The history of governments meddling in the practice of science is not a good one. The most infamous case is that of Lysenkoism -Stalin backed the ideas of Trofim Denisovich Lysenko who believed in the inheritance of acquired characteristics. His ideas became of the official sanctioned science of the Soviet government. Genetics was declared a “bourgeois science,” or “fascist science,” and many geneticists who disagreed with Lysenko were executed or sent to labor camps. Execution tends to have a chilling effect on the free exchange of ideas and the practice of science. Over seven decades later genetic science in Russia is still lagging behind.
In the US we have a similar problem – not the Gulag, but political factions that disagree with certain findings of science that are ideologically inconvenient for them. The two biggest issues being targeted (but certainly not the only ones) are evolution and climate change. Much of the focus has been on what should be taught to students in science class (my vote is for science).
Recently the North Carolina legislature proposed House Bill 819 to study the effect of climate change on sea levels, and therefore coastlines. For some reason the legislators felt the need to include in the bill specific restrictions on how the science can be done. Section 2 includes this line:
These rates shall only be determined using historical data, and these data shall be limited to the time period following the year 1900. Rates of sea-level rise may be extrapolated linearly to estimate future rates of rise but shall not include scenarios of accelerated rates of sea-level rise.
Apparently being elected to the State House of North Carolina invests politicians with superior knowledge (to actual scientists who work in the field) as to the proper way to model future trends. Extrapolating current trends into the future is always tricky. The very heart of any such extrapolation is to know what the mathematical trend is. There is a tendency to assume a linear extrapolation, but that is usually not the case.
The problem with linear extrapolation of trends is that often we are looking at a small part of the slope of a complex and changing phenomenon, and there is no justification for assuming that current short term trends will hold up in the long term. Usually this takes the form of, “If current trends continue (linearly is often assumed), then in X amount of time something horrible will happen.” However, we may just be looking at a the ups and downs of a cycle, which is only apparent when longer term data is looked at.
Further, there are often game changers that alter the factors involved in creating the trend. Systems reach saturation points, compensatory mechanisms take hold, technology advances, and times change.
On the other hand, some systems have feedback loops that can actually increase the rate of change over time, so the trend is not linear but geometric or exponential. If in 1960 you tried to predict what computer technology would look like in 2012 by extrapolating linearly from advances so far, your estimates would be off by many orders of magnitude. Advances in computer technology have been geometric, doubling roughly every two years, or 18 months if you believe Intel (a phenomenon known as Moore’s law, which specifically referred to the density of transistors on computer chips).
What is the proper way to mathematically model future changes in sea level based upon recent climate data? That is a highly technical scientific question best left to those with expertise in this area. Climate has many feedback loops and game-changing consequences that need to be taken into account. How this all works out is a legitimate matter of scientific debate, and there are many viable models. Here is a good discussion of the issue, specifically relating to the recent IPCC report on sea level change. They discuss a number of different scenarios, producing predictions of sea level rise by the end of the century between 18 and 59 cm. The extrapolations are all fairly conservative, in that they in fact rely on some linear extrapolations of some things like the flow of ice from Greenland. They acknowledge that this can dramatically change, and may even slow, or the glaciers may rapidly slide off into the North Atlantic, changing all predictions of sea level rise.
The North Carolina bill does more than dictate how publicly funded scientists may extrapolate future trends, they also dictate that they may only use historical data. In other words, they cannot use climate models – they can’t use any actual scientific understanding of climate and the forces at work that can affect climate. They must take a simplistic approach of extending the line from historical data into the future, and they must do it in a linear fashion.
This bill is an amazing combination of ignorance and hubris. It is perhaps an excellent example of the Dunning Kruger effect whereby ignorance renders one unable to detect their own ignorance. Only a scientifically illiterate legislator would be unable to determine how scientifically illiterate this bill is.