Why the Warmies are Wrong

Dave Dyer

When trying to explain complex phenomena such as the economy, the weather, or the behavior of a child, it is common to look for a single factor that causes and explains the phenomenon. For example, Harry Dent looks at the percentage of specific age groups in the population to explain the economy, Warmies look at the CO2 in the atmosphere to explain changes in the climate, and parents blame violent video games for little Johnny’s bad behavior.

These attempts are almost always simplistic, misleading, and wrong. Why?

Think about complex systems like the economy, the weather, or children. They all have multiple independent variables that influence the value of any specific measurable dependent variable. A specific measurable dependent variable might be the rate of change in GDP for the past quarter, the average global temperature at a specific time, or the number of times that little Johnny has been kicked out of school in the past year. Independent variables would be any factors that might have a positive association with the dependent variable. Knowing the value of the independent variable, and how it influences the dependent variable, allows the dependent variable to be predicted.

The problem is that complex systems contain multiple independent variables, even if only one dependent variable is under consideration. Little Johnny is influenced by many things (food, friends, heredity, newspaper stories, etc.) in addition the violent video games. A bulge in a segment of the population may influence the economy, but so will new technologies, incompetent politicians, and wars. That is why these systems are called complex.

It gets worse when you understand that a specific independent variable may influence other independent variables, in addition to influencing the dependent variable. For example, if there are ten independent variables, a change in the value of one of them may also change the values of some of the other nine. And in turn, as each of these others change, they may also change the values of additional independent variables. Some of these changes may be simple and linear, but others may be non-linear and less understandable. A feedback loop from the dependent variable may also change the independent variables, adding another level of complication. So, a change in the value of one variable could change an entire matrix of values making it hard to understand how that one independent variable influences the single dependent variable under consideration.

This is the basic problem of multi-variable process control in a continuous process like a refinery. In that world, models can be built, tested, and tweaked until they work properly. But that is a world of a closed loop process, one where all the inputs are known in advance and no new ones can be introduced without notice. In fact, refineries are designed to be controlled and optimized for certain specific dependent variables. Sensors are embedded in the process equipment to record accurate readings. It is designed by humans to be predictable. Natural systems, like the economy, the weather, or children are the product of natural selection in an environment of many influences. They are fundamentally different in an important way.

Natural systems do not allow themselves to be easily predicted from a single variable. They evolved in a multi-variable environment over time and their nature is determined by these many influences, unlike the refinery that was designed at a single moment in history and intended to be predictable. Since the natural system is the product of many influences over time, not the product of human intent, it is reasonable to expect those many influences to continue their influence into the future even after the system has reached its current state. The same forces that built the system will continue to influence it. That is not the case with something that was designed by human intent at one particular time and then left alone to run.

How does this relate to the claim that man is responsible for causing global warming, mainly by increasing the amount of CO2 in the atmosphere by industrial activity? How is it possible with any degree of certainty to pinpoint this single independent variable as the primary influence over a single dependent variable, the Earth’s temperature? Let’s look at what you would need to know in order to support this claim:

1. You would need to know all the other potential independent variables that could affect climate change besides CO2.
2. You would need to know the complex function that maps changes in the CO2 values to changes in the dependent variable.
3. You would need to know how all the other potential independent variables are influenced by changes in CO2.
4. You would need to know if there is a feedback loop from changes in the Earth’s temperature back to changes in the CO2 level.
5. You would need accurate measurements, both current and historic, of all independent and dependent variables.

This presents a daunting task for someone trying to prove that man-made CO2 is the cause of global warming. The answer, normally, is to build a computer model. This has been done multiple times, but, unlike the refinery model, this model can’t be tested by experimentation in the real world. There are lots of models for the climate, but you can’t shut down the Earth and re-run it to test any of them. Testing with historical data is problematic. How could you possibly know with any certainty what the temperature was at a specific location 400,000 years ago? Testing a model based on its predictions for the future is not very practical when the predictions are about events that will happen hundreds or thousands of years from now. An untested model is just an opinion.

How many other potential independent variables might influence the Earth’s temperature? Here is a short list:

1. Changes in solar activity
2. Continental drift
3. Changes in the Earth’s magnetic poles (they do reverse every few hundred thousand years)
4. Changes in the jet stream
5. Volcanic eruptions
6. Changes in ocean currents due to new undersea mountains
7. Erosion and the landscape changes caused by erosion
8. Deforestation by some dominate species
9. Cooling of the Earth’s molten core
10. Smoke from forest fires
11. Cosmic events like an asteroid strike
12. Changes in the tilt of the Earth’s axis
13. Changes in the speed of the Earth’s rotation

Look at the list above. Is it complete? How can you be sure? Even if it is complete, can you tell with any accuracy how each of these might influence, or be influenced by, CO2? That would be a very complex formula, and you would still need accurate empirical data on the values of each of these potential independent variables, not just currently but over time. I am not saying that the Warmies can’t come up with the right formula; I am saying that even if they did, they could not prove that it is the right one. Since the Warmies are claiming a level of certainty that requires fundamental changes to our economic activities in order to avoid an approaching disaster, this is a major problem for them.

The Earth’s climate is a complex, unstable, open-ended process that is subject to the influence of many independent variables. The formula needed to determine the influence of any single variable is probably unknowable and it is certainly untestable. Even if you did have the right formula, the accurate empirical data needed to validate it are not going to be available.

The fundamental mistake that the Warmies make is to look for a single cause in a natural system that evolved under the influence of multiple causes. So, whatever caused the climate to change in the past will probably also cause it to change in the future. There was no single cause for the current state of the climate and there will be no single cause for its future states.

What caused the climate to change before humans were around? Probably lots of things. Will those same things cause the climate to change in the future? Sure. And humans will respond by doing what humans do best…they will adapt to the changes.