Science is a path to knowledge

There are a lot of misconceptions about what science actually is, and, as a result, there are a lot of incorrect conclusions about the reliability and utility of science. I frequently encounter people who expect science to give absolute answers. They act as though science is a method for proving what is true with 100% certainty. As a result, they view cases where science led to an incorrect conclusion as evidence that science itself is flawed. You can clearly see this in arguments that a current scientific result doesn’t need to be accepted because “science has been wrong before” or “there used to be a scientific consensus that the earth was flat” (there wasn’t, but that’s another topic), etc. Similarly, there is a false view that a scientific conclusion is either 100% right or 100% wrong. In reality, science is a path to knowledge. It is a way of testing ideas and slowly building a body of knowledge based on the results of those tests. Sometimes, the path takes wrong turns, but unlike every other path to knowledge that has ever been invented, science is systematic and self-correcting and steers itself back in the correct direction, resulting in a gradual accumulation of knowledge.

Before I go any further, I want to acknowledge that this description of science as a “path to knowledge” is not original with me and was coined by my friend and fellow skeptic, The Credible Hulk. So, go check out their blog and Facebook page for more great science content.

I really love this description of science as a “path to knowledge” because it beautifully encapsulates what science is and why it works. You see, science does not give absolute results. In other words, it does not “prove” anything with utter certainty. Rather, science is all about probabilities. As I often like to say, science simply shows us what is most likely to be true given the current evidence. That probability can, however, always change with future evidence. Any scientific result can be overturned as new evidence comes to light.

The tentative nature of a scientific result is one of its great strengths, but it can lead to confusion. People often make the incorrect leap from, “science does not give definitive answers” to “science is uncertain and therefore I don’t have to accept a given result.” This is a flawed way of understanding science. Remember, it is a way for telling us what is most likely true given the current evidence. Therefore, it’s results should be accepted until such time as future evidence arises to discredit those results. Sticking with our path analogy, a lack of 100% certainty that a path is going the right direction would not justify abandoning the path altogether and wandering aimlessly through the forest. Further, a lack of 100% certainty does not mean that we cannot be highly confident in a result. There are some things that have been so thoroughly tested so many times in so many ways that it is extraordinarily unlikely that they are wrong. In other words, some paths are marked well enough that you can be really confident in them.

On the other end of the spectrum, people ignore the tentative nature of scientific conclusions and act as though it should give definitive answers, leading to the flawed arguments about science having been wrong in the past. These arguments are problematic in a number of important ways. First, they treat the inherently self-correcting nature of science as if it is a bad thing, when in fact, it is another great strength of science. Really think about this. If you are going to argue that, “I don’t have to accept a scientific result because scientists used to think sun moved around the earth,” my question would be, “why do we no longer think that the sun moves around the earth?” The answer is very clearly that other scientists continued conducting tests and discredited the previous view. Science corrected itself. This is not a weakness, but rather a strength. No other path to knowledge does this. No other system of understanding repeatedly and systematically tests its conclusions and updates its information by rejecting debunked results and accepting new results.

Further, because of the way that science advances, the argument that “science has been wrong before” is inherently self-defeating. Sticking with the orbit of the earth for a minute, we only know that the earth orbits the sun because science debunked the notion that the sun orbits the earth, so you can’t use that as an argument that science doesn’t work, because the argument inherently includes the premise that science works! In other words, if this argument gives us carte blanche to disregard scientific results, then why should we accept the result that the earth moves around the sun? That result was produced by science, and this argument claims that we don’t have to accept scientific results, so why should we accept the result that the earth moves around the sun? We only know that science was wrong before because of science. Again, this self-correction is one of the best things about science.

Additionally, it is important to realize that scientific results are often incomplete more than actually wrong, and there are degrees of wrongness. The progression of physics is a great example of this that I use frequently. Newton made enormous strides in physics. He moved us far along the path, but we later found out that he was slightly off course. Einstein showed that Newton’s work was incomplete and his conclusions did not apply universally. However, that didn’t mean that we threw Newton out the window and went all the way back to the trail marker Newton started at. Newton moved us closer to the truth, and Newtonian physics are still taught and applied all around the world, but he was incomplete, and Einstein took Newton’s results and shifted us back on track. Think of it like this: we needed to go north, and Newton took us slightly north west. He still moved us much closer to our goal, but we needed Einstein to reorient us and get us back on track.

This gradual accumulation of knowledge is another key aspect of science. Yes, science sometimes makes mistakes, but because it corrects those mistakes, we gradually get closer and closer to the truth. People who thought the sun revolved around the earth were less wrong than people who though the sun was a god. Galileo was less wrong than the people who thought the sun moved around the earth. Newton was less wrong than Galileo. Einstein was less wrong than Newton, etc. At each step, we got closer, and closer to the truth. This is also another reason why it is so absurd to blindly disregard modern scientific results on the basis that science has been wrong before. Science is a gradual accumulation of knowledge, and although there certainly are things about which we are wrong today, we are less wrong than previous generations, and we know this because we tested the views of previous generations and built on that knowledge.

To give another example, there are certainly things about which modern medicine is wrong. That is inevitable due to the tentative and probabilistic nature of science, but modern medicine is less wrong than medicine was 20 years ago, and medicine 20 years ago was less wrong than medicine 40 years ago, and medicine 40 years ago was less wrong than medicine 60 years ago, etc. Further, I can demonstrate this extremely easily. Imagine you need a major medical intervention and you can be treated using the technology and knowledge from any of the following time points: 200 years ago, 100 years ago, 50 years ago, 25 years ago, or current. Rank your choice from lowest to highest. I’m willing to bet your choices went chronologically (inverse) with your preference being treatment via our current knowledge, and there is a very good reason why that is the correct way to rank things. Namely, science works! It’s not perfect, but it is a path that moves generally in the right direction, and we all intuitively realize that science has helped us progress and, thanks to science, we know more than any generation before us knew.

Further, we can extend my medical analogy to just about any field of science. Imagine that you are on a game show run by omnipotent aliens with a perfect knowledge of the universe. They ask you a chemistry question, and you have a lifeline that will let you call a random chemist from the current year, or from 25 years ago, or from 50 years ago, etc. Whom do you call? Obviously, you call the chemist from the current year. Again, we all intuitively accept that science works and gradually builds knowledge. Even those who like to argue that “science has been wrong before” must admit that, thanks to science, we know more now than at any other point in our history. Science has a proven track record of moving us in the right direction.

Finally, if you are not convinced by anything I’ve said thus far, then my question for you is simply, “what’s the alternative?” Really think about this. What other path to knowledge can compete with science? As I’ve explained before, science is responsible for our modern society. All of the technological and medical marvels around you are the result of gradually testing ideas and accumulating knowledge. Look at all the previously fatal diseases that we can now cure or even prevent, look at the decreases in mortality rates, etc. All of that is because of science. So why should we go back to unsystematic guess work? We tried other systems (like relying on anecdotes) for millennia, and they didn’t work. It was science that brought us out of the dark ages, and it is science that will allow us to continue our advancement as a species. Again, that doesn’t make science perfect or infallible. It simply shows us what is most likely true given the current evidence, but by constantly testing, by constantly self-correcting, by constantly updating, it gradually moves us closer and closer to the truth. It’s not perfect, and it certainly isn’t a straight path, but it’s the best path to knowledge that we have.

Note: To anyone who is about to reply with a snarky remark about doctors/scientists saying that smoking is safe, please read this post. The reality is that there was never a scientific consensus that smoking was safe and, in fact, science had showed that it caused cancer all the way back in the 1930’s. Indeed, actual studies consistently showed that it was dangerous. Tobacco companies simply did a good job of creating the illusion that science was on their side; meanwhile, actual science was continuing along the correct path.

Related posts


Posted in Nature of Science | Tagged , | 29 Comments

The problem with “just asking questions”

Asking questions is generally a good thing. Indeed, questions are the very foundation of science. People become scientists because they are curious and like to ask questions, and science itself is simply a systematic method for asking and answering questions. Unfortunately, the positive perception of questions often leads to people using questions as a disguise for wilful ignorance, and the phrase, “just asking questions” has been used to justify all manner of insane and illogical beliefs. The people who use this phrase are generally not actually asking questions. Rather, they are phrasing a belief as a question in an intellectually dishonest attempt to maintain the appearance on intelligence.

There are two major problems that I am going to discuss. The first is simply that not all questions are good. I fundamentally disagree with the notion that there is no such thing as a stupid question. Good questions stem naturally from known facts and evidence. In other words, they have a basis in reality. Bad questions, however, are not based on facts or evidence and instead rely on wild conjecture. Indeed, in science, hypotheses do not spring out of nowhere. Rather, they are based on the existing evidence.

Let me give an example. In my field (herpetology) there has been a fair amount of debate and discussion about the purpose of basking behavior in turtles (i.e., why do aquatic turtles come out of the water and bask on rocks and logs?). There have been many hypotheses/questions that people have looked at. For example, is it for thermoregulation (temperature)? Does it help immune functions? Does it remove parasites? Etc. All of these are good questions. They are perfectly rational things to wonder about based on our existing knowledge of biology.

Now, however, imagine a scientist asked, “Are they basking to avoid aliens that live in the water?” That would be a bad question, because it’s not based on any known facts. There is no reason to think that aliens are involved, and we’d need good evidence of the presence of aliens before it would be rational to even consider the possibility that they are involved. If a scientist asked that question at a conference, they would be laughed out of the room, and they absolutely could not justify it by saying, “I’m just asking questions. Aren’t you scientists supposed to be open-minded?” Yes, scientists should be open-minded, but being open-minded means being willing to accept new ideas when presented evidence for them. It does not mean being willing to accept or even consider the possibility of aliens influencing turtle behavior despite a lack of evidence that aliens are living in our aquatic ecosystems. Do you see the point? You can’t just say something insane that has no evidence to support it and justify it as, “just a question.” There needs to be some reasoning behind the question. There needs to be some actual evidence to make the question worth perusing in the first place.

