Science never showed that smoking was safe

Posted on January 8, 2026 by Fallacy Man

“Science used to say that smoking was safe, so why should we ‘trust the science’ when it says that vaccines are safe and effective, climate change is real, or GMOs are safe?”

This is one of the most common excuses I hear for rejecting modern scientific evidence. People throw this argument around like a “get out of jail free” card for blindly dismissing any studies they don’t like. The argument is, however, fundamentally flawed in every way. It suffers all the usual problems with “science has been wrong before” arguments, and its core claim is simply false. There was never a scientific consensus that smoking was safe, and the true history of the science of tobacco actually reveals just how important it is to “trust the science” and why anecdotes, appeals to antiquity, and appeals to nature don’t work.

Note: This post pulled heavily from Doll’s (1998) review of the history of tobacco, and I highly recommend reading that paper in full.

Traditional tobacco use

Smoking tobacco is an ancient practice. Archeological evidence shows that people have used tobacco for at least 2500 years. It was used by the Mayans for religious ceremonies, medicine, and recreation, and those uses spread throughout the Americas. This is important to realize because there are many today who argue that natural remedies must be safe and effective by simple virtue of the fact that they are “natural” (this is the “appeal to nature fallacy”). Tobacco is, however, completely natural, yet it is very harmful to human health. So, if appeals to nature were reliable, then smoking should be safe.

Similarly, the supposed medical benefits of smoking tobacco were supported by countless anecdotes. Generation after generation saw examples of people getting better after smoking tobacco, so surely it must work, right? Wrong! Anecdotes are not evidence of causation, and coincidences and placebo effects do happen (saying “I took X then got better, therefore X works” is a post hoc ergo propter hoc fallacy)

Likewise, I frequently encounter people who criticize “western” medicine and insist that indigenous peoples have “other ways of knowing” that are superior to science. Here again, if that was true, and we should rely on traditional knowledge instead of modern science, then we’d have to conclude that smoking is not only safe, but actually medically beneficial!

Don’t get me wrong, indigenous peoples often do have a wealth of knowledge about the environment, and their knowledge provides excellent starting points for scientific research. I am all for scientists collaborating with indigenous groups and working with them. Those relationships can be extremely valuable, but at the end of the day, anything about the physical world that is actually true has to be able to pass scientific testing. The fact that something has been used for hundreds or thousands of years simply is not sufficient.

This understanding is important for things like acupuncture, cupping, and a host of other “alternative medicines.” I constantly hear people say things like, “well acupuncture has been used for thousands of years, so surely it must work. Why else would they have used it for so long?” The history of smoking clearly shows why that thinking doesn’t work. The fact that something has been used for a long time does not in any way shape or form indicate that it works (this is known as an appeal to tradition fallacy).

In short, we developed science precisely because anecdotes, appeals to nature, and appeals to antiquity don’t work. Science is, hands down, the most reliable tool we have ever invented for understanding the natural world.

Scientific history

When Europeans arrived in the Americas, they quickly adopted tobacco use and brought it back to Europe with them, where it was once again used for pleasure as well as being used to treat a range of conditions including asthma, coughing, cramps, parasites, tumors, and various other conditions. Here again, it enjoyed a long period of use and countless anecdotes.

By the early 1900s, people had largely (although not entirely) abandoned the idea that it was medicinal, but its recreational popularity was still in full swing. There was some pushback, but those efforts weren’t based on any sound science. Indeed, over the next 50 years, little serious work was conducted on tobacco use. There were various small studies that suggested possible harm, but no truly compelling evidence emerged, and most (if not all) of those studies would not have passed modern peer review. Remember, medical science was still in its infancy, and many of the tools and experimental standards we rely on today had not been invented or established yet (this is another important point that the “science has been wrong before” crowd often ignore).

In 1950, things changed with the publication of five moderate-sized case-control studies that all showed that smoking was associated with cancer (Schrek et al. 1950; Levin et al. 1950; Mills and Porter 1950; Wynder and Graham 1950; Doll and Hill 1950). Importantly, however, case-control studies are only suggestive and cannot establish causation. Thus, their publication indicated that a potential link between smoking and cancer needed to be studied more closely, but they could not actually demonstrate that smoking was the cause of cancer.

