I believe in Global Warming (and other conversion stories)
Man-made global warming predicts the past with amazing accuracy. Its record at predicting the future is less impressive. Yesterday’s weather always seems to confirm global warming but it’s never predictive: when there’s a big hurricane, it was global warming charging up the weather, and the next year is likely to be worse. When the next year ends up with an extremely weak hurricane season, it’s global warming causing the ocean to calm down, and we’re going to have mild winters. When the next winter is extremely cold and snowy, it’s global warming again. When it’s warm it’s proof of global warming; when it’s dry it’s proof of global warming, when it’s wet it’s proof of global warming, and when it’s cold it’s either proof of global warming or weather isn’t climate.
In the last several years, we’ve seen that warm weather in Texas is the result of global warming; cold weather in England is the result of global warming; a powerful hurricane is the result of global warming, and a weak hurricane year is also the result of global warming.
It never seems to matter how the predictions come out, the result is always proof of man-made global warming.
One of the recent examples of that is a statistic about the number of record highs in the last several months. You can always search the statistics of the past for confirmation bias: there are millions of statistics, and you can throw out the ones that don’t help you. What matters is how well the theory you base on those statistics can make predictions. That’s what science is.
I saw those stats on Facebook, and commented that “the real test will be what predictions are made from this data.”
My Facebook friend’s response was not a scientific prediction made from a testable theory. It was an article titled A Message From a Republican Meteorologist on Climate Change. Paul Douglas’s article starts out like it’s addressing an AA meeting, and it talks completely about generalities, isolated statistics, and blanket predictions. It predicts everything so that it can never be wrong1, but predicts nothing measurable, so that it can never be disproved. It disparages proponents of applying science to the theory as enemies of the Truth.
I grew up Catholic, and I recognize Paul’s story: it’s a conversion story. Paul Douglas was driving out of Damascus, Minnesota and Mother Nature hit him on the head and said “be my steward”. This “Republican Meteorologist” didn’t look at the predictions of anthropogenic (human-caused) global warming as a theory, see that they predicted a mild winter in 2008/2009 and that the prediction held up. He couldn’t; it was the coldest in a decade. The next winter’s predictions were even further off the mark. He didn’t say that he looked at the predictions of increased hurricane activity in 2007 and that the prediction held up; 2007 was one of the mildest hurricane years in the past hundred years. He didn’t say that he looked up the predictions of snow loss in the Himalayas, or of any of the other disproven predictions of anthropogenic global warming.
If human-caused global warming were a scientific theory, the people who made those predictions would have moved on to something else; their theories were disproven.
The test for any scientific theory is, how can it be proven false? A theory that can’t be proven false no matter what happens is religion, not science. For example, these twelve months are not proof: they weren’t predicted. But suppose sometime in the next ten years there were 12 months with similarly-record-making lows. Would that be a disproof of global warming? What if temperatures had stagnated for the past ten years? Would that be a disproof?
Science needs theories that make predictions about both what happens when they’re right and how to falsify the theory if they’re wrong. Scientists do not try to prove their theories. They try to disprove them. A scientist with a theory will tell you how to disprove their theory. That is, they will either show that their theory makes a prediction which, if it doesn’t come true disproves their theory; or they will show that the theory predicts something not happening, and if that something happens it disproves their theory.
- What is the theory?
- How does it predict some future event happening or not happening?2
- How will that prediction be measured?
Then, and only then, do you get to “perform the experiment”. You don’t get to make predictions in hindsight; you don’t get to choose the measurements afterwards. Not if you want it to be science.
But there’s bound to be some weird weather somewhere, and we will react like the sailors in the Book of Jonah. When a storm hit their ship, they didn’t ascribe it to a seasonal weather pattern. They quickly identified the cause (Jonah’s sinfulness) and agreed to an appropriate policy response (throw Jonah overboard).
You can always sift through statistics to find something out of the ordinary. Statistically speaking, there will always be something out of the ordinary: 50% of everything is above average. Parts of the United States are hot this summer; other parts were colder, but I wouldn’t be surprised if San Diego’s turns out to be colder than normal, too. We normally get at least a few hot days in July, and we haven’t yet. Now, there’s still time in August for hot days; if we do get hot days in August, I can expect that (a) there will be cold days in other parts of the world, and (b) those cold days will be ignored but our hot days will be proof of man-made global warming.
My Facebook friend wasn’t doing anything out of the ordinary proselytizing for global warming as if it were a religion rather than a science. Man-made global warming proponents tend to act more as a cult than as a religion, with instructions on how to talk with your relatives about the cause. It’s frustrating for those of us with a scientific background. Six years ago, a friend of mine was talking about how soon global warming was going to cause catastrophes. Within six years, the bay he lived one was going to be several feet higher; by 2016 it was going to be 12 feet higher. I have no idea where he heard that, but it doesn’t matter, because the failure of that prediction hasn’t changed his faith in global warming. He’s just turned to different prophets who are more careful to talk in terms of centuries rather than predictions that can be verified in a single lifetime.
Man-made global warming is an impressive theory; to become a science, its proponents need to start making measurable predictions that can prove their theory false. Instead, they hide their predictions and change their data. If man-made global warming were science, you wouldn’t see its practitioners seeing divergent temperatures between rural and urban measuring stations, and always adjusting upward to the higher temperature urban sites.
If man-made global warming were science, you wouldn’t see its practitioners predicting fifty million climate refugees by 2010 and then hiding the decline. Even natural evolution, where some things get bigger and some get smaller is proof of global warming: initially, global warming predicted smaller birds, but when birds started getting bigger, the researchers think that the trend is due to climate change.
The theory predicted one thing; they found the opposite. They decide that both results support their theory. That isn’t science. Science is making measurable predictions: predictions that specify both what will happen and how it will be measured.
Until global warming theory works in more than just hindsight, it’s not something we can spend trillions of dollars on, because there’s no way to plan for the future using a theory that can’t make predictions. If you think man-caused global warming is science, ask yourself “how can it be proven false?” When a religious leader predicts the end of the world on May 21, you know that when the end of the world fails to come, they’ll just re-predict the end of the world for October 21. Science doesn’t work that way. Science needs to be falsifiable, or it isn’t science.
- August 19, 2012: The 329th consecutive month of hot temperature adjustments
You can always sift through statistics to find something out of the ordinary.
This is especially true when you control the statistics. For example,
July was the fourth-warmest such month on record globally, and the 329th consecutive month with a global-average surface temperature above the 20th-century average, according to an analysis released Wednesday by the National Climatic Data Center (NCDC).
Actually, 864th, but matching the numbers sounds better. It’s all a matter of what statistics you choose to look at. But, as I said in July, you don’t get to make predictions in hindsight; you don’t get to choose the measurements afterwards. The most important part of this chart isn’t that the numbers are being adjusted upward, it’s that they’re being adjusted differently every year.
At first it was global warming, but when the globe didn’t cooperate it became climate change. And when the theory still didn’t allow for predictions, it changed to global weirding. The word itself half-invokes the supernatural, and is specifically chosen to avoid a testable theory.↑
Or, how does it predict the outcome of some experiment? But while there are a few experiments, such as the CERN cloud experiment, for the most part we can’t do experiments on the weather, we can only make predictions.↑