If we apply that to current events, questions like, “where did coronavirus come from?” are fine. That’s a totally reasonable thing to ask. Even asking “is coronavirus man-made?” was not entirely unreasonable at first (see below), because there is a very real possibility of people bio-engineering viruses. However, a question like, “did Bill Gates invent coronavirus so that he could microchip everyone?” is not a good question. That is a stupid question, because there is utterly no evidence to suggest that either Gates engineered the virus or that Gates is trying to microchip people. The question, “Did Bill Murray engineer coronavirus because he enjoyed being in Zombieland and wanted to try an apocalypse in real life?” is just as valid, by which I mean, just as stupid. The fact that something is phrased as a question does not make it rational.

The second major problem with people “just asking questions” is that those questions are rarely good-faith questions being asked out of honest curiosity. Rather, they are often statements of belief that are being disguised as questions. Many (if not most) of the people asking things like, “did Bill Gates make coronavirus?” don’t actually want the answer. Rather, they are confident that they know the answer, and that’s a problem.

Asking questions is only a good idea if you are willing to accept the answers to those questions. In other words, asking a question like, “is coronavirus man-made” is fine if it is being asked out of a genuine sense of curiosity and desire for knowledge. There is nothing wrong with asking that question if you are then willing to look at the evidence and accept the answer provided by that evidence (in this case, the answer is a clear, “no, it was not man-made”). The problem is that many people asking the question won’t accept that answer. They refuse to accept the evidence, but also don’t want to admit that they are denying evidence. So, instead, they claim to be “just asking questions.”

To be clear, I don’t think most people are deliberately using the phrase “just asking questions” because they know that they are denying evidence and don’t want to look foolish. Rather, this is simply one of many cognitive traps that people fall into. Most of the people who go around justifying nonsense by saying that they are “just asking questions” probably truly think that they are being rational and are simply asking good questions. So, the point of this post is really to act as a warning. Be conscious of your views and biases, and if you find yourself “just asking questions” stop and ask yourself, “why am I asking this? Is there actual evidence to suggest that this is a good question?” Then, if you think that it is a good question, actually look at the evidence. If you aren’t willing to look at the evidence, then you are stating a belief, not a question. Once you’ve been shown the facts, it is no longer rational to keep asking the same question. Once you’ve been given the answer, your choices are either to accept it or deny it. You cannot claim to be rationally asking questions if you’ve already been given the answer to your questions and simply refuse to accept it.

Finally, it is worth explicitly stating that when I say to look at the evidence, I mean actual evidence from reputable sources. Youtube videos, conspiracy websites, outlets on either extreme of the political spectrum, someone you know on Facebook, a cherry-picked expert, etc. do not count. To quote Will Turner, “that’s not good enough.” In science, your evidence needs to come from the peer-reviewed literature, and you need to look at the entire body of literature, rather than cherry-picking, and for topics like politics and current events, you should get your information from multiple reputable news outlets. Don’t accept the first source you come across. Rather, cross-reference it using multiple other sources and see if they all say the same thing (the Media Bias Chart is a very useful tool for seeing if the sources you are using are neutral and reliable).

My point with all of this is simple. You should ask questions. You should think critically and evaluate what you are told, but your questions need to be based on known facts, and they need to be good-faith questions that are asked out of an honest curiosity. You must be willing to answer them by actually looking at evidence from reputable sources and accepting facts.

Related posts

Posted in Rules of Logic | Tagged , | 6 Comments

Shoddy statistics and false claims: Dr. Erickson dangerously misled the public on coronavirus

By now, you have likely seen the viral video of two doctors in Bakersfield, California (Dan Erickson and Artin Massihi) holding their own press briefing in which they argued that COVID19 is no deadlier than the flu, shelter in place orders are doing more harm than good, and schools and businesses should re-open. Clips of the press briefing have rapidly been latched onto by many people for a variety of reasons ranging from political leanings to desperation for hope. Unfortunately, these doctors have no clue what they are talking about, badly blundered the statistics, made numerous false claims, and have enormous financial conflicts of interest (i.e., re-opening businesses would be tremendously financially beneficial for them). In short, they are emergency doctors, not virologists, microbiologists, immunologists, or epidemiologists, and they should leave the statistics to people who are properly trained to analyse them. To demonstrate that, I’m going to go through their nonsense point by point, starting with what I consider to be the core issues. Please note that I will only be discussing the science and logic behind sheltering in place, not the politics.

Note 1. I based this post on the entire 1+ hour interview, not the short 2–5 minute segments.

Note 2. It’s worth mentioning that the American College of Emergency Physicians condemned their statements.

Bad statistics

Incorrect mortality rates

Their entire argument rests on the notion that the mortality rate from COVID is actually very low, even less than 0.1% (roughly the typical mortality rate from the flu). Actual studies have found that the mortality rate varies from  3.6% (Baud et al. 2020) to 1.4% (Wu et al. 2020). I have yet to see an estimate based on confirmed cases that was anywhere near the number these emergency doctors came up with (see Note 3). So how did they get such a low number? Easy: they’re bad at statistics.

To get their numbers for a given county, state, or even country, they took a series of simple steps. First, they took the number of tests that had been conducted and calculated the percentage of positive results. Next, they “extrapolated” that by applying that percentage to the entire population of the geographic region in question to calculate the total number of positive cases. Finally, they divided the number of deaths by their calculated number of cases, and lo and behold, the death rates were low, way lower than actual epidemiologists have calculated (see example in Note 4). Why is that? Anytime you see one or two “experts” present a value that is vastly different from what all the other experts have arrived at, you should be suspicious, especially if they announce their findings in a blog, press conference, etc. rather than the peer-reviewed literature (where real scientists present their findings).

In this case, there are two glaring problems with their analyses. First, you simply cannot extrapolate the percent of positive tests to the entire population because it’s not a random sample. Imagine, for example, that we have a bag with 1,000 marbles some of which are black and some of which are dark blue. We don’t know how many of each there are, so we reach in and pull out several random handfuls and count them, and we find 50% black marbles and 50% dark blue marbles. From this, we’d conclude that there are roughly 500 black marbles and 500 dark blue marbles in the bag. That would be a fine extrapolation, because we took a random sample. Now, however, let’s say we can see partially into the bag. It’s a bit dark so we can’t always tell the color of the marble for sure, but we deliberately select the marbles we think are dark blue. From this, we find that 20% are black and 80% are dark blue. Can we conclude that 800 of the marbles in the bag are dark blue? Obviously not! We clearly took a biased sample, which means we can’t extrapolate from it. This is experimental design 101.

Coronavirus testing thus far has been a very biased sample. It has not been truly random sampling. Rather, it has been heavily biased towards people who had symptoms, people who were in contact with someone who developed COVID19, people at high risk, etc. In other words, the percentage of positive cases in the testing is probably much higher than the actual state or country-wide percentages, just as our estimate of dark blue marbles was unrealistically high. This means that our intrepid doctors overestimated the total number of cases, thus vastly underestimating the mortality rate. They calculated mortality by dividing known deaths by their estimated cases, which means the higher the number of estimated cases, the lower the death rate.

The other problem is that they are only using the people that have died thus far, but that number is going to keep going up, even if no one else ever becomes infected. In other words, some of the people who are infected with COVID right now are going to die. So, you can’t take the ongoing infection data and divide the number of deaths by the number of cases, because people are still dying. That number is going to keep going up. To illustrate, let’s say that we have 10,000 currently infected people, plus another 10,000 who have either died (100) or recovered (9900). It would be stupid to take those deaths (100) and divide by all those cases (20,000) and conclude that there is only a 0.5% death rate. We can’t do that because we still have 10,000 people who are infected, some of which will die. Do you see the point? Using these numbers midway (as they did) biases the results towards a lower death rate.

These very basic problems with their analyses completely nullify their results. The numbers they are basing their arguments on are invalid, which means that they have nothing to back up their claims.

Note 3: Both the number of confirmed cases and the number of confirmed deaths are almost certainly large underestimates, but since we don’t know those values, it’s hard to know what the true death rate is. That does not, however justify the type of shoddy statistics they used. Also, there have been several recent prevalence estimates based on antibody tests that argued for a much higher disease prevalence than is currently documented, but the estimates thus far have been riddled with problems (including non-random sampling) that are beyond the topic of this post.

Note 4: To work through the math of one of the examples they gave, at the time they collected their numbers, California had done (according to them) 280,000 tests, with 33,865 positives, giving a prevalence of roughly 12%. They then assumed that 12% applied to the entire state and multiplied 0.12 by 39.5 million (CA’s population), resulting in 4.7 million calculated cases (according to them; they didn’t round correctly). Now, if we just divide the current number of deaths (1,227) by that number and multiply by 100, we get a mortality rate of 0.03%. (I have not checked their numbers as far as number of cases and such; I’m just reporting their shoddy math).

