Climate models play a central role in the attribution of global warming or climate change to human causes. The standard argument takes the following form: “We can get the model to do X, using human causes, but not without them, so human causes must be the cause of X.” A little digging reveals that this is actually a circular argument, because the models are set up in such a way that human causes are the only way to get change.

The finding is that humans are the cause of global warming and climate change is actually the assumption going in. This is circular reasoning personified, namely conclude what you first assume.

This circularity can be clearly seen in what many consider the most authoritative scientific report on climate change going, although it is actually just the most popular alarmist report. We are talking about the Summary for Policymakers (SPM), of the latest assessment report (AR5), of the heavily politicized UN Intergovernmental Panel on Climate Change (IPCC). Their 29 page AR5 SPM is available here.

The “smoking gun” of circularity lies in just two figures, specifically Figures 5 and 6.

First look at Figure 5 on page 14. It is a bit technical, but as the caption says in the first line, these are “the main drivers of climate change” (according to the IPCC that is). There are just eleven of these main drivers, each with a colorful horizontal bar, although some are broken down into components.

The length of each bar represents the IPCC’s best guess at how much each driver has done since 1750 (and through 2011), which they call a “forcing.” So any observed climate changes over that 261 year period must be due to these drivers and this is just what the climate models assume.

In fact the modelers are required to assume it if they want to feed their results into the IPCC assessment. About 100 major climate models are used to do this modeling, in a huge coordinated effort that takes several years to pull off. That effort is coordinated by the US Department of Energy.

The smoking gun begins with the fact that 10 of these 11 drivers are from human causes. The only one that is natural is changes in solar energy input (called “irradiance”) and that bar is so small that it basically does not count. In fact you can barely see it in the figure.

So for all practical purposes, the IPCC assumes that all of the drivers of climate change are human caused. There are many other drivers that are discussed in the scientific literature but they are simply ignored, here and in the models.

Obviously, if all the drivers of climate change are human caused, then all the observed changes must also be caused by humans. And this is just what we see in Figure 6, on page 18.

This figure shows 18 separate little pictures. Each picture shows three things:

(1) An observed climate change.

(2) What the model output looks like without human forcing.

(3) What the model output looks like with human forcing.

Given that the only significant forcing allowed is human, it is no surprise that the model output only matches the observations when human forcing is included. How could it be otherwise? The 18 model outputs without human forcing do not change, indeed they cannot change, because no natural drivers are allowed.

But rather than seeing that this result simply follows from the initial assumptions, the IPCC amazingly claims that it proves that human are the sole cause of climate change. It proves nothing of the sort and this completes the smoking gun of circularity.

The simplicity of this fallacy is stunning. If you assume that only humans can cause climate change then of course it follows that all climate change is caused by humans. But this is just true by assumption, not by science. The reasoning is perfectly circular.

A legitimate assessment of the science would consider the many other drivers that are presently under consideration. Legitimate scientific modeling would explore these drivers, to see what contribution they may be making to the observed climate changes.

Unfortunately neither the IPCC nor the extensive, expensive modeling that is done to support it is legitimate. That the US Department of Energy is at the center of all this politicized non-science is especially bad and needs to change.

As it is, all of this modeling is just junk, because it assumes a false answer going in. There is no such thing as science by assumption. If the models are not going to do science then they should not be running.