Chapter 7 - Emergence and Science
Like a rat in a maze
the path before me lies.
And the pattern never alters
until the rat dies.
And the pattern never alters
until the rat dies.
Patterns, Simon and Garfunkle
Seventh grade science grabbed me and and I was sucked in by Ms. Helen Kicklighter. (She said we could remember her name as follows – “Don’t kick so hard, kick lighter.”) I didn’t notice it at the time, but she was truly a nerd. She had taught my two older brothers and seemed pretty old when I was in seventh grade. She had been in a car accident years before I was in her class and could not turn her head. That made for some amusing times when we’d try to get her to quickly look our way. Rather than turning her head, she had to move her entire body, shuffling her feet as she did. It seemed funny for a bunch of 12 year olds. Ms. Kicklighter got me to put together an exhibit for the Dade County Science Fair. She suggested I do something about the effect of smoking on grades. It was a pretty daunting research project for a 7th grader but she was there to make sure it got done. It was a very scientific project – I put together a questionnaire that I gave to all 7th and 8th graders. Somehow, I was able to get the names of people who indicated they smoked and Ms. Kicklighter was able to get copies of the smoker’s grades. (I’m pretty sure this was a violation of someone’s civil rights, but things were simpler in the 1970s.) I put together bar charts using plastic tape on poster board instead of PowerPoint, which was still many years from its invention. My final results showed that people who said they smokers got lower grades, on average, than people who did not smoke. I have no idea if the results were statistically valid as I didn’t know about things like sampling error. I had a conclusion section that postulated why people who smoked might have lower grades but the reality is I had no idea. People who smoked in 7th and 8th grade tended to be rebellious and rebels didn’t do things like homework or studying for tests. I had found a correlation and we’d like correlations to lead to a cause and effect explanation. With a cause and effect, we can then determine a pattern and once we have a pattern we can predict the future.
A Short History of Science
Science grew into the large enterprise we see today because of one simple fact – it does a better job of predicting the future than any other human invention. People talk about how they get into science to deepen their understanding of a particular subject (Just like people say they get into politics to better the world.) You don’t get too far in the scientific world if you don’t find patterns. Without patterns, the world is a scary and unknowable place. If there were no patterns you’d never go outside because you’d never know what would happen. Fortunately, the universe is built on a combination that allows for patterns and chaos, which breaks the pattern. At its heart, science is a methodology for looking at observational results, finding a pattern and using the pattern to predict the future. We saw in the earlier chapters that most interactions lead to patterns and those patterns allow the universe to have a stable base from which to evolve.
There are a lot of patterns in the world early on we noticed the pattern of daylight/night time and the seasons. It took a while for nerds to evolve, taking delight in observing the patterns more closely and starting to see things that others did not. The closest planets – Venus, Mars, Jupiter and Saturn – were discovered early on and a model of the solar system, with the earth at the center, evolved from there. Calendars were created as a byproduct of the seasonal patterns. Sundials were early devices that took advantage of daily patterns. The relationship between science and religion is documented well in other places so we’ll leave that part of the discussion aside. It took a while, but eventually science became a separate field of study. Galileo was not the first scientist but he is credited with promoting the cause more than any one before him. Early scientists used observations to look for patterns. Galileo performed experiments (like dropping balls of different weight to see if the fell the same or different speeds) and looked for patterns in the observations. Once a pattern is observed, a prediction can be made and more experiments can be performed to test the accuracy of the prediction. One of Galileo’s first experiments was with pendulums. He noticed and then measured how long it took a pendulum to swing back and forth. We performed experiments and determined that the length of time it took for a pendulum to complete one cycle (the period of the pendulum) was related only to the length of the pendulum. He put together a mathematical formula based on some experimental evidence and then used the formula to predict the results of new experiments. His predictions were spot on and today his discoveries are taught to all new physics students. Mathematical formulas were a new way of expressing patterns and were a great advance from physical models, like the initial models of the solar system. Modeling is the way science advances, so the introduction of a new way of modeling allowed science to advance even more quickly. Let’s look more closely at modeling.