Sweden vs Norway

A core piece of these doctors’ argument is that sheltering in place doesn’t work. They claim that there is peer-reviewed evidence to support this (I haven’t seen any), but the only “evidence” they present is a comparison between Norway (which shut things down) and Sweden (which did not shut things down officially, but still many work from home, engage in social distancing, etc.). The numbers, according to them (which seem at least close to correct for a few days ago) are as follows: Norway (shut down) has a population of 5.4 million and has 7,191 confirmed cases and 182 deaths. Sweden (not shut down) has a population of 10.4 million and has 15,322 cases and 1,765 deaths. According to them, those death rates are “statistically insignificant,” a term that they clearly don’t understand. You can’t just eyeball the numbers and assert that they aren’t significant. You have to actually do some statistical tests. Based on those numbers, Norway has had 34 deaths per 1 million people. In contrast, Sweden is all the way up at 170 deaths per million! If we do an actual statistical test (chi square), that difference is, in fact, highly significant (P < 0.0001; this means that there is less than a 0.01% chance that a difference this great or greater could arise by chance; more on stats here). Again, these guys are emergency doctors, not statisticians or epidemiologists. They are talking about things they do not understand.

Now, to be clear, that comparison I just made is not great either. I only made it to show the absurdity of their claim that those numbers aren’t different. The reality is that there are tons of differences between those countries that make it very difficult to make such a comparison. For example, not only did Norway lock down, it also has done substantially more testing than Sweden which can also have a huge effect (it’s also why the number of cases per million is similar between the countries even though the per capita mortality rate is so different). Any sort of country comparison like that is inherently problematic, particularly if you have a sample size of 2 countries. Also, a better approach is to look at trends, not snapshots. For that, I’d take a look at the graphs presented by the BBC (they are on a log scale, so the difference between Norway and Sweden is substantial). My point is simply that their analysis is totally bogus. Once again, the data aren’t on their side.

Shutting things down works

Another core piece of their argument is that shutting things down and sheltering in place aren’t effective. This is based on, as far as I can tell, utterly nothing. They claim there are studies to support their claim, but they don’t cite them and I can’t find them. They also try to use the Norway/Sweden comparison, but as I showed above, if anything that actually suggests that shutting down works. Finally, when pressed by a reporter for their evidence, one of them seemed frustrated and said, “I don’t need a double-blind clinically controlled trial to tell me if sheltering in place is appropriate. That is a college level understanding of microbiology.”

Now that statement is interesting for a number of reasons. First, up until that point (it was late in the interview), they kept insisting over and over again that they were just following the science. They repeatedly claimed to be the ones objectively looking at facts. Yet when pressed for their evidence, they retorted by saying they didn’t need studies, because they just knew (a very common science-denier strategy). Further, the effectiveness of sheltering in place is clearly not something you’d test with a double-blind clinically controlled study, which makes me suspect that they know very little about experimental design.

In reality, despite their claims to the contrary, very basic math tells us that sheltering in place will work, and you can very clearly see the pattern across the world of countries shutting down, followed by flattening the curve and, if they stay shut long enough, levels dropping. Indeed, if you just think about this for a second, it should make perfect sense. You can carry COVID for roughly 2 weeks without symptoms. So, if you are out and about, you are spreading that everywhere. If you are at home, you aren’t spreading it. Further, if you are at home, then you aren’t being exposed to others who might spread it to you. Even a very, very basic understanding of epidemiology is enough to realize that the rate of viral spread in a population is strongly influenced by the number of interactions people have with other people. The more interactions, the greater the spread; the fewer interactions, the less the spread. I can’t believe I even have to explain that. This is why density is such a critical component of disease outbreaks. These guys can present themselves as experts all they want, but they clearly don’t know what they are talking about, which is why actual epidemiologists and health officials say they’re wrong.

Death rates are low because we took action. Also, the outbreak is ongoing

At several points, they criticized the early models that predicted hundreds of thousands or even millions of deaths. They cited them as evidence that people over-reacted and the disease is not much deadlier than the flu. This argument is, however, based on a poor understanding of the models. These sorts of models don’t show what will happen, rather they showed what could happen under a range of scenarios. We run them precisely so that we can change our behaviour and avoid the worst outcomes. That’s literally their purpose. We don’t run them for the fun of it. We run them so that we can learn how to save lives. The number of deaths is much lower than originally predicted because we shut down schools, implemented shelter in place orders, etc. The things that these doctors want to undo are the very things that prevented us from having millions of deaths. This is very much like anti-vaccers arguing that we don’t need the measles vaccine because measles deaths are rare. They are only rare because most of the population is vaccinated. Even so, the death rates are “low” because we implement measures to make them low.

The other thing to keep in mind is that the situation is very much ongoing. At the time I’m writing this, the US has nearly 57,000 deaths from COVID-19, and it is still adding well over 1,000 (often over 2,000) deaths daily. Erickson frequently cited annual US deaths from the flu as being 24,000-62,000 (at another point they said 37,000–67,000; it’s actually 12,000­–61,000). He used this as evidence that COVID is no worse than the flu, but stop and think about that for a second, we are already at nearly the highest end of that range, and given the rate at which people are dying, we are going to shoot well past it, and that is with things shut down. Just think how much worse this would be if things weren’t shut down! The only reason they are even able to make that comparison is because we shut things down. They are simply wrong that COVID isn’t deadlier than the flu.

Comorbidity doesn’t mean COVID isn’t responsible

They also spent a great deal of time arguing that COVID isn’t really the killer, it’s actually the other conditions like being immunocompromised, being a smoker, etc. This is a very stupid argument. Yes, most mortalities are associated with other factors, but that does not in any way, shape, or form change that fact that those people would not have died if they had not caught COVID. Indeed, that’s a big part of why keeping things shut down is so important: it protects the people who are the most vulnerable.

Interestingly, this argument is another one that they lifted straight out of the anti-vaccine playbook. Anti-vaccers frequently make the same argument claiming, for example, that measles doesn’t kill anyone; it’s the secondary infections that kill. This, of course, ignores the fact that those infections happen because of the measles. Even so, yes, COVID generally has help in killing patients, but that doesn’t negate its role. This argument is like talking about a gunshot victim and arguing that, “Bullets aren’t dangerous, because the bullet didn’t kill him; it was really the loss of blood.” It’s a very dumb argument.

 Sheltering in place won’t destroy your immune system

A final core thrust of their argument is the notion that sheltering in place will harm your immune system and make you sick. You see, according to them, viruses and bacteria are the “building blocks” of your immune system, and if you aren’t regularly exposed to them, they will all disappear somehow, they won’t protect you, and your immune system will be weakened. They also extended this to arguing that you shouldn’t disinfect things in your house, shouldn’t wear a mask in public, etc. They justified this by smugly saying, “we’re not wearing masks. Why is that? Because we understand microbiology, we understand immunology, and we want strong immune systems.”

To quote a famous meme, “That’s not how this works, that’s not how any of this works!”

There are a ton of issues here, but let me start by acknowledging the grain of truth in their silo of stupidity. It is true that you have a microbiome consisting of many bacteria, viruses, etc. and they do play important roles in your body, potentially including helping fight some diseases. Also, there is some evidence that exposure to microbes early in life helps to train the immune system to respond correctly, and a lack of those microbes results in autoimmune disorders and allergies (this is known as the “hygiene  hypothesis”).

None of that, however, supports their claims. First, there is no reason to think that staying home for a month or two is going to dramatically alter your microbiome. They acted as if all the bacteria living in and on you will die if you don’t go outside. That’s nuts. They will keep living and reproducing and doing bacterial things. You are their home. That’s where they live. To be clear, a change in your routine might shift the microbiome around slightly, but it is constantly shifting around slightly, and there is no scientific evidence that sheltering in place is going to shift your core microbiome in a detrimental direction, and it certainly isn’t going to deplete your body of bacteria.

Further, they acted as if your house is totally sterile (except when it was convenient for them to act otherwise; see later), which is insane. Even if you disinfect your counter (as they waxed on about), your house is crawling with bacteria. Do you live with other people? They have bacteria. Do you have a pet? They are coated in bacteria. Even inside, you are constantly exposed to bacteria other than the ones on and in you.

Additionally, microbes are not the “building blocks of the immune system.” They aren’t even part of the immune system. Sure, they train your immune system, but only in that it learns which microbes to attack (and how to attack them) and which ones not to attack. Many people (these guys included) talk about exposure to bacteria “strengthening your immune system” as if exposure to bacteria A will result in a general improvement in your immune system and ability to fight other bacteria, but that simply isn’t how it works. As I explained in detail here, exposure to bacteria A simply teaches your immune system how to kill bacteria A and whether it needs to. It doesn’t “strengthen” it against other bacteria/viruses.

Note 5: Unlike these two, I have actually published papers on host microbiomes.

They lied when they said experts agreed with them

At one point, they said that they had shown their results to local health officials, and those officials agreed with them and were just waiting on permission from the governor to re-open things. That was, however, a lie. A spokesperson for the Kern health department said “our director has not concurred with the statements that were made yesterday about the need to re-open at this time.” As a general rule, I don’t trust people who make such brazen lies.

They aren’t experts/don’t cherry-pick your doctors

This is a point that I have touched on repeatedly throughout, but it is worth stating again: these two are emergency care doctors, not microbiologists, not immunologists, not virologists, not epidemiologists. They are not experts on a topic like COVID. They are not people you should be treating as authorities. Having an MD does not make you an expert on all aspects related to medicine. They state early on that they have taken courses on these topics, which I’m sure they did back in pre-med/med school, but that doesn’t make them experts. I took courses on these topics, as well, as part of my training in biology, and, as part of my PhD, I even studied microbiomes and the effects of an emerging infectious disease on the ecology of wildlife populations, but that doesn’t make me an epidemiologist. The fact that I have a little bit of training and experience in that field does not make me an expert in it, and it certainly doesn’t put me on par with people like Fauci who have spent their entire lives studying these topics.