Other studies quickly followed, including two large, prospective cohort studies (Doll and Hill 1954; Hammond and Horn 1954) which also showed that smoking was strongly associated with cancer. Cohort studies are more powerful than case-control studies and can strongly suggest causation, though they are not as conclusive as randomized controlled trials. These results caused serious debate among scientists (as is good and healthy in science), and not everyone was immediately convinced, but studies kept pouring in, including research demonstrating that tobacco causes cancer in rats and isolation of known carcinogens in tobacco smoke.

Taken together, all of these studies soon built a compelling case, and by the end of the 1950s, health agencies around the world were acknowledging that smoking causes cancer. This is a beautiful demonstration of how science should work: initial evidence suggested harm was present, and scientists, being a skeptical bunch, wanted more conclusive evidence before reaching a verdict. So more, larger, high-quality studies were done until a consensus of evidence was reached, at which point, a consensus of experts emerged and health agencies updated their conclusions and recommendations.

Critically, at no point was there compelling evidence that smoking was safe, nor was there ever a consensus of experts that it was safe. There was a long period where it was unstudied, during which many doctors smoked (as did the majority of people), but there was never a large body of evidence showing that smoking was safe, and there is a world of difference between saying, “many scientists and doctors use an untested product” and “this product has been extensively tested and scientists and doctors agree that, based on that evidence, it is safe and effective.” See the difference?

So, if you are arguing that we should not trust scientific conclusions today because science once concluded that smoking was safe, then you are wrong both logically and historically. Even before the 1950s, what little scientific evidence we had mostly suggested that smoking was dangerous, and once proper evidence emerged, scientists adopted it pretty quickly. Also, remember again that, in contrast, non-scientific approaches (other “ways of knowing”) brought us thousands of years of thinking smoking was not only safe, but beneficial.

“But what about those ads with doctors smoking?”

I don’t know how to tell you this, but companies lie and mislead.

Tobacco companies never actually had a consensus of experts, and they certainly didn’t have a consensus of evidence. They did, however, have a very good advertising campaign and invested a great deal of money in trying to mislead the public.

Admittedly, they did also attempt to corrupt the science. This included paying prominent scientists to testify on their behalf, funding studies showing smoking was safe, and targeting doctors at conferences and their practices. So, they did have some scientists and doctors supporting them, especially prior to the 50s when the evidence was far from conclusive. However, the studies that they funded were dwarfed by the much larger, more robust studies discussed earlier, and they were never able to purchase a consensus (see Jackler and Samji [2012] for more details about companies’ attempts to pollute the scientific literature and deceive the public).

Critically, while their campaign was very successful among the general public, and they did make a bit of a mess in the literature, they were completely unable to stop the large studies described earlier from coming out. They never managed to buy or influence a majority of scientists, they never had more than a few shoddy studies, and they were unable to suppress the truth. Once large, high-quality studies started coming out, their influence on science essentially disappeared (though they kept trying).

So given that tobacco companies where ultimately unable to block scientific progress, how does that compare with the modern science on topics like vaccines or GMOs? Several things have changed that actually make it even harder for companies to buy a consensus today.

For one thing, all modern journals require authors to declare their sources of funding and any potential conflicts of interest (including not just research funding for the current work, but also past research funding, fees for speaking engagements, etc.). Lying about conflicts of interest is a serious offense in modern science that could easily cost a researcher their career, and universities keep track of all their scientists’ funding, so an audit could easily catch the lies. None of this was true decades ago, and tobacco company-sponsored research generally did not declare who was funding it or how the researchers might be influenced. Thus, today’s science is much more transparent, and we can (and should) look at conflicts of interest in papers. They can bias results and should be taken seriously. That said, you need to actually look for conflicts of interest (rather than assuming they exist), and they should make you more cautious, but depending on the nature of the conflict, generally are not grounds for automatic dismissal of the paper.

Second, the scientific world has grown massively since the early to mid-20th century. Unlike tobacco where the debate centered around a handful of fairly small studies, topics like vaccines have thousands of studies conducted by tens of thousands of scientists from all over the world. The number of institutions and scientists working today make it unthinkable that a company could buy a consensus. In the modern world, hundreds of papers are published daily. It’s just not possible for a company to fundamentally shift that volume of literature.

