief in God. If you believe in God and he doesn't exist, then things aren't that bad (some wasted time and effort). If you don't believe in God and he does exist, then you're in big trouble. The expected value of your reward definitely is maximized by avoiding the worst-case scenario.

There are several problems with this argument, as it applies to global warming, many of which apply to the original argument as well. First, one has to look at the probabilities of the events, and not just their existence. It is true that one of the following two statements is correct:

  1. there is global warming, and humans are causing it
  2. there is not global warming or humans are not causing it
If we assign probabilities to the individual events, then the correct policy may not be to address point (1), if it is sufficiently unlikely. Further, even if we assume there is global warming, and humans are causing it, both the amount of the warming and impact of the warming are needed for making policy. So, if the warming is most likely 0.5 deg in 100 years, versus 2 deg in 100 years, then the policy choices should be modified accordingly. Even though there may be a slight chance of a catastrophic warming, we do not need to plan for it if the chance is slight enough. Finally, dumping resources into a non-problem is not just a "waste of money", but of lives. The amount of money being discussed here is in the trillions, which can go a long way to fighting problems that we know are here and are a problem (hunger, corruption, extreme poverty, religious extremism, etc...). We have to look at where we need to place resources in the most efficient way to address the many, and serious, problems that we know exist in the world.

Now a similar, but more persuasive motivation for climate change regulation is simple: we are dependent on unstable governments for our energy and our energy sources have serious environmental impacts. These facts are currently costing us two unwinnable wars and an oil spill that will take decades to clean up. If we could divert the costs that go into those wars and industries, and put them into non-fossil fuel alternatives (i.e. nuclear fission and fusion), then we have a chance of significantly improving our national security, environment, and future. So, in dealing with a known, significant problem we can solve another (possible) problem (i.e. global warming) as a bonus.

Tuesday, June 8, 2010

Religion and Scientists

A friend of mine posted this Slashdot brief on Facebook, which also links to this summary. The articles both imply a much stronger religiosity among scientists than is traditionally assumed, quoting the work of Elaine Ecklund. Some key quotes from these two short summaries are:


  • "Fully half of these top scientists are religious. Only five of the 275 interviewees actively oppose religion"

  • Even among the third who are atheists, many consider themselves "spiritual."

  • The study reveals that scientists often practice a closeted faith, worrying about how their peers would react to learning about their religious views.

  • The '"insurmountable hostility" between science and religion is a caricature, a thought-cliche, perhaps useful as a satire on groupthink, but hardly representative of reality,' writes Ecklund.



So, when I read this, I thought "Wow! Half of these scientists are religious? Spiritual atheists? Closeted faith? That's amazing stuff!". So, as a scientist myself, I decided to look at the study itself. I confirmed what another blogger stated: essentially the conclusions come from some "new math", and vague statements.

The study included these results:

  • 34% chose “I don't believe in God,”
  • 30% chose “I do not know if there is a God, and there is no way to find out.”


which equals 64%. So "fully half" must mean "less than 36%". That's some good arithmetic! Oh, and


  • 8% chose “I believe in a higher power, but it is not God.”


So I guess religious doesn't imply belief in God, so it goes in the very vague category. If religious includes a belief in God (and I guess in her book she uses "traditionally religious"), then "fully half" must mean "less than 28%".

So about 25% of all scientists profess a belief in God, compared to a public of about 80%. That's a pretty stark contrast! I'd wager that there must be something (or somethings) that makes this negative correlation between science and religion. I think it is difficult to argue that there is not some inherent conflict between the two given this contrast. What about the closeted religious scientists? Those shouldn't appear in these anonymous polls, and thus do not affect any of these numbers. If a scientist doesn't want to bring up their religion with their peers, then 1) what does that say about the strength of their beliefs and 2) perhaps the religious aspects of their life are irrelevant to their scientific work. The closeted religious were not mentioned at all in the study, by the way.


Spirituality



In another monograph, Ecklund discusses her work. In it she states:


For many of the natural scientists, in particular,
knowledge of the spiritual comes directly from their work. For example, according to one physicist,

When I travel to observatories...and when I finally just have enough time to try to
think of my place in the world and the universe and its vastness, it’s then that I feel
the connection to the world more than I do, say, sitting here in my office. And so that
for me, that’s the closest I can come to a spiritual experience.