When pressed on why actual infectious disease experts fundamentally disagree with them, they first tried to dodge the question by going on rabbit trails about how Fauci’s actions weeks ago were justified because he didn’t have all the data (which completely ignores the actual question of why people like Fauci still disagree with them now). Then, they eventually argued that the disagreement was because people like Fauci have just been doing research from afar for years, whereas they are “in the weeds” seeing how things are on the ground. This is, of course, an insane argument. Beyond the fact that many (probably most) other healthcare workers who are “in the weeds” with them disagree (we’ve all seen the photos of nurses blocking protesters who are trying to open things up), treating COVID patients does not in any way shape or form make someone an expert on the factors and conditions that allow the virus to spread. Emergency patient care and epidemiology are two very different things and doing one does not qualify someone as an expert in the other.

Nevertheless, I’m sure there will be some who continue to insist that these two know what they are talking about because they are doctors, at which point my question becomes, “why trust them?” There are thousands of doctors with far more relevant experience who disagree, so why trust these two? Why cherry-pick them out of all the experts? What makes them more trustworthy than all the other MDs and PhDs? Is it possible that you are blindly believing them not because they have good data (they don’t) or because they are experts (they aren’t) but rather simply because they are saying the things you want to hear? You should carefully consider this possibility, because it is a very easy cognitive trap to fall into.

Massive conflicts of interest and probably biases

It’s always a good idea to see if people have something to gain from making public claims like this, and in this case, the conflicts of interest couldn’t be clearer. Erickson started off by saying that many hospitals have been furloughing staff, shutting down wings, etc., and he returned to that point frequently. Then, at the end, he pointed to the various news people in the room and asserted that if COVID had cost them their jobs and they weren’t getting a paycheck, they might have a different view of the situation.

I found that very interesting, because Erickson and Massihi aren’t simply doctors. Rather, they own a series of urgent care facilities (Accelerated Urgent Care), which, as they admitted, aren’t getting many customers right now. In other words, the shutdown is hurting them financially, and re-opening would be hugely beneficial to them, but I’m sure that had no influence, right? Never mind the fact that they literally said finances would influence people; I’m sure they are just after the public good (sarcasm).

There also seem to be some very strong political biases at play. Erickson hinted at a conspiracy throughout, with frequent statements about “something else going on.” For example, he asserted (with no evidence whatsoever) that ER doctors were being pressured to write “COVID” on death certificates for some political motive. “Why are we being pressured to add COVID? To maybe increase the numbers and make it look a little bit worse than it is? I think so.” This is a nonsense conspiracy that I have not been able to find any evidence to support. Further, he admitted that it would be administrators who would be pressuring doctors, but when asked why administrators (who are hurting financially from COVID) would try to artificially increase the death rate, he didn’t have an answer.

Things really went off the rails after the main chunk of the interview (and a related moment part way through) where Erickson started ranting about the word “safe.” In regards to official health recommendations, he said, “when they use the word safe, the word safe, if you listen to the word safe, that’s about controlling you.” Ah, the ubiquitous and amorphous “they.” Who is “they”? Who knows? Probably some evil government entity. Later he said (his emphasis), “who says what’s safe? Are you smart enough to know what’s safe for you or is the government gonna tell you what’s safe for you…They are using this to see how much of your freedom can they take from you.” This is, of course, pure pablum. I wonder if he also applies this line of reasoning when the government tells people how to be “safe” when a hurricane is coming. I also wonder if he has more trust for health officials when they determine the “safe” doses of drugs. It’s just insane to argue that health guidelines designed for public safety are really about public control, and, more importantly, it makes his biases abundantly clear. It is extremely obvious that he is motivated, at least partially, by a strong distrust of the government. Despite all of the claims to be just looking at the evidence, they gave a lot of indications of following political biases, not facts.

Note 6: In cases like this, the general public does not know enough to know what is safe. To be clear, it’s about knowledge, not intelligence, but most people simply don’t have the training and experience to actually evaluate epidemiological data and determine what is safe. That’s why we should listen to experts like Fauci, not some random person on the street. This assault on expertise is yet another very common anti-science strategy.


At this point, I have covered the bulk of this nonsense and hit the most important points, but I want to touch on a few more minor issues.

Quarantining the healthy

At several points they asserted that quarantines have never applied to healthy people before. This is a lie. During the 1918 Spanish flu outbreak, things were shut down very much as they are today.

Coronavirus on plastic objects

At one point, they argued that because coronavirus can live on plastic for up to three days, sheltering in place is pointless because you are just bringing it into your house when you buy things. They even asserted that they can probably find COVID in your house. This is just stupid. The odds that a shovel from Home Depot (to use their example) has COVID on it are very low, and much, much lower than the odds that one of the many people you’d be near in Home Depot would give you COVID (seriously though, you probably don’t need to go to Home Depot right now). The fewer people you are in contact with, the lower your odds. It’s that simple. So yes, sheltering in place absolutely does minimize your risk.

Also, note that by claiming that things you bring into your home commonly harbor micro-organisms, they have just totally negated their own argument about staying at home preventing you from being exposed to bacteria and viruses. You can’t have it both ways.

The need for more testing

At one point, they accidentally gave up the game and said, “In order to re-open the economy, you have to have widespread testing, that’s #1.” That is actually something I agree with. If we had sufficiently wide-scale testing, we could start slowly re-opening because we could monitor cases and quarantine the sick. The problem is that we aren’t there. Not even close. We don’t have nearly enough testing to be able to do that. Thus, this admission defeats their entire argument.

When a reporter asked when they thought testing would be sufficient, they dodged the question and started arguing that people aren’t getting tested because they are just so scared to leave their home (never mind all the testing shortages).

More contagious than the flu

They also shot themselves in the foot by admitting that COVID19 is more contagious than the flu. Think about this with me. Their core argument is that COVID’s death rate (i.e., deaths among infected individuals) is the same as the flu’s. Even if that were true (it’s not), COVID would still be more dangerous because of the higher transmission rate. Being more contagious would mean more people get it, and, as a result, more people die.

Claimed New York ordered 30,000 ventilators and used 5

In another odd segment, Erickson went on a tangent about how ventilators aren’t saving anybody (which is untrue) and claimed that New York only used 5 of the 30,000 ventilators it ordered. I have not been able to find any evidence that NY over-ordered ventilators, and I certainly haven’t found evidence that only 5 were used.

Misuse of the word “theory”

At another point one of them said, “you’d better have a very good scientific reason and not just theory.” This is a minor issue, but I really hate the misuse of the word “theory.” In science, a theory is not an educated guess. It is an explanatory framework which has been rigorously tested and shown to have a high predictive power. Theories don’t graduate to become facts; rather, they explain facts. Saying that something is a theory does not indicate that we are uncertain about it. The very notion that viruses cause disease is a theory (i.e., the germ theory of disease).


In short, these doctors have no clue what they are talking about and seem to be motivated by money and politics, not science. Their statistics are bogus and their facts are faulty. ABC should be ashamed of itself for broadcasting their nonsense.

I think information should be shared and available. So you may re-blog, re-post, quote, etc. as you see fit, just please include a link/credit back to my page. Thanks

Posted in Uncategorized | Tagged , , | 1 Comment

Is sex binary? Let’s look at the biology

Are there more than two sexes? This is a question that has caused an enormous amount of social and political debate in recent years, but at its core, it is a scientific one, and I want to treat it as such. In other words, what we do with the answer to that question certainly has social and political ramifications, but the question itself is one of biology, not politics. Therefore, I am going to try to answer it in this post from a strictly scientific standpoint. I am not going to make any statements about politics, morality, religion, etc. Instead, I am going to talk only about the biology. As always, if you are going to read this, then all that I ask is that you lay aside any ideologies and views you might hold and look solely at the facts. Political and social positions must be based on facts, not the other way around. So, in this post, all I am going to do is present the facts.

Terms, definitions, and critical background information

On topics like this, it is always a good idea to define the opposing positions at the outset. In this case, there are basically two camps. One holds that sex is strictly binary and is determined by the presence or absence of a Y chromosome (sometimes stated more explicitly as XX = female, XY = male). The other position argues that sex is more complicated than this binary and follows more of a spectrum rather than a clear dichotomy. Some people misunderstand this and construct a straw man about people arguing for the existence of “third sex.” The argument is not that there is a third sex, but rather that sex cannot be adequately defined by two discrete categories because there are many people with both male and female traits. In other words, one position argues that sexes can be defined as two distinct boxes into which all individuals fit. The other argues that the situation is more complex and there are some individuals who do not fit cleanly into either box and are actually somewhere in the middle. I think part of the confusion arises over the way that we talk about this, and I fully admit that I have been guilty of this as well. We often say things like, “there are more than two sexes” as convenient shorthand, but what we really mean is that sex cannot be adequately defined using a simplistic dichotomy in which all individuals with a Y chromosome are males and all individuals without a Y chromosome are females. It is more complicated than that, and there are many intersex individuals that do not fit neatly into traditional categories of males and females.

Note: I opened with the question, “Are the more than two sexes?” simply because that is terminology that is familiar to most readers and introduces the topic.

Note for clarity: based on the comments thus far, I want to clarify that I don’t have a problem with a more nuanced position that says something to the effect of, “based on reproductive physiology most individuals are binary in that they either have the physiology for producing sperm or the physiology for producing eggs, but there are a variety of exceptions to this. So the binary classification is useful in some contexts, but we should acknowledge that there are exceptions and situations where the binary classification does not work.” Indeed, that is more or less what I am arguing (also note that I am only talking about biological sex, not gender or gender identity).