Likewise, the strength of evidence presented in today’s studies massively overshadows the evidence for or against smoking available prior to the 1960s. Look at the studies I presented in this post on vaccines and autism, for example. We have multiple massive case-controlled studies, multiple cohort studies with over 100,000 children, and even a meta-analysis with over 1.2 million children. Tobacco companies were never able to get anything even approaching the type of agreement or strength of evidence.

To be 100% clear, I totally agree that modern companies are corrupt and would happily lie to the public and fund favorable studies, but topics like vaccines and climate change are supported by such a massive volume of high-quality, international research (much of which had no conflicts of interest) that it is unthinkable that companies have secretly bought nearly all the world’s scientists and thoroughly corrupted the entire body of literature. You would need some extremely strong evidence of such extensive corruption, and no such evidence exists. Further, if you think that all science is “sold to the highest bidder,” then why did tobacco companies ultimately fail? If they could not buy a consensus at a time when there were far fewer scientists and ethical standards and transparency were far looser, then why would you think that modern companies have succeeded?

Finally, I want to reiterate one of my most frequent points on this blog: do not cherry-pick studies or experts. On literally any well-studied topic, you can find a handful of outlier studies and experts. Don’t blindly believe them. Be critical and look at the entire body of evidence.

Conclusion

The true history of smoking provides a compelling example of why science is so important and why we should trust scientific results rather than anecdotes, appeals to antiquity, and appeals to nature. The latter brought us thousands of years of thinking smoking was safe and even beneficial. Science was the thing that finally showed us it was dangerous. Further, the scientific evidence was consistent from the outset. There was never a consensus of experts or evidence that smoking was safe. Yes, there was a period where most doctors and scientists smoked due to a lack of evidence regarding safety, but a lack of evidence either way is a very different thing from a large body of evidence showing that smoking is safe. Similarly, yes, tobacco companies tried very hard to deceive the public, buy off scientists, and prevent the dangers of smoking from being known, but they ultimately failed. They were never able to buy a scientific consensus. In short, tobacco companies had a good ad campaign, not a consensus, and once solid evidence of the harm from smoking started to emerge, they were powerless to prevent studies from being published. Scientists debated the evidence and kept conducting studies until a clear consensus of evidence emerged. That’s how science works and why it is so important to follow scientific evidence.

Related posts

Literature cited 

  • Doll. 1998. Uncovering the effects of smoking: historical perspective. Statistical Methods in Medical Research 7:87-117
  • Doll and Hill. 1950. Smoking and carcinoma of the lung. Preliminary report. British Medical Journal 2: 739-748.
  • Doll and Hill. 1954. The mortality of doctors in relation to their smoking habits. A preliminary report. British Medical Journal 1:1451-1455.
  • Jackler and Samji. 2012. The Price Paid: Manipulation of Otolaryngologists by the Tobacco Industry to Obfuscate the Emerging Truth That Smoking Causes Cancer. Laryngoscope 122:75-87
  • Hammond and Horn. 1954. The relationship between human smoking habits and death rates: a follow-up study of 187,766 men. Journal of the American Medical Association 155: 1316-1328.
  • Levin et al. 1950. Cancer and tobacco smoking. A preliminary report. Journal of the American Medical Association 143: 336-338.
  • Mills and Porter. 1950. Tobacco smoking habits and cancer of the mouth and respiratory system. Cancer Research 10:539-542.
  • Schrek et al. 1950. Tobacco smoking as an etiologic factor in disease. Cancer Research 10: 49-58.
  • Wynder and Graham 1950. Tobacco smoking as a possible etiologic factor in bronchogenic carcinoma. Journal of the American Medical Association 143: 329-336.
Posted in Nature of Science | Tagged , , , , | 1 Comment

Climate change is undeniable: Evidence from satellites

Posted on December 30, 2025 by Fallacy Man

climate change global warming satellite data smoking gunThe evidence for anthropogenic climate change is overwhelming, but the inherent complexity of the climate can make it difficult to communicate the science to the public. The basic concept is simple enough (CO2 traps heat, we have increased the CO2 in the atmosphere, therefore more heat is being trapped), but the details quickly get convoluted and conversations get bogged down in details of climate models, forcings, feedback loops, etc. So in the post, I want to talk about a really key piece of evidence that is, in my opinion, very straightforward and easy to understand and also extremely compelling. Namely, the results of satellites measuring heat leaving earth.