Models are a simplified representation of the real world. For example, early models of the solar system had the Earth at the center (because everything fell towards Earth, it must be at the center) and the Sun, Moon, planets and stars orbiting the Earth attached to crystalline spheres. They were spheres because that was the perfect shape and it was a pretty close approximation to the real orbits. The spheres were crystalline so that the heavenly bodies didn’t fall into the Earth and you could see through the different levels. For many years this model served us humans well. There were some tweaks to be made to explain some anomalies in orbits of the planets but all-in-all things worked for hundreds of years. Ultimately, new and more rigorous observations lead to our current solar system model with the Sun at the center. It is more accurate than the spherical model and over time has supplanted it. Now a day, we can’t imagine thinking the Earth is at the center of the universe but when you stop and think about it, it seems to make sense.
How do these models come about? Scientists create them and the process they follow is amazingly simple (of course, we’ve learned that simple interactions can lead to complex behavior so don’t assume science is simple). They look at observations or make observation by conducting experiments and look for patterns. The scientific method evolved in such a way that scientists create a Null Hypothesis to predict what they think might happen and then conduct experiments to test the accuracy of the Null Hypothesis’ predictions. Accurate predictions elevate a hypothesis to the lofty state of theory. Those that are tested and found to predict a wide class of behavior are promoted to Theories. “Capital T” Theories are few and far between but the Theory of Gravity and the Theory of Relativity are two examples. As with any human endeavor, the easy problems were solved first. In the early days of science, it was a hobby for many people and a number of famous scientists were part timers. For the record, Albert Einstein was a patent clerk when he did some of his seminal work. Over time, the work got more specialized and people started pursuing science as a career. Teaching science and research became a way to make a living. Of course, when a large number of people are part of a profession, you can expect politics to follow. Before going into that, I’d like to talk a little more about science and the nature of truth.
Science and Truth
George Box was a statistician who made the famous statement, “All models are wrong, some are useful.” One of the first things a scientist will do is make assumptions that simplify the thing they are studying. Edward Lorene wanted to study the weather but it was too complicated so he developed an experiment using a hot plate sitting underneath a spinning Bundt pan. The hot plate was like the sun shining on the equator and the spinning Bundt pan was a 2 dimensional representation of the rotating Earth. Simplifying a system allows you to see patterns that may be hidden in the larger system. That simplification comes at a price. You give up accuracy in order to gain predictability. These days, the methodology for making observations can be so convoluted that even these observations, the facts of science, can be in dispute. I’ve read too many science discussions expounding the truth as defined by science and I want to correct that notion. There is no truth in science! By its very nature of modeling, it gets things wrong. We should not confuse usefulness with truth. Usefulness is pragmatic, it is good enough. In a confusing and chaotic world, we need to find a way to find our way forward; to get up every day and live in an unknowable world. We accomplish that by simplifying the world around us. Scientific studies show we process very little of the input that enter our senses. We throw away information at an amazing rate because we can only process a limited amount of it. So even we humans have to be aware of the limitations of our senses in understanding the world around us. We build models all of the time to help us navigate through our lives and we know we make assumptions that simplify the number of things we need to think about. For example, marriage is a human invention that allows two people to pledge they will be faithful to one another. (Among other things.) Trusting your spouse and having a marriage vow allows you to assume they will not cheat on you. You can go through your day not worrying if they are cheating on you and focus your attention on other things. All models are wrong and sometimes the assumptions you made (your spouse is faithful because they said they would be) turn out to be wrong. We then have to revise our assumptions, change our model and figure out how that affects the relationship to our spouse. We’ll talk more about this in the chapter on good and evil.