This excerpt and the many others like it show that, for some scientists, rather than science
replacing religion, spirituality may be replacing religion.


In this way, it is possible for atheists to be "spiritual". To me, I think this is really bad terminology and easily misused. For most people, "spiritual" implies some some "spirit", or an existence separate from the body. These scientists are making no claim like that, they simply mean a feeling still bound to the biology of our bodies. Unfortunately, by using the term "spiritual" it can be used (as here) to enhance the notion that these scientists are religious, or leaning toward religion, which is misleading at best. A religious friend of mine once recently said that she thought I was spiritual. I thought the term was basically meaningless and ill-defined, and still do.

Religious Scientists



Is it surprising that there are religious scientists? Not at all. Because most of the public is religious, most scientists are raised with some religion. It is also human nature to separate different, conflicting, modes of thinking in different contexts. It is very easy for someone to study superconductivity and not have it interfere or conflict with any of their views of the divine. I've written about religion in other posts, and will only add that I believe that much of religion comes from the skills we have to help our survival: it is better to see a tiger when there isn't one than to not see a tiger when there is one. From there, combine with cultural structures, it is easy to see how religious ideas propagate. The great thing about science is that its methods allow us to transcend these false-positives, and arrive at ways to discern truth from fiction. Because we're human, it doesn't work 100% of the time, but in the long run it has been our most successful human creation.

It is no surprise to me that some scientists are religious. It would have been a surprise to me to hear that "fully half" are religious, and as it turns out that isn't true.

One last TV post, about Battlestar Galactica

No TV posts for a while after this one...just trying to clear my desk.

There are a number of things that bother me about BSG. This list is here for anyone who
wants to challenge me on any of this, comment please! It's in the form of loose-fitting
notes, written over the course of watching the series. There are spoilers, etc...

1)
Once they find out that Cylons look like us, and they consider a Cylon
detector, they seem to hand the entire problem over to Dr. Baltar.
There seems to be no consideration about whether he is a Cylon, or
in league with the Cylons. What could they do differently? Have the
Dr make something that is independently testable, perhaps by Gaita or
others. They don't even consider it.

2)
I find it straining in the extreme to believe that there is no
straightforward test to tell a Cylon from non-Cylon, given the actions
and words of the various people. Cylons:
1) are stronger than people
2) do not tire
3) are able to be uploaded, at least at death
4) have backs that glow under certain conditions
5) have fiber optic interfaces in their arms!

points 1 and 2 would almost certainly leave indicators in the muscle
cells, as well as brain waves. point 3 on brain waves, and point
4 certainly should be detectable. point 5 should be obvious!

At other points, they say that they are basically indistiguishable,
which to my mind means that they have organic processes (even if
artificially made), and even their brain activity is the same, to the
point of being indistinguishable from a person. If that is the case,
then it seems weird to talk about them having "software, not emotions".
It's a looks like a duck, acts like a duck, every possible measurement
confirms it's a duck, but it's not a duck? Sorry, that's just too hard
to swallow.

So you have the problem that, either, they are very different in which
case they should be detectable, or they are identical, in which case
they are no different than humans. The writers seem to want it
both ways.

What would my solution be? If these models were designed for the express
purpose of infiltratration into the fleet, then there would have been
a lot of work to make them undetectable. I would say that the innards
would be clearly mechanical/organic, with an organic outside. The inside
would broadcast signals to fool scans, like MRI and CAT scans, while
the outside would have blood, sweat, etc... to fool observation there.
It wouldn't be too hard to imagine that the only way to test for such
a thing would entail tests that would be lethal to a person, which
might solve some of the writers' problems.

Whenever someone said "they aren't human, they are machines" I cringed, and wanted
to reply: what do you mean by machines? In which ways are they machines and humans
aren't? throughout the series, the cylons were humans when convenient, and
machines when convenient. I don't think the writers had a consistent vision of
what the cylons were, or wanted.


3) WTF is the cylon plan. They want all humanity destroyed, but then
they seem to want humans for procreation and love. They say they want
to protect Sharon's baby, and then the next episode (!) they launch
an all-out attack on the fleet. They have well hidden spies in the
Galactica who could have done damage to the ship, that was going to
be a museum anyway, and they were unable to destroy it? How hard would
it have been to put one of those Cylon viruses in the computer, dormant,
and then when the fleet was assembled have it relay random jump points
to all of the fleet, and then send the Galactica into a star?