To set the stage for this, let’s quickly review the basic facts. Energy enters the earth as high energy (short wavelength) radiation from the sun. Some of that energy is lost as it passes through our atmosphere and reflects off the earth’s surface. That reflected radiation is then lower energy (long wavelength) heat that goes back into space. CO2, methane, and other greenhouse gases absorb some of that long wavelength energy, temporarily trapping it in our atmosphere, just as the glass of a car windows traps heat in your car. All of this is basic science that we have understood since the late 1800s. Everything so far is universally agreed on.

Changes in CO2 levels subsequently result in changes in the earth’s climate, and we can clearly see from studying past climates that CO2 increases result in increased temperatures (note: it is not true that the temperature increases lag behind the CO2; details and sources here).

We also know that we have greatly increased the levels of CO2 and some other greenhouse gases in the atmosphere. This led to the simple prediction that our greenhouse gases would cause the planet to warm (just as past natural increases in CO2 have), and, of course, our planet is warming. Scientists have studied this by carefully examining natural drivers of climate change, and none of them can explain the current warming (Foster and Rahmstorf 2011; Meehl, et al. 2004; Wild et al. 2007; Lockwood and Frohlich 2007, 2008; Lean and Rind 2008; Imbers et al. 2014). However, when we plug the known physical effects of our greenhouse gas emissions into the calculations, we get a very good fit between observations and what we expect based on the physics (Hansen et al. 2005; Santer et al. 2012; Stips et al. 2016; Stott et al. 2001; Meehl et al. 2004; Allen et al. 2006; Lean and Rind 2008; Imbers et al. 2014).

Note that we know the current CO2 increase is from us due to the isotope ratios. Details and sources here.

All of this is already very compelling, but modern satellites provide another line of evidence that is a metaphorical smoking gun. We have used them to measure the outgoing long wavelength radiation, and that radiation shows a clear and undeniable signature of our effects on the planet.

Science fundamentally works by making and testing predictions, and we judge the strength of an idea by how well it predicts observations and experimental outcomes. So let’s look at the predictions that were made prior to analyzing the satellite data and see how they hold up.

If our understanding of climate change is correct, then when we look at the satellite records we should see:

  1. Energy entering the earth remains relatively stable, while the outgoing long wavelength energy decreases (because it is being trapped by greenhouse gases)
  2. The decrease should be specifically at the wavelengths that our gases trap (thus indicating that it is our gases doing the trapping)
  3. The rate of decrease should match the models that are based on the physics of how greenhouse gases trap heat.

These are really strong, falsifiable predictions. If it turned out that outgoing radiation was not decreasing (prediction 1), that would sink climate change. That prediction has to come true for the notion of our gases causing climate change to be true. Likewise, if radiation decreased, but it wasn’t at the frequencies that greenhouse gases trap (prediction 2), that would show that our greenhouse gases weren’t the cause of the reduction in radiation. Finally, if predictions 1 and 2 came true, but the rates of decrease were way off, it would suggest that the fundamental concepts of anthropogenic climate change might be right, but our understanding of the physics is way off and, therefore, our predictions are unreliable.

In contrast, no other driver of climate change makes these three predictions. There is no other reason to expect outgoing radiation to decrease, and we certainly would expect a decrease to happen at exactly the frequencies trapped by greenhouse gases, nor would we expect the rate of decrease to match the modeled predictions based on our understanding of how greenhouse gases work.

Note that these measurements are typically taken on “clear sky” days to eliminate cloud cover as an explanation.

This is a great set of predictions because they make it possible to falsify climate change and they should only come true if our greenhouse gases are driving the current warming.

So what have we found? You probably guessed it: multiple studies tackling the problem from multiple angles have consistently found that all three predictions are true. When we use satellites to measure outgoing radiation we find a decrease specifically at the frequencies of our greenhouse gases, and those decreases match the models (Harries et al. 2001; Griggs and Harries. 2007; Kramer et al. 2021; Whitburn et al. 2021; Raghuraman et al. 2023; Teixeria et al. 2024). These results also match similar land-based research that directly measured the impact of CO2 (Feldman et al. 2015).