As science has moved into more complicated venues, like psychology and medicine, we see even more clearly how limited the scientific approach becomes. My 7th grade science fair project on smoking and grades did not show any causation, only a correlation. (I forgot to mention I got an honorable mention and a $25 savings bond from the South Florida Lung Association – my first indication that science could make you money.) This is a much weaker statement and isn’t really a prediction. I couldn’t say that if a student started smoking, their grades would go down. (That would be a prediction.) All I could say was that students who admitted they smoked got lower grades than students that didn’t say they smoked. I found a correlation, which is like a pattern, but not as good at predicting the future. That doesn’t prevent people from acting on these correlations, since in a highly complex world, it seems to be the best we can do. Let’s explore how correlations came about and how they work.
Correlations and Causation
Almost every day, a new study is published that discusses a correlation between foods eaten and rates of certain cancers. How did we get to this? Remember in Chapter 5 when I wrote about the study on extra-marital affairs? – “Forget the statistics, I want names and addresses?” The simplest explanation for how we got here is evolution – specifically, scientific evolution. (Remember, we need to qualify the word evolution.) As science evolved as a way to find patterns, it started with the easy problems. Galileo found the pattern of pendulums using string, weights and his pulse. The next hundred plus years were dominated mostly by amateurs, with professionals sprinkled in here and there. Scientists went after problems that were interesting to them or interesting to their bosses and since the tools they had to work with were primitive, they were limited to simple problems. Scientific evolution took over as people tried different approaches to solving more complicated problems and in the 1800s, Ludwig Boltzmann made a creative leap in solving problems involving large numbers of molecules and atoms. He realized you could treat them like identical billiard balls and use statistics to combine their individual motion into group motion. The results were nothing short of amazing. A box 1 foot on a side has an insanely large number of molecules in it, too many to track individually. (Remember Avogadro’s number – 6.022 x 1023? That is a 6 with 23 zeroes after it and is pretty close to how molecules are in the box.) The nature of the evolutionary process is to take an emergent property and spread it throughout the environment. In this case, the emergent property was statistical analysis and scientific evolution took it to other fields of science where it was tried out. Nuclear physics was able to use it with great effect, since elementary particles are indistinguishable. As a methodology for dealing with large numbers of things, it worked well and at some point in the 21st century, it was applied to biological and biochemical problems. Why? Mostly because the problems were so complex that no other way to approach a solution was available and a partial solution seemed better than no solution at all. Looking for a pattern with a large number of human actors is basically a crap shoot. While as far as we know all elementary particles, atoms and molecules are identical, we know that isn’t true of animals and humans. The assumption that the actors are identical is violated so any conclusions derived from these assumptions will be less valid. All models are wrong and if your basic assumption is wrong, you’re bound to be limited in how useful the results will be. Medical studies are trying to move into a world where they consider the individual differences; DNA testing and genetic sequencing are the first steps. We’ll need another emergent property (an A-ha! moment) to take us to the next level of pattern recognition in human biology. Psychology is even further away and there’s more than one reason they call economics the dismal science.
Now there are some places where the assumption that humans are all alike is more valid – like in the e-commerce field – and we see those models are more useful. It is especially important that you understand the assumptions that are made at the beginning of any scientific modeling exercise. Sometimes, the assumptions are so off-base that they have more in common with make-believe than science. As long as there is emergence, there will be places where the old models fail – like in the economic models of the 2000s. The effect of emergence is so large, why isn’t there more science published on emergence? Where’s the science of emergence?
How do forces emerge?
Just to be clear. No one has any good ideas how new forces emerge. The history of science has been based on the assumption that all forces can be derived from the three quantum forces and gravity. I’d like to talk about some of the conditions needed for a force to emerge and some details of how a new force might emerge. Chaotic behavior is needed for a property to emerge. Nothing will ever emerge from the classical world and their attractors. A necessary condition for emergence is strange attractors. But chaotic behavior isn’t enough. There needs to be a high level of interaction in order under some constraints to obtain emergence. In fact, I’ve come to look at emergence as developing from chaotic behavior under constraint. Think of an explosion as an uncontrolled event where the energy literally blows things apart. Emergence has the level of intensity of an explosion, but instead of flying apart, it is focused inwardly and there is a transition to a new level of interaction. For example, a nuclear explosion uses the power of the strong nuclear force to generate a lot of energy that spreads out quickly. The sun uses the same strong nuclear force but, under the constraint of gravity, gradually produces energy that is released much, much more slowly. This controlled energy produces new, heavier elements from the basic building blocks of electrons, protons and neutrons. These elements are then spread out throughout the galaxy when the sun finally explodes, only to be brought together by gravity into planets where they can be used to create ever more complex properties – like life.