The writers seem to use Cylon attacks, and then Cylon pleasantness or
incompleteness, for convenience. It's back to the "not like us, just
like us" problem above which leads to inconsistent behavior. I think
there was no plan, so these inconsistencies keep compounding over time.


4) In flight of the Phoenix, when they disabled a huge number of cylon
raiders, why didn't they capture some of them for use later? Seems
like a major tactical failure.

5) In the 1st season episode when they were looking for Starbuck, who had
crashed on a planet with a cylon, Com. Adama said to his son that if it
were his son that was lost, that he'd never stop searching. That family
was most important. Why, then, in the last episode, is Com. Adama quick
to leave his son, knowing that Roslin was only going to be around for
a short while, and that both would now be alone.

6) In Pegasus, why is Baltar's angel surprised at seeing a beat-up
number 6? Why does she say "it's me", when it really isn't?

7) if ressurection ships are so important, why not guard them better?
why not have 2 ressurection hubs? why not a more distributed system?

8) how far away really is the Cylon homeworld?

9) in E16 2nd season, Sacrifice, we have people who claim the fleet are
cylon sympathizers and hold hostages. Not 2 episodes before, it was all
about cylon sympathizer groups trying to force gallactica to try to get
peace with the cylons. Each lasts 1 episode, and no more. so much
of this would have been better with a plan, built up over several
episodes.

10) at the beginning of each episode they have a 1-minute "what has
gone before", and then after the credits they have a 30-second flash
of images from the episode. it is as if they don't trust the extended
storyline structure, and they feel that a gimic is needed to entise
viewers to stay.

11) in Season 3 episode 8, Bulldog returns in a cylon raider after
being gone for 3 years. Doc Cottle does a quick DNA test to match it
against his military record, and concludes he's not a cylon. Wow! They
couldn't do that in Season 1? It certainly would have made a huge
difference. As plot holes go, that one is pretty darn big.

12) I find it mindboggling that a race that has mastered organic
technology cannot
a) do a proper quarrantine
b) diagnose a simple virus, that Doc Cottle can in a short while

13) why do the cylons want a new home? why Earth?

14) the incessant use of flashbacks I think is because there isn't a
plan, and this makes it look like there is a consistent backstory

15) the episodes with the coup, ending in the execution of Zarek and
Gaeta...some of the best that the show has had. I'd say, that the best
episodes have been:

Miniseries
Pegasus - Resurrection
New Caprica rescue
Coup (S4ep 13-14)

Notice that these are all the human stories. the cylon stories are not
nearly as good.

16) Episode 15, when Tyrol tells Adama that the ship is slowly breaking
up, and there is cylon tech that can help suddenly Adama is anti-cylon
tech? he changes his mind later, but still, this is inconsistent with
the previous episodes when he was immediately fine upgrading jump drives.
why the sudden change of heart?

17) when the BG makes it's final jump, Adama says: "whereever we are, that's where we're going to stay"

...of course, until the rest of the fleet gets there, and we continue with the
original plan. Why did Adama say this?

18) don't build a city at the end? don't use the technology? that
seems like an idea from an idealist but not very practical - a sure way
to reduce life expectancy in your new home


19) Adama is a cylon. Kara is the harbinger of death. Baltar and Six will be the mother
and father of the human race. prophecies that don't pan out.

20) from someone else, but exactly what I was thinking:
"""
When the season 1 [of Babylon 5] was aired JMS knew why Babylon 4 had
appeared and what
meant the visions saw there... when the first Shadow cruiser appeared
in season 1 he fully knew what it was and who was behind it.. when in
season 1 they talked about why Sinclair was so important and what
secret he had..

All was shorted out in advance.. he even writed "scape doors" in case
some actor stepped out of the series.. like happened with Michael
O'Haire (Sinclair) left at the end of season 1.. he used a scape
door introducing Sheridan..JMS has said he had scape doors for Delen,
Garibaldi, etc..

You think BSG writers knew there was going to be 2 earths?
Or who the remaining 8 models were going to be? Or even that Tigh
and Tyrol were Cylons?... you think they really knew
what the Opera house visions really meant? Or what the plan was?
"""

21) It seems to me that there would be a fool-proof way of killing a cylon
without it able to download: instantly drop it in the engine. unless the
download is able to be done in nanoseconds, there wouldn't be anything left
to download from.