This is it. If there was ever any doubt, this settles it. Case closed. This a smoking gun, a clear fingerprint of our actions, and any other courtroom analogy you can think of. Really think about how extraordinary these predictions are and ask yourself, “why did they come true if our gases are not driving the change?”

We KNOW that our gases are trapping extra heat. We’ve measured it! This isn’t a “theory” or a “model.” It is a direct measurement of our gases trapping heat and warming the planet. This is the most conclusive evidence we could ever hope to have.

If you want to continue to deny that humans are causing climate change, then you have some really hard questions to answer like, where is the heat going? We know it’s not leaving earth, so what’s trapping it? Further, what is managing to trap it at the frequencies that we know our gases trap? And how is it doing it at exactly the rate we expect based on the increase in greenhouse gases?

Any alternative to anthropogenic climate change has to present compelling answers to those questions. Good luck.

Note: It is simply not true that climate change predictions have been wrong. Certainly there have been some sensational claims in the media as well as bad distortions of scientists’ predictions, and if you really try, you can cherry-pick off the cuff comments from individual scientists, but the core, consensus models scientists rely on have actually done a very good job of predicting trends. Details and sources here

 Literature cited

  • Allen et al. 2006. Quantifying anthropogenic influence on recent near-surface temperature change. Surveys in Geophysics 27:491–544.
  • Feldman et al. 2015. Observational determination of surface radiative forcing by CO2 from 2000 to 2010. Nature 519:339-343
  • Foster and Rahmstorf. 2011. Global temperature evolution 1979–2010. Environmental Research Letters 7:011002.
  • Griggs and Harries. 2007. Comparison of spectrally resolved outgoing longwave radiation over the tropical Pacific between 1970 and 2003 Using IRIS, IMG, and AIRS. Journal of Climate 20:3982-4001.
  • Hansen et al. 2005. Earth’s energy imbalance: confirmation and implications. 308:1431–1435.
  • Harries et al. 2001. Increases in greenhouse forcing inferred from the outgoing longwave radiation spectra of the Earth in 1970 and 1997. Nature 410:355–357.
  • Imbers et al. 2014. Sensitivity of climate change detection and attribution to the characterization of internal climate variability. Journal of Climate 27:3477–3491
  • Kramer et al. 2021.  Observational Evidence of Increasing Global Radiative Forcing. Geophysical Research Letters 48: e2020GL091585
  • Lean and Rind. 2008. How natural and anthropogenic influences alter global and regional surface temperatures: 1889 to 2006. Geophysical Research Letters 35:L18701.
  • Meehl, et al. 2004. Combinations of natural and anthropogenic forcings in the twentieth-century climate. Journal of Climate 17:3721–3727.
  • Lockwood and Frohlich. 2007. Recently oppositely directed trends in solar climate forcings and the global mean surface air temperature. Proceedings of the National Academy of Sciences 463:2447–2460.
  • Lockwood and Frohlich. 2008. Recently oppositely directed trends in solar climate forcings and the global mean surface air temperature. II. Different reconstructions of the total solar irradiance variation and dependence on response time scale. Proceedings of the National Academy of Sciences 464:1367–1385.
  • Raghuraman et al. 2023. Greenhouse Gas Forcing and Climate Feedback Signatures Identified in Hyperspectral Infrared Satellite Observations. Geophysical Research Letters 50: e2023GL103947
  • Santer et al. 2012. Identifying human influences on atmospheric temperature. PNAS 110: 26-33
  • Stips et al. 2016. On the causal structure between CO2 and global temperature. Scientific Reports 6: 21691
  • Stott et al. 2001. Attribution of twentieth century temperature change to natural and anthropogenic causes. Climate Dynamics17:1–21.
  • Teixeria et al. 2024. Direct observational evidence from space of the effect of CO2 increase on longwave spectral radiances: the unique role of high-spectral-resolution measurements. Atmospheric Chemistry and Physics 24: 6375-6383
  • Wild et al. 2007. Impact of global dimming and brightening on global warming. Geophysical Research Letters
  • Whitburn et al. 2021. Trends in spectrally resolved outgoing longwave radiation from 10 years of satellite measurements. 4: 48

Posted in Global Warming | Tagged , | 5 Comments

How does evolution explain complex mimicry?

