Andrew On Everything

One day, when my daughter was seven years old, we were looking at some old pictures and there was a picture with “Santa” visiting some gathering of children. I said “Oh, there’s Santa”. She said “No, that’s just Bob (a family friend) dressed up as Santa”. I said “How do you know?”.

She went on to use some logical arguments, based on how the “real Santa” acts, to show that the person in the picture could not have been the real Santa.

What interested me was the fact that, even though she got the conclusion right (i.e. that person was not Santa, it was the family friend my daughter suspected), and even though her arguments were logical, her arguments were based on an erroneous assumption, namely, that Santa Claus exists and that he behaves in a certain way.

She could not even fathom (at that age), that the central assumption in her arguments was false. Which made me see how futile her efforts at logical reasoning based on this false assumption were.

This in turn made me think how futile our efforts most probably are when we try to reason about the nature of existence/reality. This is because many of our assumptions (about what it means to exist, what it means for something to be “outside” our universe, etc) are most probably wrong.

It seems futile to logically debate about the ultimate nature of reality, when, no matter how logically correct our arguments might be, they are likely based on some erroneous assumptions, and we can never know which assumptions are correct or not.

To put it simply, to make a logical argument, you need some assumptions and then you need to follow the rules of logic to come to some conclusion. Sometimes logic can be used to show that the assumptions are false (e.g if they are logically inconsistent), but most “well constructed” assumptions are not logically inconsistent, and therefore cannot be shown to be false using logic.

In that case, you are arguing correctly, but your conclusions are meaningless.

My experience that day made me think that maybe all of our (as humanity) metaphysical discussions and resulting logical reasonings are meaningless.

When I hear people mention that the electoral red/blue state map of the US looks mainly red (which is even more pronounced when you look at the red/blue divide at the county level), and which I’m reminded of when reading Paul Ryan’s statement about Obama winning due high urban turnout, the thing that comes to mind is simply: “Land can’t vote. Get used to it.”

No matter how much land some people occupy, they still get one vote each. It’s really not a mystery.

Originally posted November 13, 2012

While looking at the medal rankings of the London 2012 Olympics, the US is the clear winner, with 104 total medals over the second country’s 88 medals, and 46 Gold medals of the second country’s 38 Gold medals. China, Great Britain, and Russia also make a strong showing.

Below are the top 10 countries:

However, since the US is around 300 million people, and China is even larger (1.3 billion), to make the comparison a bit more fair to the many smaller countries that make up the European Union, I added their medal count and created a new table, shown below.

We can see that the European Union totally crushes the other entrants. For a more in-depth look, I wanted to see what effect population size and GDP had on the medal counts, so for the top 10 countries in the above table I calculated a bit more info, based on population and GDP data from Wikipedia.

The above table shows that in terms of medals per 100 million people, most of the above countries are similar, hovering between 30 and 60 medals per 100 million people. The two outliers are China and Australia, with 7 and 152 medals per 100 million people, respectively. Note that the EU, Russia, South Korea have about twice the number of medals per 100 million people as the US does.

In terms of medals per 100 million people per USD 1 trillion GDP, there is a large variation, from 1 for China to 382 for Kazakhstan. Among countries with a large GDP (above 4 trillion), Japan does best, with 7 medals, followed by the EU with 4 medals, followed by the US with 2 medals, and finally followed by China by 1 medal, all per 100 million people per USD 1 trillion GDP.

From the above it looks like GDP is not a great predictor of medal count, but population size is a somewhat decent predictor. One big caveat is that the above calculations were done only for the top 10 countries in table 2, so one would need to extend these calculations to all countries in the Olympics to come to more definite conclusions.

Originally posted August 24, 2012

Originally posted June 26, 2012

It has been evident since 2000, after the Gore-Bush decision debacle, that the US Supreme Court is less a bastion of Constitutional Truth, and more a motley crew of partisans & ideologues, who were selected based on accidents of history regarding which justice died during whose administration, and who can and will put their ideology/party before any quest for Constitutional Truth.

Since then, with every 5-4 decision along ideological/party lines, the above impression of the Supreme Court only gets stronger.

Essentially, the US Supreme Court is useless. They decide most cases along conservative/liberal lines and justify already-arrived-upon opinions with complex legal terminology. As a result, they add no real value to questions of constitutionality. A Supreme Court decision is less a revelation of what the constitutionally-correct interpretation of an issue is, and more a confirmation of the decision everyone knew they would arrive at before they announced it, based simply on the party/ideology of the justices.