Battlestar Galactica, Lost, and Babylon 5: To Plan or not to Plan

I'll get back to more academic things shortly, but I wanted to get this one off my chest. I just finished watching the entire series of Battlestar Galactica, which I did in little bits over the course of a year or so (I don't watch much TV, so I catch it when I exercise, and an occasional lunch time). I've always liked science fiction, was a big Trek fan, and followed Star Trek Next Generation from the beginning through all 7 years. I am a huge fan of Babylon 5, which I still believe to be the best Sci Fi ever to be on television (see full episodes here). Battlestar Galactica could have been that, but failed in one very particular way which I'll get to. Other people said that I'd like Lost, but I refused to watch it until it was done, so I could be told whether it failed as well, which I've been told it did in the same way as Battlestar Galactica (BG). How did it fail?

BG failed, not because of budget (which was probably 10 times that of Babylon 5 (B5)), or acting, cast changes, director problems, or writing (which is some of the best I've ever seen), but because they didn't have a plan. BG and Lost claimed they had a plan, but didn't. On the other hand, the creator of B5 shorted out all of the episodes for 5 years, so he knew where things were going from day 1. B5 is the only show that I know of that has done that, and it is (in my opinion) the only way to do long-running shows like these. Why does this make such a difference?

  1. Inconsistencies. When you have a plan, you don't have nearly as many inconsistencies. You're not trying to hack together a concluding episode, trying to make all of the lose ends tie together in some haphazard way, because you knew how they tied together at the beginning. Star Trek Next Generation solved this problem, mostly, by not having any long-running plot lines. Each episode ends with the characters in exactly the same condition as they started. BG had some good ideas, but its resolution seemed like a hack.

  2. Foreshadowing. When you have a plan, you can have foreshadowing. Not just vague, unexplained claims, but real references to future events. In the first episode of B5, a character refers to a prophetic dream they have about their death 20 years from then. At the time the viewer thinks they understand the dream, or think that it is a toss-off comment, but late 3rd season we see some of the details, and late 5th (and last) season we see the lead-up to those events. The flashback method is the poor-man's (or poor planners) way of doing this, basically saying "we need these events to have happened in the past, to make sense of what we want to show now, but we didn't think of it before so we'll show it now". BG used this a lot.

  3. Larger Story Arc. Even at its weakest points, B5 benefited from its plan. A weak single episode was lifted up if it contributed to the larger story arc. In this way, weak episodes were improved because there was a plan, and the entire series seemed more consistent.

  4. Rewatchability. From a marketing point of view, this seems to me to be a no-brainer, although it is almost never done. Once B5 finished, I wanted to go back and watch it again to pick up on all of those things that were planned, but I missed. How much foreshadowing did I miss? How many little details in the background were there that became important, but I didn't realize? How many decisions of the characters contributed to their final roles? How many seemingly throw-away lines were really important? Now that I've finished watching BG, I don't have any urge to watch it ever again. There's nothing new to see. Finding out in season 3 that Tigh is a Cylon doesn't modify your perspective on any of his actions in season 1. None of the fates of the characters can be seen in the early parts of the series. The so-called prophecies are each stated and resolved in a couple episodes, or so vague as to be meaningless.



I've heard that Lost suffers from this same problem: no plan leads to inconsistent storylines, convenient flashbacks, and an unsatisfying conclusion with loose ends.

Now, B5 wasn't perfect. It could have used more money, better actors, better dialog. It's writing can be corny at times, and there was a studio snafu that condensed some of the plot in Season 4, and made Season 5 a bit thinner than one would like. However, the universe is entirely original, and the 5-year plan was just amazing and makes the series hold up well over time.

I now won't watch much of anything without a plan...it's just not worth the ride.