Next, we need to talk about how we define sexes, and before we get specifically to humans, it is really important to look at the biology of sex more broadly, because this gives us context and important background information. So, let’s start with the general definition of male and female. If you were to ask professional biologists to provide a general definition of “male” and “female,” the one answer you are not going to get is, “if a Y is present it’s a male; if Y is absent, it’s a female.” There’s a very good reason biologists don’t use that definition. Namely, because it doesn’t work for a very large number of organisms. You see, many organisms don’t have sex chromosomes; instead, male vs female is determined by some environmental factor (I’ll come back to that in a minute). Further, even for species with sex chromosomes there are lots of exceptions and atypical situations (again, more later).

Because of these problems biologists have historically defined sex based on the production of gametes (sperm and egg). The sex that produces small (usually mobile) gametes is considered to be the male, and the sex that produces the large, stationary gametes is considered to be the female. Thus, it is the production of gametes that defines sex, not the presence of a particular chromosome. To put that another way, sex is defined by gamete production, but in some cases, it is determined by chromosomes. In others, it is determined by environmental factors. This may seem like pointless semantics, but it is actually really important (as will become increasingly clear as we go), because the biological definition of sex is not about chromosomes. This already puts the “Y = male” position on shaky ground (it’s also worth noting that in many species it is the female that has two different sex chromosomes, not the male).

Having said all of that, there is a caveat that needs to be explained. Namely, the broad definition of male vs female that I have given can run into trouble at the individual level because some individuals are sterile, so by this definition, it seems like they simply shouldn’t have a sex. In reality, we define sex practically based on the physiology that would result in the production of a particular gamete under normal circumstances. This is important, because physiology is rarely binary. There aren’t, for example, two distinct groups of people with regards to metabolism: high and low. Rather, there is a whole spectrum of metabolic activity.

The next thing we need to talk about is genotype vs phenotype. The genotype is what a person is genetically. In other words, what their genes code for, whereas the phenotype is the physical characteristics of the individual. This is important to understand because different genotypes can lead to different phenotypes, but also the phenotype does not always match the genotype. This becomes particularly true when we start talking about epigenetics. An epigenetic effect occurs when something other than genetics affects the expression of the trait. In other words, the phenotype is determined not only by the genotype, but also by the environment, enzymes, etc. and in some cases, those factors can override the genotype.

Sexes in the animal kingdom

With all of that background in place, let’s look at the animal kingdom and see what sort of variation exists for the sexes, because there is a lot we can learn from this broad perspective (I promise I will talk about humans later). Even a cursory knowledge of zoology will quickly tell you that sex is complicated. There are, for example, many species that are hermaphrodites. This means that they simultaneously have the physiology to produce eggs and the physiology to produce sperm. They are not “male” or “female;” they are both.

Many other organisms can switch between the sexes, and in many cases do so obligately (i.e., all individuals start out as one sex and switch later in life). This is one of the places where epigenetics comes in. Anemonefish (aka clown fish) are a good example (Todd et al. 2016). Anemones are inhabited by a male-female pair, where the female is larger and dominant. Individuals start off life as males and pair up with a female, but if that female dies, this causes epigenetic changes in the male, resulting in it changing sexes and becoming a female. Thus, if Finding Nemo was biologically accurate, when Nemo’s mother died, Marlin (his father) should have changed sex and become Marla.

In many other species, individuals do not change sex as adults, but their sex is determined by the environment as they develop. Some (but not all) turtles provide a good example of this (as do crocodilians, some lizards, etc.). They are what we call temperature sex determined (TSD), and the temperature at which the eggs are incubated determines the sex of the offspring. I don’t want to get too technical here (and indeed there are important pieces of information that we don’t have yet), but I do want to briefly walk through some of how this works because it is instructive (see a more detailed overview here: Lance 2009). During early embryonic development, sex has not been determined (this is true in humans as well) and whether an embryo becomes a male or a female depends on the hormones present. Under many conditions, the embryo will develop as a female, and this seems to be largely driven by the hormone estradiol, which is made from testosterone via the enzyme aromatase. At certain temperatures, however, aromatase stops converting testosterone into estradiol, ultimately resulting in the development of male characteristics.

I went through all of that info on TSD because that background knowledge lets us look at some import questions. For example, what happens if we raise eggs at a male-producing temperature, but we supply them with estradiol? The answer is usually that females develop (Lance 2009). In other words, even though temperature usually determines sex we can over-ride that and produce a different sex. Further, the fun doesn’t stop there, because in at least some cases, we can take turtle species that do have sex chromosomes, paint the eggs with estradiol, and get hatchlings with female physiology even if they are genetically male (Freedberg et al. 2006)! In other words, we can make turtles that have male sex chromosomes develop female phenotypes, including the ability to lay fertile eggs. This is why I’ve been arguing that chromosomes sometimes determine sex, but they don’t define it. We can change the sex to be something other than what was determined genetically. To put that another way, even though chromosomes usually determine sex in these species, we can override that and make the estradiol treatment determine sex.

Similarly, there are some lizards that are usually genetically sex determined (i.e., sex is based on chromosomes) but at certain temperatures, there is an epigenetic effect and the temperature overrides the genetics and determines the sex of the hatchlings. In bearded dragons, for example, at high temperatures, animals that are genetically male (based on chromosomes) develop as females and produce fertile offspring (Holleley et al. 2015). So, if you want to insist that chromosomes define sex, rather than determining it (under normal circumstances), then you must claim that these lizards who are running around laying fertile eggs are actually males. This is a notion that any biologist would scoff at because, again, that’s not how we define sex. If you are going to claim that males are laying eggs, then you have invented your own definition of “male” that biologists do not accept.

Finally, you may be wondering, given all this complexity with TSD and chromosomes, can you ever get intermediates? The answer is, yes! There are situations where individuals don’t develop entirely as male or entirely as female and instead end up developing partially as both (Ewert and Nelson 1991), which makes it pretty impossible to maintain a view that sex is binary. In other words, up until this point, you could have tried to make a post hoc change to the original argument and claim that, “there are only two sexes, and it is determined by physiology,” but that doesn’t work, because some individuals have aspects of both male and female physiology.

The point that I’m trying to get at here is that sex is complicated. It is clearly not as simple as a binary state determined strictly by chromosomes, because we know that you can have reproductive “females” who are genetically “males.” We know that there is more to sex than simply the chromosomes. and we know that environmental factors can override the genetics. Now, you may protest to this because I have been using examples from non-human animals, but that counterargument misses the point. The point is that traits are more complicated than a simplistic understanding of genetics would lead you to believe, and there is no reason to think that sex is only complicated in non-human animals. Indeed, as I’ll explain in the rest of the post, sex is extremely complicated in humans. To put that another way, using non-human animals is a good way to get people to lower their biases and look at the evidence, and as you’ll see, the bizarre situations in other animals are highly analogous to what happens in humans.

Sexes in humans

Let’s being by looking just at the sex chromosomes. In humans, you have probably heard that there are two possibilities for sex chromosomes: XX and XY, but that is not correct. In reality, there are many possible combinations, and it’s not that uncommon for someone to have an atypical number or arrangement of sex chromosomes. Indeed, one large study found that 1 out of every 426 people (2.34 out of 1,000) had one of these conditions (Nielsen and Wohlert 1991).

For example, some people get extra X chromosomes. When this is associated with a Y chromosome, it is known as Klinefelter syndrome, and people with it can be XXY, XXXY, or even XXXXY. These unusual genotypes are associated with a combination of male and female phenotypes (with female traits being more prominent when more X chromosomes are present). People with this condition have male genitalia, but they are often have small testes and are sterile or have reduced sperm counts, they have less body hair and often no facial hair, they have lower testosterone levels, and in some cases they develop breasts (Visootsak and Graham 2006). So here, we have people who have two X chromosomes, but also a Y chromosome, breasts but also a penis, testes but low testosterone levels, etc. They simply don’t fit neatly into the discrete boxes of “male” and “female.”

Extra X chromosomes can also occur without the presence of a Y, and you can have someone who is XXX (sometimes called “superfemale”). People with this present mostly as normal female phenotypes, but they are taller on average, and often have learning disabilities (Tartaglia et al. 2010). Things often become more severe when there are four X chromosomes (“tetrasomy X”; XXXX). Some people with this develop normally, but others do not experience normal puberty, don’t develop a normal female phenotype, and are infertile. Beyond this, some individuals actually have a full 5 X chromosomes (XXXXX) and experience even more severe symptoms. Here again, we have atypical chromosome arrangements resulting in different phenotypes.

There can also be unusual numbers of Y chromosomes. For example, some people are XYY. These individuals have mostly normal male phenotypes and are usually fertile. Others may have XYYY or even XYYYY. These conditions are quite rare making it hard to generalize, but behavioral problems such as aggression have been reported in several cases (Abedi et al. 2018).

Additionally, there are XXYY individuals. These individuals are largely similar to XXY individuals, though there are some differences (Tartaglia et al. 2008). Like XXY individuals, they are generally sterile, and have reduced male features (e.g., small testes).

Finally, there is a condition known as an X monosomy (Turner syndrome; XO). This occurs when an individual has a single X chromosome and either no Y or sometimes a partial Y. These individuals appear female, but are generally infertile and do not have properly developed gonads (Fryns and Lukusa 2005). I want to pause here for a second to note that you can get a situation where someone has part of a Y chromosome. So if your definition of sex is based on the presence or absence of a Y, how do you define someone who has part of a Y? Are they only partially male?