In effect, the laws/decisions that get upheld simply reflect the conservative/liberal mix of the Supreme Court at the time, and have little, if any, to do with any legal “truth” or constitutionality of the issue under consideration.

To be more accurate, for the truly divisive cases, it’s already known what the four judges on the left and the four judges on the right will decide, and there remains only one source of uncertainty: which way the ‘swing vote’ justice (currently Anthony Kennedy) will decide. This situation is making a travesty of democracy, since so many major decisions that affect the lives of 300 million people are de facto decided by only one person, the current swing vote justice.

I wanted to quantify the decision making process of the SC, to see if we can gather some statistics that back or refute our assumptions about the SC. Using data for all the SC decisions in 2011, we build the following table which shows the percent of the times each justice justice agrees with each other justice, on decisions with 6 or fewer concurring justices. In each row we see what percent of the time the various justices agreed with the justice shown at the beginning of that row.

Percent agreement between justices on decisions with 6 or fewer concurring

Some observations:

• From row 1, we see that Scalia and Thomas agree with Roberts 94% of the time in such cases, while Ginsburg agrees 12% of the time on such cases

• From row 2 we see that Scalia and Thomas agree 100% (!) with each other on such cases. You could replace Scalia on the bench with a cardboard cutout, and just always double Thomas’ vote, and there would be absolutely no difference in judicial outcomes of the Supreme Court in these cases.

• From row 4 we see that Alito and Ginsburg are diametrically opposed, since they agree on 0% (zero!) of such cases. You could replace both of them on the bench with cardboard cutouts, and just always add 1 vote for each side of an issue, and there would be absolutely no difference in judicial outcomes of the Supreme Court in these cases.

• Comparing rows 1-4 (conservative justices) versus rows 6-9 (liberal justices), we see that the right-leaning justices are quite monolithic, since they agree with each other between 88% and 100%, while the left-leaning justices are less monolithic, since their agreement varies between 56% and 81% of the time. This seems to show that the left-leaning justices don’t as easily fall prey to groupthink as the right-leaning justices do.

• As expected, Kennedy is the swing vote, agreeing 50% of the time with Ginsburg and 50% of the time with Alito, the two diametrically opposed justices. His agreement with the other justices ranges from 44% with Breyer to 62% with Roberts. So Kennedy is quite balanced, with a slight bias towards right-leaning justices.

Another way to analyze the 2011 Supreme Court data is to look at the percent of the time that a justice is in the majority of 5-4 decisions, which is what the next table shows

Percent presence in the majority in 5-4 decisions

Observations:

• We see that Kennedy is in the majority in most 5-4 decisions (75%), cementing his swing vote status
• He is followed by the four right-leaning justices (Roberts, Scalia, Thomas, Alito) all with 62% of being in the majority of 5-4 decisions. This is a remarkable consistency and is very illustrative of the monolithic voting and groupthink on the part of the right-leaning justices.
• The left-leaning justices are less successful, being in the majority of 5-4 decisions between 38% and 50% of the time.

Overall, the data from 2011 quantifies what we had already suspected, namely that SC justices vote mainly based on their ideology/party, and not based on the legality or constitutionality of the issue at hand. The fact that they are so monolithic in their voting (especially the right-leaning justices) proves this. Their record is there for all to see, and it is quite damning.

Apparently, I’m not the only one who is skeptical of the SC and the way in which they arrive at decisions. According to an article in the New York Times, “many Americans do not seem to expect the court to decide the case solely along constitutional lines. Just one in eight Americans said the justices decided cases based only on legal analysis”. Not surprisingly, the article also states that according to a new poll, just 44 percent of Americans approve of the job the Supreme Court is doing, which is a historical low.

It’s not clear what the best solution to this problem might be.

Referendums for the most contentious cases are a no-go for many reasons, but they would at least reflect the conservative/liberal mix of the population at the time of the issue at hand, and not the conservative/liberal mix of a court that was determined years ago, based on accidents of history regarding which party had the White House when past justices died or retired.

Perhaps more realistically, maybe we should have a fixed duration of tenure for Supreme Court justices. This way, they rotate out more frequently, and better reflect the political reality of the day. Having them be life-term judges was supposed to keep them “above the fray” and above politics, but in practice, they have turned out to be as much in the fray as everybody else. So, getting rid of the life term for Supreme Court justices should not have any practical effect on how non-partisan the decisions will be.

An alternate solution is to keep things as they are, but force decisions to be least 6-3. If something is 5-4, the result should be, legally, as if the Supreme Court refused to hear the case, i.e. the decision of the lower court is upheld. This new rule may get the justices to be less partisan and compromise more in order to reach the 6 votes needed. But, even if it doesn’t do that, it will at least remove the travesty that is the one-man-swing-vote.