Monday, June 7, 2010

Lord of the Rings

I have a backlog of blog posts that I wanted to get caught up on. This blog has been pretty academic lately, so I figured a change of pace would be nice. I figured I'd post some of my thoughts on the Lord of the Rings. To set the record straight, I really like the books and the movies, but I noticed something about the movies which disturbs me somewhat: in nearly every case where the movie deviates from the books, it is in the direction of weakening the characters. The more I thought about it, the more I realized that the pattern is held. Here I give a summary list of the changes to the characters. Perhaps there are more examples. Are there any counter examples?

















































CharacterComment ([M]=movie, [B]=book)
Gandalf

  1. hits his head in Bilbo's home [M]

  2. staff is broken by the Nazgul [M]

Frodo

  1. falls down, defenseless, when approached on Weathertop [M]

  2. strikes back bravely (although ineffectually) at Weathertop [B]

  3. ditches Sam in the middle of Mordor! [M]

Aragorn

  1. nearly gets killed by an orc, and needs to be revived [M]

  2. is not strong enough when looking in the Palantir, to counter Sauron (drops it) [M]

Sauron!

  1. information about the attack on Minas Tirith is leaked to Pippin's mind during the contact with the Palantir [M]

  2. Everyone knew where the next attack would be, and the attack is rushed due to Aragorn's confrontation in the Palantir [B]

Eowyn

  1. Eowyn is clearly afraid up to and including the battle [M]

Faramir

  1. Faramir is seduced by the ring, and takes Frodo to Osgiliath [M]

  2. Faramir is only momentarily seduced by the ring, but quickly comes to his senses and lets Frodo go [B]

Eomer + Minas Tirith

  1. Rohan and Minas Tirith are not enough to combat the first army of Sauron, and need the dead (called by Aragorn) to clear them out. [M]

  2. Only the back-up armies and navies of Sauron are cleared out by the dead called by Aragorn, and the Rohan is enough to fend off the armies at Minas Tirith itself. [B]

King Theoden

  1. Theoden needs Aragorn's prompting to help Minas Tirith. [M]




Any more?

Monday, April 19, 2010

Nice talk about science and vaccination

The site http://www.ted.com/ is a really great site to find talks on lots of different topics. The one I recently listened to is http://www.ted.com/talks/michael_specter_the_danger_of_science_denial.html, which highlights some of the dangers of the recent anti-vax movement. Lots of other great links at http://www.sciencebasedmedicine.org/?cat=36.

Friday, February 26, 2010

The Not-so-Hidden Flaw in this Climate Argument

There are sometimes people think that I am a global warming denier (I'm not), but I am not entirely convinced that the dire predictions from the global warming camp are supported by the evidence. I am skeptical of conclusions based entirely on models, and I am really skeptical of anything where I perceive less-than-open information exchange combined with vested interests and a lot of money. Given all of that, I feel free to rip apart anyone's argument regardless of which side they happen to be on. :) That's how science works!

So I was pointed to an article called "The Hidden Flaw in Greenhouse Theory" which has such flawed logic itself that I am surprised someone wrote it.

Introduction



To summarize the article, I quote the last paragraph:
An idea has been drummed into our heads for decades: that roughly 1% of the atmosphere's content is responsible for shifting the earth's surface temperature from inimical to benign. This conjecture has mistakenly focused on specifically light-absorbing gases, however, ignoring heat-absorbing gases altogether. Any heated atmospheric gas radiates infrared energy back toward the earth, meaning that the dreadful power we've attributed to light-absorbing molecules up to now has been wildly exaggerated and must be radically adjusted -- indeed, pared down perhaps a hundred times. Because all gases radiate the heat they acquire, trace-gas heating theory is an untenable concept, a long-held illusion we'd be wise to abandon.


How does he come to such a grand conclusion? He starts with a quote of a NASA elementary school guide, which has:

Question: Do all gases absorb heat?
• Answer: No. Only some gases have the unique property of being able to absorb heat.


Then he (correctly) criticizes this guide saying
So how does NASA go wrong? By consistently confusing light and heat, as you see in the illustration below, where infrared light is depicted as heat.


Clearly NASA should have said that greenhouse gases absorb infrared light, or radiation.

Then the author continues with this:
Why does NASA go wrong? Because it has a flimsy yet lucrative theory to foist on the taxpaying public, that's why. As the space agency explains in the Main Lesson Concept, the core idea of greenhouse theory is that downward radiation from greenhouse gases raises the earth's surface temperature higher than solar heating can.


This is amazing. Regardless of ones stance on global warming, the greenhouse effect (badly named) is a well known, well established consequence of the basic laws of thermodynamics. Without it, Earth would be far less habitable!