Posted on December 20, 2025 by Fallacy Man

Evolution is, in my opinion, the most fascinating topic in all of science. It provides elegant, compelling, and enthralling answers for everything we observe in biology. It really is true that nothing in biology makes sense without evolution. Unfortunately, not everyone shares my passion for this topic, and evolution is often poorly taught and badly misunderstood.

Some examples of the comments people were making

As an illustration of this, I was recently wasting time on Twitter (I refuse to call it X) when I came across someone sharing an article from a few years ago about the discovery of a new species of rove beetle (Austrospirachtha carrijoi) that had evolved a specialized abdomen that made it look like the beetle was carrying a termite larvae “puppet” on its back. The puppet provided a disguise for the beetle and allowed the beetle to live inside termite tunnels and even be fed by the termites!

It’s a very cool discovery, and a fascinating example of the crazy stuff evolution can produce. However, the person sharing the article had a different take, commenting, “random mutation is a retarded explanation for this btw.” Many others chimed in agreeing with him and committing one straw man argument after another. As if that wasn’t bad enough, even many of the people who were commenting in defense of evolution were also badly misrepresented evolution. These comments predominantly followed two paths, either erroneously claiming that mutations are not random or doubling down and asserting that random mutations could create this given enough time.

The reality is more complex and fascinating, and both the original poster and many of the commenters* were missing three critical aspects of how evolution works:

  1. While mutations are random, natural selection is, by definition, non-random
  2. Evolution is “blind,” meaning that it has no direction or goal that it is trying to achieve
  3. Evolution doesn’t happen in isolation. Species co-evolve together.

In the post, I want to walk through each of these using the rove beetle as an example and try to give you a better understanding of how evolution by natural selection actually operates. To any young earth creationists who might be reading, as always, I ask simply that you hear me out and try to actually understand these evolutionary mechanisms. As I’ve written about before, I used to be a young earth creationist, and although I thought I understood evolution, essentially all of my objections to it were actually straw men. Put another way, in this post, I am not trying to convince you that evolution is the explanation for this rove beetle, rather I am simply trying to demonstrate that when properly understood, evolution does, in fact, offer a compelling and rational explanation (see my other posts for evidence that evolution is correct, e.g., here, here, and here).

*Note: some commenters did get it correct by pointing out that mutations alone are a bad explanation but mutations plus natural selection is a very good explanation.

 Mutations and selection

First, to be completely clear, genetic mutations are random. Drop any notions you have about mutations being guided, or consciousness affecting mutations, or mutations being biased towards beneficial ones. Mutation is a random process that happens during reproduction, and it (accompanied by genetic crossing over and the independent assortment) results in enormous genetic variation. Most mutations are neutral, a few are harmful, and a few are beneficial (see this post for more details on mutations).

Second, I completely agree that mutations alone would be a very poor explanation for the diversity of life that we see. Even with millions of years and millions of individuals mutating and reproducing, the odds of the exact right set of mutations occurring in the exact right order to create something like a rove beetle (all while avoiding harmful mutations) are astronomical, and when you apply that math to all of nature, it becomes wildly implausible. That’s why natural selection is so important.

This is one of the single biggest misunderstandings about evolution by natural selection: it is not random. Any time you hear a creationist scoff about something evolving “just by chance,” they are committing a straw man fallacy. Mutations are random, but that’s just step 1. The next absolutely critical step is natural selection, which is, by definition, not random.

In each generation, numerous mutations occur, most of which are neutral (those do evolve randomly via genetic drift), but every once in a while, one of them is beneficial. The individuals with that beneficial mutation survive/reproduce just slightly better than the ones who lack that mutation. As a result, they produce more offspring than the other individuals, and that inherently means that the beneficial mutation becomes more common in the next generation (more offspring = more copies of the mutation). Those offspring carry that mutation with them, and, just like for their parents, the mutation gives them an advantage allowing them to produce more offspring, which means even more copies of that mutation in the next generation. Each generation, that mutation becomes more and more common in the population all thanks to simple math. That’s it. That’s how natural selection works. There’s nothing mystical or atheistic about it: it’s just simple math.

Note that I am glossing over some complexities of inheritance that are irrelevant for the core argument, but in many cases it would take two copies of the mutation to have a benefit.