By this point, it should be abundantly clear that sex in humans is far more complicated than XX vs XY, and there are lots of genotypes and lots of phenotypes. It should be obvious that chromosomes determine sex rather than defining it, but there are still more layers of complexity that we haven’t gotten to yet. What if I told you, for example, that it is possible to be born with normal female genitalia, even though you have a Y chromosome? This is a condition known as Swyer syndrome, and it’s often a result of a mutation on the SRY region (aka testis-determining factor) of the Y chromosome, but many other genes can cause it as well (Thomas and Conway 2014). These genes often play key roles in activating the right chemicals for an embryo to develop into a male (think back to the turtles earlier for an analogous situation), so when they are modified, those chemicals don’t get produced at the right amounts. As a result, people with Swyer syndrome have a predominantly female phenotype, but instead of having either testicles or ovaries, they have “streak gonads” which are undifferentiated pieces of tissue that can produce neither eggs nor sperm. People with this condition typically don’t go through puberty and require hormone treatments to develop secondary sexual characteristics such as breasts. However, people with this condition can usually carry a child and give birth if an embryo is artificially implanted. I want you to stop and think for a second about just how complex this is. Here we have people who have a Y chromosome, but also have vaginas, don’t have either testes or ovaries, but have all the other female reproductive physiology and can carry a child if implanted with it. The line between male and female is really blurred in this situation.

The inverse of Sewyer syndrome is “XX male syndrome.” This condition produces individuals with typical male genitalia despite the fact that they do not have a Y chromosome. The cause of this is usually a mutation that resulted in the SRY region ending up on an X chromosome (Anik et al. 2013). Much like Sewyer syndrome, individuals with this condition are generally sterile and often have reduced testes.

There are other situations that are even more bizarre. For example, there are documented cases of people developing “ovotestes.” These are gonads that have some of the features of a testis and some of the features of an ovary. This often occurs in people who are XX but have a mutation on the RSPO1 gene (Tomaselli et al. 2011), which results in ambiguous gonad development. Others actually have both an ovary and a testis and were historically referred to as “true hermaphrodites.” This can occur in both XX and XY individuals (though XX is more common) as well as individuals with some of the chromosome abnormalities described earlier. Further, some individuals with this condition are actually fertile and have children (this usual happens when one gonad is developed and the other is an ovotestis; Krob et al 1994). In other words, there are people who are reproducing even though they have both ovarian and testicular tissue (this is more common in mothers but there also people who are fathers despite this condition). You may remember from the beginning of this post that biologists have typically defined sex based on the physiology required for producing sperm vs eggs. So how are we supposed to classify these individuals who have both physiologies?

There are also cases of individuals who are chimeras. In other words, they have two sets of DNA, and in some cases, one of those sets is XX and the other is XY. In some cases, this has little effect on individuals, and they can reproduce, but in other cases, it results in the development of either ovotestes or other odd combinations of gonads as described earlier. Nevertheless, some of these individuals can still reproduce (Verp et al. 1992). To put that another way, there are people who have a Y chromosome, and have testicular tissue, but still produce eggs and give birth. Now, if you are going to insist that things are as simple as, “if you have a Y you are a male,” then you must argue that these people are males, even though they have mostly female phenotypes and give birth. This is, again, not something that any of the biologists I know would accept.

Beyond all of that, we know that there are epigenetic effects at play in sexual development (Gunes et al. 2016). There are, for example, epigenetic effects on the expression of the SRY region. Exactly how this plays out in developmental sex disorders (DSD) is still poorly understood because epigenetics is such a new field, but we know that there are epigenetic effects that influence the development and expression of male and female traits (phenotypes), and as this field expands, it is likely that we are going to discover that sex is even more complicated than we currently realize (we’ll have to wait and see).


As you can hopefully now see, the topic of sex is extremely complicated, and there is far more to it than simply XY = male, XX = female. There is a whole suite of genotypes and phenotypes, including individuals that are XO, XXX, XXXX, XXXXX, XXY, XXXY, XXXXY, XYY, XYYY, XYYYY, and XXYY. Further, there are individuals who are XX yet develop mostly as males, and there are individuals who are XY but develop mostly as females. There are literally people who give birth, despite having a Y chromosome. There are people who have both ovaries and testicles. There are people who only have part of a Y chromosome, etc.

So, if you are going to insist that Y = male, you are going to have to make some bizarre claims. For example, you are going to have to say that XY individuals with an SRY mutation are, in fact, males, despite the fact that they were born with vaginas, lack testicles, and, if implanted with a fertilized egg, can carry a fetus to term. You are literally going to have to say that a male can give birth. Similarly, you are going to have to say that some XX individuals are females, despite the fact that they have mostly male physiology (including penises). Those are, of course, nonsense positions that biologists don’t accept. Biologically, sex is defined by the physiology needed to produce particular gametes (eggs or sperm), not by sex chromosomes, but recent years have shown that this simply is not a binary situation. There are many individuals that have aspects of both male and female physiology, thus making it impossible to use binary categories.

Let me put that another way. Given the existence of individuals with conditions like XXY who have some female traits and some male traits, the existence of individuals who appear female despite being XY, the existence of individuals with both an ovary and a testis, the existence of people who give birth despite having a Y chromosome, etc., which of the following descriptions seems more accurate, “sex is strictly binary; if you have a =Y you are a male, if you don’t you are a female, no exceptions” or “sex is a complex trait with many genotypes and phenotypes as well as epigenetic factors. It is a spectrum of traits and cannot adequately be described using strictly binary categories.” Which of those does a better job of describing the enormous variation that I have discussed in this post?

Again, to be clear, I’m not making any political or social arguments here. What you do with this information and how it affects your views is up to you, but you must accept facts, and the facts clearly show that biologically, sex is more complicated than a simple binary dichotomy.

Rules for commenting on this post

 As explained, this post is solely about the science. If you think I am wrong about the science, feel free to explain, but I do not want the comments to divulge into endless political and social debates. As I said, for the sake of this post, I am just presenting the science. What you do with that is up to you. Comments that are not about biology or that tack political arguments onto biological ones will be deleted. Similarly, if you think I am wrong, please actually explain why rather than just saying, “no, Y = male.” Actually deal with the points I raised and evidence I presented. Also, be civil (see the Comment Rules for my more general policies).

Literature cited

(see this post if you have trouble accessing these for free)

  • Abedi et al. 2018. Rare 48, XYYY syndrome: case report and review of the literature. Clinical Case Reports 6:179–184.
  • Anik et al. 2013. 46,XX Male Disorder of Sexual Development: A Case Report. Journal of Clinical Research and Pediatric Endrocrinology 5:258–260.
  • Ewert and Nelson 1991. Sex determination in turtles: diverse patterns and some possible expliantions. Copeia 1991: 50–69.
  • Freedberg et al. 2006. Long-term sex reversal by oestradiol in amniotes with heteromorphic sex chromosomes. Biology Letters 2
  • Fryns and Lukusa 2005. Monosomies. Encyclopedia of Life Sciences.
  • Gunes et al. 2016. Genetic and epigenetic effects in sex determination. Birth Defects Research Part C Embryo Today Reviews 108:321–336
  • Holleley et al. 2015. Sex reversal triggers the rapid transition from genetic to temperature-dependent sex. Nature 523: 79–82.
  • Krob et al 1994. True hermaphroditism: Geographical distribution, clinical findings, chromosomes and gonadal histology. European Journal of Pediatrics 153:2–10
  •  Lance 2009. Is regulation of aromatase expression in reptiles the key to understanding temperature-dependent sex determination? Journal of Experimental Zoology 311:314–322.
  • Nielsen and Wohlert 1991. Chromosome abnormalities found among 34,910 newborn children: results from a 13-year incidence study in Arhus, Denmark. Human Genetics 87:81–83.
  • Tartaglia et al. 2008. A new look at XXYY syndrome: Medical and psychological features. American Journal of Medical Genetics A. 146A:1509–1522
  • Tartaglia et al. 2010. A review of trisomy X (47, XXX). Orphanet Journal of Rare Diseases 5
  • Thomas and Conway 2014. Swyer syndrome. Current Opinion in Endocrinology & Diabetes and Obesity 21:504–510.
  • Todd et al. 2016. Bending genders: The biology of natural sex change in fish. Sexual Development 10.
  • Tomaselli et al. 2011. Human RSPO1/R-spondin1 Is Expressed during Early Ovary Development and Augments β-Catenin Signaling. PLoS One 6:e16366
  • Verp et al. 1992. Chimerism as the etiology of a 46,XX/46,XY fertile true hermaphrodite. Fertility and Sterility 57:346–349
  • Visootsak and Graham 2006. Klinefelter syndrome and other sex chromosomal aneuploidies. Orphanet Journal of Rare Diseases 1.
Posted in Uncategorized | Tagged | 29 Comments

The overwhelming consensus on climate change

The climate is changing, and we are the primary cause. These are simple facts that are supported by a vast body of evidence and agreed upon by virtually all experts. Nevertheless, many people continue to think that the science isn’t “settled” and there is widespread disagreement among experts. Unfortunately, these myths have been propagated and supported by very active misinformation campaigns, so I want to take a few minutes to explain why they are incorrect. First I will explain what we mean when we say that a topic is “settled” or that there is a “consensus,” then I will demonstrate that such a consensus exists for the topic of anthropogenic climate change.

“Settled” science

 First, I need to explain what I mean by “settled science,” because there are many people who argue adamantly that science is never “settled” because it is always possible that some future discovery will overturn the current thinking. That is technically true, but it can be misleading and requires clarification.

It is true that science, by its very nature, does not provide “proof.” Rather, science shows us what is most likely true given the current evidence. So to that extent, it is true that science is never 100% “settled,” because it is always technically possible there is something we have missed. However, there is a huge difference between a technical possibility and practical doubt. For example, it is technically possible that we are wrong about smoking causing cancer. It is technically possible that all of the countless studies on smoking and cancer are wrong and smoking is actually safe or even beneficial. Further, you can even find a handful of doctors that argue that we are wrong about smoking causing cancer. Does that mean that the science isn’t “settled” or that there is serious debate on the topic? Of course not! The topic has been so well studied so many times by so many people that the odds that we are wrong are insanely low. They are so low that for all intents and purposes, we can treat them as if they are zero. The notion that smoking causes cancer is “settled” in the sense that it is supported by such a massive and consistent body of evidence that it is extraordinarily unlikely that it is wrong, and we must act as if it is correct until such time as compelling evidence arises to the contrary.