Conduction, Convection, Radiation, Oh My!




He continues:

To make this idea seem plausible, therefore, it's crucial to fix people's attention on the 1% of the atmosphere that can be heated by radiant transfer instead of the 99% and more that is heated by direct contact with the earth's surface and then by convection.


As a nitpick, it would actually be more like 5 percent or so, because water vapor is the most abundant greenhouse gas. Even so, his argument is fallacious, that a trace gas cannot cause significant warming. Let's take another example of a trace gas that increases atmospheric warming: ozone. In the stratosphere, at the "ozone layer", the concentration of ozone is about 2 to 8 parts per million. That is 0.0002%! Despite this trace amount, we get significant warming of the stratosphere as shown here.

His reasoning on this issue is
Consider too that since most air molecules are infrared-transparent, they can't be heated by the infrared that CO2 and water vapor emit. This means that downward radiation from "greenhouse gases" can only explain how the earth's surface might get warmer, not the rest of the atmosphere.


No one is suggesting that the non-greenhouse gases are being heated by the re-radiation of the greenhouse gasses. They are heated by the collisions with greenhouse gasses, thus raising the overall atmospheric temperature. This feedback loop continues until the new balance, with the surface, is established. It's strictly conservation of energy. Some is radiated, some is lost in collisions, and the gasses in the atmosphere as well as the surface adjust until an equilibrium temperature is reached. A simple process, really.

At the crux of his argument is this:

For meteorologists acknowledge that our atmosphere is principally heated by surface contact and convective circulation.


This is just wrong. What physicist do is follow the money, or the energy in this case. A very simple balance can be shown, quantified, and verified by numerous types of measurements. A very simple picture of it is here.


5ADD81CC-8005-4BA1-915D-5294F80A36C5.jpg



(From http://oceanworld.tamu.edu/resources/oceanography-book/radiationbalance.htm, which cites Houghton et al., (1996: 58), which using data from Kiehl and Trenberth (1996).)


As a fraction of the 342 units of solar energy coming in, about 168 is absorbed directly but about 324 is absorbed from the atmospheric radiation! The fraction of conduction and convection comes only to about 30%. It is true that the atmosphere gets some energy from conduction, but not most of its energy.


Scientific-sounding Junk



It is a real challenge for people without science background to sift through things like this. It sounds like science, but it is incredibly wrong. One way to help tell the difference is to observe how the author responds to criticism. Concerning the errors, do they offer an errata or do they ignore them, or worse, inject ad-hominem attacks on the critic? Saying that NASA "has a flimsy yet lucrative theory to foist on the taxpaying public" is playing into people's distrust of government, and trying to use that to bolster an argument, without actually strengthening the argument itself. It is implying motive without substantiation. I could do that too by suggesting that the author is starting with an anti-global warming position, and wants to prop it up with whatever emotional verbiage he can get away with to hide spurious arguments. Instead, I simply show how elementary and spurious his arguments are, and predict that he will not submit any corrections. I can point out the ad-hominem attacks on NASA as unbecoming of a scientific argument, and is useful only in a propaganda piece.


Another way that someone might tell if the argument is sound, is what sorts of errors are being raised. If the errors are of a technical nature, like the benefits of particular statistical tests, then it can be hard for a non-expert to tell. However, if the arguments have basic errors that can be confirmed in encyclopedias then it is much easier to tell. If the arguments include non-scientific, emotional, exaggerated language then it is much easier to tell that it is a propaganda piece and not a scientific piece.

Are there other ways that non-experts in a field can decide the merits of arguments? Perhaps someone will comment and add some more, and perhaps I'll think of some more. Mostly, being able to skeptically think like a scientist, regardless of your personal belief in something, is the most important skill to use in cases like this. I think it is an important thing to consider, in order to better win over the public to science.

Monday, February 22, 2010

A bit of must-read economics

In the face of Obama's new healthcare initiative, which includes provisions to cap the increases in insurance costs, you can read here a nice description of the problems of fixing maximum prices. When the Government caps prices to help people, it is a policy which goes in the category of "really simple, intuitive, well-intentioned, and wrong."

More people need to understand the proper functioning of the free market. This is true of people on "Main Street" as well as those on Wall Street. I think I may be becoming more Libertarian in my old age. :)