Originally posted June 22, 2012

Original responses:

• HA responded: A couple of comments:
  - In your comment about Scalia and Thomas, you had it backwards. There should be a cardboard cutout of Thomas and a doubling of Scalia’s vote. Thomas has never broken with Scalia’s vote. Moreso, Feb. 2011 marked a 5-year anniversary of Thomas’ silence from the bench— the man had not made a comment or asked a question from the bench in 5 years. If he’s uttered anything since I haven’t heard about it.
  - Hey, wouldn’t it be funny if Roberts read your post before the SC decision June 28? :-)

• Pat Powers responded: Chief Justice Roberts’ decision on the Constitutionality of the Affordable Care Act, while statistically insignificant, might just be an indicator that the conservative monolith is breaking up. Also, I don’t think it’s fair to describe the liberal justices as “monolithic” when their rate of agreement is only 50 to 80 percent, compared to the conservative rate of agreement of 88 to 100 percent. When your rate of agreement is only 50 percent you are WAAAAY away from being “monolithic.” They are not just less monolithic, they are NOT monolithic.

Imagine some aliens for whom our solar system is an atom in their universe. Also, suppose their physicists are studying the Earth like our physicists study subatomic particles in our universe.

One can imagine them discovering some mass flowing/moving on the surface of this Earth, and it turns out that this moving mass corresponds to all the Earth’s animals and humans. Since we are too small for them to individually count us, to their instruments we just appear like a flowing mass moving around the surface of the Earth. As a result, they develop theories that model this movement or flow.

I assume that they can find some periodic events, like the great migration of the Wildebeest in Africa:

This means that their equations will have some periodic components. However, the equations have to be a bit more complex than that, since the timing of these annual migrations isn’t totally predictable: “The timing of their migration in both the rainy and dry seasons can vary considerably (by months) from year to year.” (Wikipedia)

What’s more interesting though, is what happens when their equations try to predict the flow of humans across the planet. Most of the population is located in cities, but there is a lot of movement on a somewhat periodic basis. For example, some events cause the movement of tens or hundreds of thousands of people:

• With a period of 4 years, we have the Olympics and the Soccer World Cup
• With a period of 1 year, we have the Super Bowl in the US and Soccer cup finals in Europe
• With a period of 1 week, we have football and soccer games in cities around the world

These are somewhat predictable, and could be incorporated quite easily into the model, except for the fact that they are sometimes pre-empted by other events. For example, the 1940 and 1944 Olympics were canceled due to World War II. How would the model account for that? Or how would it account for the fact that on some weekends, games are canceled due to strikes or other phenomena?

And how would it account for the fact that once in a while, wars break out between humans, and thousands and sometimes millions of people move around the surface of the Earth?

I assume that they would take care of these unpredictable-to-them anomalies using something similar to what our physicists use: the wave function.

“Although ψ is a complex number, |ψ|2 is real, and corresponds to the probability density of finding a particle in a given place at a given time, if the particle’s position is measured.”

That is, we can’t tell where a subatomic particle is with certainty, so we have a probability density function for its location. Similarly, the aliens wouldn’t be able to predict the exact location of “human mass” with certainty, so they would have a probability density function for its expected location.

However, remember that if the aliens could hear and understand our newscasts and followed the goings-on in the human world, they might know that a baseball strike means no flow of humans to baseball stadiums this Sunday. Or they might know that the assassination of an ambassador means war between two rival countries resulting in a high flow of humans in the next few days or weeks.

Since they don’t know all the underlying reasons behind the events, they are stuck with the probabilistic model. Could something similar be happening with the probabilistic models we use to describe our subatomic world?

I’m not claiming that subatomic particles in our world contain sentient beings, and if only we could understand these beings, we could accurately predict their behavior without probabilistic models. What I am claiming, and what the analogy in this post is trying to show, is that it’s possible that there may be some information about the subatomic world which is unobtainable to us, and which, if we knew it we could predict certain events without the need for probabilistic models such as the wave function.

This sort of idea has been proposed before, and it’s referred to as the hidden variable theory. It seems to have been mostly debunked: “Bell’s theorem would suggest that local hidden variables are impossible”.

However, whenever I see references to quantum physics, the wave function, and probabilistic models of the locations of subatomic particles, I still get a kick out of imagining aliens trying to tweak their model to account for the latest royal wedding in England or the latest Justin Bieber concert.

Originally posted June 21, 2012