One physical characteristic that seems to be related to emergence is phase transitions. We all know that water comes in three phases – vapor, liquid and solid (ice). Scientists have spent a lot of time studying the process where water vapor liquefies and water freezes. It is a most complicated scenario and one that even today is not fully understood. (So it isn’t farfetched to think that since we don’t understand phase transitions that we don’t understand emergence.) The transition from solid to liquid has some of the properties of emergence, in that liquid water has different ways to interact than ice, but there is more to emergence.
We really should discuss entropy at this point, because it is related to emergence.
Entropy is a measure of organization. Larger values of entropy mean a system is more random (less organized). So your car is a highly organized piece of equipment. Someone had to put a lot of energy to build your car. A car will not organize itself out the parts all by itself. But it is how you apply the energy that makes a car “emerge” out of the individual parts. If you just took an explosive, put it into the pile of parts and lit the fuse you would certainly put energy into the system, but you would blow it apart. That same energy, applied in a directed way, can create an automobile. The car has less entropy than the parts that went into it.
Left to its own devices, your car’s entropy increases and the level of organization decreases. Eventually, you are left with a rusted pile of junk where the car used to be. Things fall apart unless you do something (which requires energy) to maintain and organize them. Even worse, the universe is put together in such a way that entropy always increases. The only way you can decrease entropy is to increase it somewhere else. We see that on Earth. The Sun is increasing its entropy by burning as a “controlled” nuclear reactor. The heat from this reaction warms the Earth and allowed life to emerge (decreasing entropy). Our gain is the Sun’s loss and it is always the case, there are winners (entropy decreases) and losers (entropy increases). In fact, if you hearken back to our discussion in Chapter 1 on hierarchy you’ll notice that as the level or organization increases in the universe, the total amount of organized material decreases. There are more quantum particles than solid matter. There is more inert matter than living matter. There is more un-intelligent life on the Earth than intelligent life. As you organize things more and more, you must leave behind more and more un-organized detritus to balance out the scale. So there is a limit to how organized physical systems can become and I suspect we’ve reached that level on Earth. The emergent process appears to have reached a physical limit in the origin of human intelligence. It isn’t because the physical part of the planet earth has stopped the emergent process, but the emergence of intelligence endowed us with the ability to create things with ideas much more quickly and free us from slowness of physical emergence.
In addition to physical entropy, entropy can also correspond to information. Once intelligent creatures created the concept of information, we were freed from physical constraints. While there is only so much physical mass in the universe, information has no limits to the things that can emerge. Information and it associated properties of language (spoken and written) free us from physical limitations. Information still follows the U-ROC, things change by interacting. Information is just something that forms that basis of interaction between intelligent objects. Technology allows us to increase the speed of information interaction (we can talk on phones where the information travels at the speed of light rather than the speed of sound) and the range of interaction (we can talk to anyone in the world now and are not limited to just talking to people near us). It is worth noting that while technology can increase the speed and range of information interactions, it has no effect on the quality of the interaction. I think this is a basic limitation of technology. Technology can improve productivity but it does not necessarily improve the quality of life. Remember when people said we’d eventually have a 20 hour work week because technological advances would make us more productive and we’d have more leisure time? That didn’t happen because in a capitalistic society, you need to maximize profits so the increased productivity allowed us to produce more in the 40 hour work week, not work 20 hours. Economic evolution trumps technological evolution sp the work week remains at 40 hours, we just get more done. It is effects like this that make us question the definition of progress.