This is true for a very large number of scientific topics. There are many things that have been so thoroughly studied that they are as close to “settled” as science can possibly come, and it does not make sense to talk about them as if there is any practical doubt. It would be absurd, for example, for a politician to say, “science is never settled, therefore we can’t really be sure that smoking causes cancer.” The link between smoking and cancer is “settled” in the sense that it is supported by such a vast body of evidence that it is extraordinarily unlikely that we are wrong about it. As I will demonstrate, the same is true for climate change.

See these posts for more information about “settled science.”

Note: please read to the end before arguing that there used to be consensus that smoking was safe. That is a myth and I deal with it below.

What is a scientific consensus?

 There are really two different levels at which we can talk about a consensus, and this can become confusing because most people are bad at specifying the level at which they are talking (I have been guilty of this myself). At one level, there is a consensus of experts. In other words, this exists when the vast majority of experts agree on something. This is what most people think of when they think of a scientific consensus, but it is not actually the best level to look at.

You see, when we say something like, “this is a fact” or “the science is settled,” we aren’t basing that on a consensus of experts, but rather a consensus of evidence (i.e., a large body of studies that all agree with and support each other). The consensus of experts is a secondary by-product of the consistent body of evidence. This is really the level we need to look at when asking questions like, “is there any serious debate on topic X.” Science is not a democracy. It is about evidence, not authority. So simply finding some people with advanced degrees who disagree with X does not mean that there is serious scientific debate about the topic. Rather, if there is serious debate, it will be reflected in the peer-reviewed literature, because people will be publishing papers presenting evidence that X is not correct. So that is really the level we should focus on when we talk about a scientific consensus: the evidence, not the experts.

Having said that, there is value in having a consensus of experts when it comes to the general public. No one can be an expert on everything, so even though we would ideally always look for a consensus of evidence, there are many topics on which a given individual simply is not equipped to do that (this is true for everyone, myself included). So, when encountering those topics, it makes sense to look for a consensus of experts, because, on average, an expert will know more about the topic of their expertise than a layperson will, and a consensus of experts usually reflects a consensus of evidence.

I do want to pause here for a second to emphasize the “on average” bit of my last statement. On pretty much any topic, you can cherry-pick an expert who holds an extreme position. You can find doctors who think HIV doesn’t cause AIDs, immunologist who think vaccines are dangerous, etc. That does not mean that there is not a consensus of experts on those topics. You won’t find 100% agreement among experts on just about any topic. So that is not the standard by which we assess a consensus of experts. Further, the fact that you found an expert who agrees with you absolutely does not mean that your position is legitimate or scientifically valid. It is always possible to cherry-pick experts who agree with you, and it is imperative that you avoid falling into that trap. If you go to 100 doctors and all but one of them says you have cancer, you shouldn’t trust the one who disagrees and proclaim that doctors just aren’t sure about your diagnosis. It is intuitively obvious that you should listen to the 99 who said you have cancer.

See this post for the difference between deferring to experts and appealing to authority.

 See this post for more information on why you should avoid cherry-picking experts, including discussions of some of the handful of climatologists who deny anthropogenic climate change.

The consensus on climate change

 With the semantics now out of the way, let’s look at climate change. The most famous (or infamous depending on your point of view) study looking at the consensus on climate change is the Cook et al. 2013 study that produced the 97% statistic that we have no doubt all heard. This paper has been widely criticized by the good people of the internet, mostly for invalid reasons. Nevertheless, there are some issues that are worth talking about (I have talked about it at length here, so I’ll be brief).

In short, this study did not actually look at a consensus of experts, but rather looked at a consensus of evidence (studies), from which a consensus of experts (authors) was inferred. It took 11,944 papers (by 29,083 authors) and scored them based on whether or not their abstracts stated agreement, uncertainty, or disagreement with the notion that humans are causing climate change. They found that of the papers that made an explicit statement about humans causing climate change (agree, uncertain, or disagree), 97% agreed that we are causing climate change.

That seems like pretty straight forward and compelling evidence of a strong consensus of evidence, so why the hoopla over this study? It mostly comes from the large number of papers that did not explicitly express a view one way or the other. 62.5% did not express an opinion, so the 97% agreement figure comes only from the subset of papers that did express an opinion, and that fact has drawn criticism from both sides.

On one hand, climate change deniers often erroneously claim that Cook et al. “threw out” nearly two thirds of the studies, and actually only 36.9% of the papers agreed that we are causing climate change. This is a faulty argument because it tries to treat papers that did not express a view as if they expressed uncertainty or disagreement, which is clearly false. If they did not express a view, then we cannot draw any conclusions about whether or not the authors agree with anthropogenic climate change, therefore they must be removed before calculating percentages. If we translate this into a more standard survey, if you sent a survey to 10,000 people asking if they thought smoking causes cancer, and 1% said it does not, 49% said it does, and 50% didn’t respond, you would not conclude that only 49% of people think smoking causes cancer, because that would assume that the 50% who didn’t respond are all either uncertain or think that smoking does not causes cancer, which is clearly absurd. It’s the same thing with the Cook et al. study. You can’t conflate “did not express an opinion” with “does not have an opinion” or “disagrees with the consensus,” yet that is exactly what this criticism does.

Further, a lack of explicit statement is precisely what we’d expect on a topic that has reached a consensus. Most studies on vaccines, for example don’t include a statement of “vaccines are safe” in their abstract, because that is so well-established that there is no need to explicitly state it (abstracts have tight word limits). Nevertheless, by the criteria used in this study, we would relegate those studies to the “did not express a view” category, even though the authors almost certainly think that vaccines are safe. Indeed, Cook et al. actually found that the number of explicit statements decreased over time, which is exactly what we expect from a growing consensus.

This does, however, lead to the other criticism. Namely, that Cook et al. actually severely underestimated the consensus because most of the papers in the “did not express a view” category probably were by people who do actually accept anthropogenic climate change. This is actually a fair criticism (though it is often stated too forcefully and unnecessarily denigrates the work of Cook et al.). An awful lot of those studies were on the impacts of climate change, and it is quite a stretch to think that most of those authors disagree with fundamental scientific facts about climate change. Indeed, many of those papers were by authors who are known to agree with anthropogenic climate change, and some of their other papers were categorized into the “accepts the consensus” category. So, it does seem extremely likely that Cook et al. underestimated the consensus.

To solve this problem, James L Powell took a different approach. He argued (I think correctly) that if there is actually disagreement on a topic, that disagreement should be prominent in the literature. It should be easy to find papers that explicitly reject anthropogenic climate change, whereas if there is a consensus, most papers simply won’t make an explicit statement one way or the other (just as they don’t for most “settled” topics). Therefore, if we want to look for a consensus, we should simply count the number of papers that explicitly reject anthropogenic climate change. As an example, he cited 500 recent studies on plate tectonics, none of which either explicitly endorsed or rejected the theory. Based on the Cook et al. criteria, this would erroneously lead to the conclusion of no consensus, whereas based on the criteria of explicit rejection, we would correctly conclude that there is a strong consensus.

scientific consensus on global climate change, global warming

Image via Powell

When we apply this rejection criteria to the climate change literature, we find almost no studies that argue against the position that humans are causing climate change. For example, Oreskes (2004) reviewed 928 papers published between 1993 and 2003 and failed to find a single one that rejected anthropogenic climate change. Similarly, Powell has looked at this at several time points, always with the conclusion of a very strong consensus. For example, he examined 13,950 articles published from 1991 to November 2012, and only found 24 that rejected anthropogenic global warming. That’s a 99.83% agreement among studies. He later followed that up by looking at the 2,258 climate change papers published from November 2012 to December 2013. This only revealed 1 paper that rejected anthropogenic climate change (a 99.96% consensus). Admittedly, neither of those were published in peer-reviewed journals (but you are welcome to replicate his results), but a subsequent analysis of papers in 2013 to 2014 was peer-reviewed. In it, he examined 24,210 papers by 69,406 authors, and found a grand total of 5 articles published by 4 scientists that rejected the notion of anthropogenic climate change (Powell 2015). That gives us a consensus of evidence (studies) of 99.98%, and a consensus of experts of 99.99%. To put that another way, for every 1 publishing climatologist who disagrees with anthropogenic climate change, there are 9,999 who agree with it. That’s a pretty extraordinary consensus of experts.

Powell also examined the same papers used in Cook et al. 2013, only 24 of which rejected anthropogenic climate change (99.78% agreement; Powell 2016; many of these were the same papers in his non-peer-reviewed analysis). Finally, part way through last year (2019), he examined all of the studies on climate change that had been published so far that year (11,602), and not a single one rejected anthropogenic global warming (Powell 2019).

These surveys of the literature are extremely compelling evidence that a consensus has been reached and the topic is “settled.” If there was actually serious debate, if actual evidence existed discrediting anthropogenic global warming, we would see that in the literature. We would see numerous studies publishing evidence against anthropogenic climate change, but we don’t see those studies because that evidence doesn’t exist. All of the available data very clearly shows that we are causing climate change. The scientific consensus on this topic is truly overwhelming. Nevertheless, I am sure many people are preparing to fire off responses, so I want to spend the rest of this post preemptively dealing with them.

“But what about those petitions/letters where thousands of scientists said we aren’t causing climate change?”

