 In a debate, just having a bunch of people agree isn't enough. Groups of people agree and turn out to be wrong all the time. So what makes a scientific consensus different? Scientific consensus is different because it relies on three important ingredients. Evidence. Everyone involved agreeing upon standards for that evidence. And agreement from many different groups from many different backgrounds. In the study of the philosophy of science, we would say that consensus is likely to be correct when it is knowledge-based. And consensus is knowledge-based when it meets three important criteria. Consilience of evidence, social calibration, and social diversity. Consilience basically just means having many lines of evidence that are independent from but in agreement with one another that all point to the same conclusion. You can think of it like this. A friend is cleaning out his old games and gives you a bunch of different puzzle pieces. He was cleaning in a hurry and just threw pieces in an old shoebox. You don't know what the puzzle will look like or even if the pieces are really from the same puzzle. Maybe the pieces look like they're all showing something similar but don't actually fit together. Or maybe the puzzle pieces sort of fit together but look like they would have to come from very different pictures. When pieces of the puzzle fit together and show a picture that makes sense, you can be confident that you're on the right track. Science depends on evidence coming together and telling the same story. When we look at the average temperature of the Earth, we can see a bunch of different lines of evidence pointing to the same conclusion. Thermometers on the ground, on ships in the ocean, and on balloons in the air all show an increase in temperature. Glaciers around the world are melting, sea level is rising, moisture in the air is increasing. All of these things tell us that the world is getting hotter. The pieces of the puzzle fit together and the picture is clearer. Evidence is a big part of what makes science successful and that means we have to make sure that we're all using the same standards of evidence and speaking the same language, figuratively at least. We call this social calibration. It may sound obvious but people need to be in agreement about the concepts that they were discussing before they can come to a meaningful conclusion. To be able to address the question of whether the planet is warming, you have to agree on some basic concepts. It might sound silly to you and me but there are climate king trarians who actually deny that there is even such a thing as global temperature as a concept. But of course we can take temperature measurements from across the planet to get an average of the whole global temperature. People need to agree on what counts as a valid way of answering a question as well. Someone might feel that the answer was revealed to him in a dream. Someone else might have claimed to have found an answer in an ancient prophecy. But when dealing with scientific questions, it's important that we're relying on the rigorous standards of scientific inquiry. Consilience of evidence and social calibration are important but it might still not be enough. To be really confident that the consensus is correct, it also helps to see agreement coming from many different groups from many different backgrounds. In other words, we want to see social diversity. To understand why, it helps to look at cases in which a lack of social diversity can lead to the wrong conclusions. One is just plain old bad luck. It's always possible to reach a conclusion of yes when the answer correctly is really no or vice versa. It could be a statistical fluke or contaminated materials or even something dependent on the location of the group performing the experiment. Having many different groups from many different backgrounds can do a lot to rule out such problems. Another way in which a lack of diversity can lead to agreement that turns out to be wrong is group think. There's a tendency for groups of small numbers or highly similar groups to attempt to minimize disagreement and promote conformity. A desire for harmony within the group can cause people to ignore reservations they may have and reach agreement for the sake of agreeing rather than based on standards of evidence. A large diverse group will have less of an inclination towards group think as differences among the group exist from the outset. Cultural bias is another way in which a lack of diversity can lead to incorrect conclusions. Scientists are product of their cultures and different cultures have different preferences for how the world should be. Including scientific viewpoints from as many different cultures as possible helps ensure that agreement isn't the products of values rather than evidence. Over 80 national science academies around the world agree that humans are causing global warming. None disagree. Having a socially diverse consensus also guards against self-deception or outright fraud. Those with no stake in the outcome or who stand to lose rather than gain from outcome reaching the same conclusion as those who might benefit from it increases our confidence that the conclusion is correct. The consensus on climate shows clear social diversity. Keeping these conditions in mind we can look at examples from the past when consensus supposedly has been wrong. Climate contrarians are fond of holding up such examples as reasons to doubt the current consensus on climate as correct. A frequent example is that there was a consensus in the scientific community against continental movement which plate tectonics later proved to be wrong. But is this an example of a knowledge based consensus that failed? Does this example meet the conditions of conciliance of evidence, social calibration and social diversity? Let's assume that we're talking only about physical scientists actively engaged in the issue. For the sake of argument we'll say the condition of social calibration was met. There was no conciliance of evidence against continental drift. Some evidence that land masses moved was known for hundreds of years but the evidence didn't point conclusively towards a coherent picture from movement. Perhaps more interesting was how disagreement on this issue was related to nationality. Scientists with basically the same training, looking at the same evidence, were coming to different conclusions based on where they lived. Science historian Naomi Oreskes has shown how the cultural ideals among geologists in North America differed from those from Europe and how that affected their scientific views. Scientists in America tended to value democracy and were hostile to authoritarian decrees. They favored bottom-up inductive methods of interpreting the evidence. They were leery of top-down, big-man, big-theory explanations. And so they were resistant to the evidence for a continental drift in a way that their peers from Europe or South Africa and Australia weren't. Opposition to continental drift was not a consensus and definitely not a knowledge-based one. Other examples used to cast out on the modern consensus on climate fare about the same. When the pieces of the puzzle fit together you have conciliance of evidence. When everyone is using the same standards of evidence and speaking the same language, you have social calibration. And when agreement is widespread across many different groups of people from many different backgrounds, you have social adversity. And when you have all three, you have a knowledge-based consensus and you can be confident that it's correct.