There have been many attempts to discredit the consensus of experts by accruing lists of signatures, but if you examine those lists, their fraudulent behavior becomes apparent. Probably the most famous is the “Oregon Petition” which (depending on the source commenting on it) received signatures from 16,000, 30,000, 31,000 or 32,000 scientists. That sounds impressive, until you do even a modicum of fact checking, at which point you’ll realize that this petition is a fraud.

First, there was virtually no verification process. As a result, there were lots of fake signatures, including celebrities and fictional characters. Further, even for the signatures that were real, the only requirement was a B.Sc. in science, which hardly makes someone a scientist, and certainly doesn’t make them an expert on climate change. A huge portion of people who get undergraduate degrees in science never actually use their degrees. Further, even for those who went on to obtain additional degrees and pursue careers in science-related fields, many were experts in totally unrelated fields. For example, how does an orthopedic surgeon, veterinarian, or mechanical engineer qualify as an expert on climate change? When you cut through all the crap, you are left with only 39 people who actually have relevant degrees and expertise in climatology. That is hardly an impressive number and certainly doesn’t discredit the notion of a consensus of experts (more details here, here, and here).

A more recent attempt was a letter to the UN, supposedly signed by 500 scientists, arguing that there is no climate emergency. Once again, however, when you start looking at the signatures, most weren’t even scientists, let alone climatologists. Further, many of them had conflicts of interest, and the claims made in the letter aren’t supported by actual scientific evidence (more details here and here).

There is also another more fundamental issue with these attempts to discredit the science. Namely, they are only about the consensus of experts, not the consensus of evidence. Science is not a democracy, and even if there were hundreds of climatologists who rejected climate change, that would be irrelevant unless they actually had data to back up their position, which they don’t.

“But what about the list of 500 studies showing that climate change isn’t happening/is natural?”

Science deniers love lists of studies. I have, for example, written extensively about the lists anti-vaccers have assembled. The problem is that these lists are inevitably assembled without an actual understanding of the science, and when you look at the papers, they don’t say what the science-deniers think they say. For example, when I went through anti-vaccers’ lists of 160 studies that supposedly showed that vaccines cause autism, I found that 33 of their studies weren’t about autism, 82 weren’t about vaccines, multiple studies explicitly stated that vaccines don’t cause autism, and only 13 were actual human trials arguing that vaccines caused autism (all of which were riddled with problems). The exact same thing has been true of every list of papers I have ever been shown that supposedly discredits anthropogenic climate change. The lists are inevitably filled with papers that talk about regional trends (not global), talk about past climates (without addressing the current warming), talk about how natural climate forcing work (without discussing the current warming), etc. As with the anti-vaccine lists, nearly all of them are misrepresented, most are irrelevant, and many actually argue the opposite of what science-deniers are claiming.

“But what about the few studies that do actually argue against anthropogenic climate change?”

At this point, you might be thinking, “fine, most climatologists agree, and very few studies disagree, but there are a few studies that disagree, and in science, any position can be overthrown by new evidence, so what about those studies?” This is a fair question given two caveats: first, we always have to examine all studies in the context of the broader literature. Given that the context in this case is literally thousands of studies with numerous lines of evidence showing that we are the cause, the evidence in the dissenting studies had better be pretty good.

This brings me to the second point. We always have to critically examine studies rather than assuming that they are valid, and when we do that, we find that these studies used weak designs, shoddy statistics, and are full of problems (Benestad et al. 2016). So they do not in any way discredit the overwhelming mountain of evidence.

“But scientists are just following the ‘dogma’ of their field”

This well-worn trope argues that lots of scientists actually have evidence against anthropogenic climate change, they just don’t publish it because in science it is supposedly forbidden to go against the “dogma” of your field. This is one of those fundamental misunderstandings of science that just will not die. Science is extremely adversarial. We love to prove each other wrong. Further, every scientist who was ever considered great was great precisely because they discredited the views of their day. No one gets anywhere in science by blindly going with the “dogma” of their fields. If anyone actually had compelling evidence that we weren’t causing climate change, they would publish in a high-ranking journal and collect their Nobel Prize. No one has done that precisely because those data don’t exist.

“But scientists have been wrong before”

This is another trope that I have dealt with many times before, so I’ll be brief. First, there are few (if any) examples where modern science has been wrong about something with the same level of evidence that we have for climate change. The “evidence” that was used for things like the sun orbiting the earth is not even remotely comparable to the evidence for climate change.

Second, past mistakes do not automatically negate the evidence for climate change. If it did, then you could use it any time that you wanted to discredit any scientific study. “You think that smoking causes cancer? Well science has been wrong before, so I don’t have to accept that.” See how stupid that is? You need actual evidence to discredit climate science.

Third, this argument is inherently self-contradictory, because it is only through science that we know that previous scientists have been wrong, but based on this argument, we can’t trust science. Therefore, we have no more reason to trust the evidence that the earth moves around the sun than we do for the discredited evidence that the sun moves around the earth. In other words, if the fact that scientists have been wrong before means that we can’t trust scientific discoveries, then we can’t trust the scientific discoveries that were used to show that scientists had been wrong before. It’s a paradox.

See these posts for more details

 Also read this post before arguing that “most scientific studies are actually wrong”

“But there used to be a consensus that smoking was safe”

This is just a special case of the “science has been wrong before” argument. Further, it’s not even true. Tobacco companies certainly ran a good misinformation campaign (much as fossil fuel companies do today), but actual scientific studies have consistently shown that smoking is dangerous. Indeed, scientists suggested that smoking was dangerous way back in the early 1900’s, and essentially all of the research since then (minus a few industry-driven papers) confirmed their suspicions (you can find an overview of this history in Proctor 2012).

“But in the 70s there was a scientific consensus on global cooling”

No, there wasn’t. There was certainly media hype about this, but it was never a prominent scientific position. Indeed, there were a grand total of 7 papers on it, compared to 42 during the same time span that argued that we were causing global warming (Peterson et al. 2008).

“But scientists don’t agree about the extent to which we are causing climate change”

This is a very common tactic among science deniers: taking a minor disagreement and conflating it with a major one. There is some disagreement among analyses about exactly how much we are contributing to climate change, but they all agree that the majority of the change is being caused by us. There is no serious disagreement that we are the primary cause. If there was, this would, once again, be easy to find in the literature, but good luck finding many studies that argue that we are only playing a minor role. They are virtually non-existent.


There are tons of other invalid counterarguments that I’m sure I’ll get assaulted with, but I have already addressed most of them in previous posts so please read them before making inane comments. Also, if you want to more information about why simply looking for papers that reject climate change is a good approach for testing a consensus, read Powell’s papers (cited at the end) as they explain things in much greater detail.

  • This post covers most common arguments and counter points.
  • This one explains the evidence that makes us so certain that the current warming is not natural and is being caused by us
  • This one goes over the evidence that climate change is already having serious consequences
  • This one debunks the absurd notion that scientists are just in it for money
  • This one talks about Cook et al. 2013 in more detail and discusses other attempts to estimate a consensus


In short, there is an overwhelming consensus that we are causing climate change. This consensus exists both among studies and among scientists. Indeed, recent estimates put it at over 99.9% agreement that we are causing the climate to change. Thousands of studies have confirmed that we are the cause, and virtually none argue that we aren’t. Further, the handful of contrarian studies are riddled with problems and are easily debunked. Every shred of evidence confirms that we are causing climate change, and there is no serious debate among experts.

This level of consensus is important, because it means that there is no valid reason for doubting the reality that we are causing climate change. The level of consistent evidence for it is on par with the evidence for things like smoking causing cancer. Both topics have been extremely thoroughly studied, both topics have a huge and remarkably consistent body of evidence (i.e., a consensus of evidence), and for both topics, that body of evidence has resulted in a nearly unanimous consensus among relevant experts. Nevertheless, on both topics, it is possible to cherry-pick studies and experts that disagree with the consensus, but doing so is folly! As I explained, it is always best to look at the evidence itself, but for most people, that’s not possible, in which case it is rational to simply listen to experts, but why would you choose to listen to the 0.01% of experts who disagree with the consensus? You wouldn’t do that on a topic like smoking causing cancer, so why would you do that with climate change? If 9,999 doctors diagnosed you with cancer and told you to immediately start treatment, but 1 doctor told you that you had nothing to worry about, would you blindly follow that one doctor? I highly doubt it, so why would you do that with climate change? Why would you listen to the 1 scientist saying we aren’t causing it rather than the 9,999 who are saying that we are causing it and need to change our actions?

Related posts 

 Literature Cited

  • Benestad et al. 2016. Learning from mistakes in climate research. Theoretical and Applied Climatology 126:699–703.
  • Cook et al. 2013. Quantifying the consensus on anthropogenic global warming in the scientific literature. Environmental Research Letters 8:024024
  • Oreskes. 2004. The scientific consensus on climate change. Science 306: 1686.
  • Peterson et al. 2008. The myth of the 1970s global cooling scientific consensus. Bulletin of the American Meteorological Society 89:1325–1337.
  • Powell 2015. Climate scientists virtually unanimous: Anthropogenic global warming is true. Bulletin of Science, Technology & Society 35:121-124.
  • Powell 2016. The consensus on anthropogenic global warming matters. Bulletin of Science, Technology & Society 36:157-163.
  • Powell 2019. Scientists reach 100% consensus on anthropogenic global warming. Bulletin of Science, Technology & Society
  • Proctor 2012. The history of the discovery of the cigarette-lung cancer link: evidentiary traditions, corporate denial, global toll. Tobacco Control 21: 87-91
Posted in Global Warming, Nature of Science | Tagged , , , , | 75 Comments