When a negative mutation arises, the process is the reverse. Individuals with that mutation produce fewer offspring, which means that the mutation becomes less common each generation (i.e., nature selects against it).

Note that this process is not random. Which mutations stay and which mutations go is determined by the effects they have, and they are “selected” for or against simply by causing the production of more or fewer offspring. This is critical, because it means species can accumulate beneficial mutations rather than accumulating mutations randomly.

Let’s use dice as an example. Let’s say you have 10 regular, 6-sided dice, and you want to get all 10 on the number 1. We’ll think of each throw as a generation reproducing, and each number as a mutation. The odds of tossing the 10 dice and getting all 10 to land on 1 (i.e., random mutations) are extremely low. In fact, they are 6^10 or 1 in 60,466,176. You could sit there throwing the dice for days and never get it. That’s the random mutation model, and I agree that it is absurd; so now let’s add selection into the mix.

Suppose instead, that every time you get a beneficial mutation (i.e., a 1) it is kept (in the same way that nature selects the beneficial mutations). So now, each time you throw the dice, you keep any 1s, then throw the remainder for the next generation. In this scenario, you’d actually get to a set of ten 1’s very quickly (go try it yourself if you don’t believe me).

Out of curious (and because I’m something of a nerd), I programmed a quick commuter simulation to try this and see how long it would actually take to get ten 1s, and on average, it only took 16.5 throws. We went from odds of 1 in over 60 million to averaging 16.5! That’s the incredible power of selection.

This is why the mathematical arguments against evolution fail: they are focused on mutations while ignoring the selection component. Once you add selection into the equation, it becomes entirely plausible to evolve something like a rove beetle with a termite puppet. The math works.

Evolution is blind

Now that we have cleared up the math, let’s look at the “blindness” of evolution. I really like the dice example I used above except for one important caveat: it gives the false impression that evolution is working towards some ultimate goal (like us trying to get all 1s). In reality, nature is not trying to accomplish anything. There’s no goal in mind. Each generation, the genes that result in the production of more offspring inherently get passed on to the next generation in higher numbers, while the genes that result in fewer offspring inherently are less common in the next generation. In other words, evolution is working one generation at a time.

Species evolve based on their current environment, which means that a trait that was beneficial in one generation can become detrimental in the next generation if the environment changes. Likewise, a trait that was being selected for one reason can get repurposed for something else if the environment changes or the right mutation comes along. Take the wings of a penguin, for example. For the penguins great, great, great, etc. ancestor, the wings were selected for flight, and evolution evolved them accordingly. Then, conditions arose that made being a strong swimmer more important than being able to fly, so evolution repurposed the wings into paddles for swimming. Again, there was no conscious process, it was simply that each generation, the individuals who had slightly better wings for swimming were able to do a better job escaping predators and/or catching food, which resulted in more offspring and more of the genes for improved swimming in the next generation.

Turning back to our rove beetle friend, there is a huge spectrum of insects (including many other types of rove beetle) that have evolved to raid termite and ant colonies, and they range widely from very ordinary beetles (and other insects) that get attacked constantly by the ants/termites to critters like the one in question that are so highly specialized that the termites accept them rather than attacking.

Many rove beetle species have an enlarged abdomen that sticks up over top them and looks something like a halfway point between a regular rove beetle and our termite puppeteer. What good is that blob? Glad you asked. Termites perceive the world largely through touch and chemical pheromones, and in some cases, that swollen abdomen seems to produce pheromones that help the rove beetle go unnoticed in the termite colony. So, selection has been acting on the abdomens to turn them into enlarged hormone factories.

Now we can easily imagine a scenario in which these pheromone-producing rove beetles do great when the termites sniff them, but once the termites start touching, they run into issues, because the beetles don’t feel like a termite (thus blowing their cover). Then along comes a mutation that makes their swollen abdomen slightly more termite-like. Perhaps it constricts at one point like a body segment or has a tiny protrusion like a leg. This makes it slightly less likely that the rove beetle will be detected, which lets it produce slightly more offspring, which results in that mutation becoming more common in the next generation. Then, just like with our dice example, beneficial mutations start to accumulate as each of them is selected and all negative ones are selected against, until eventually, we end up with this stunning example of mimicry. Keep in mind that the mutation doesn’t need to provide an enormous advantage. Any slight increase in survival/reproduction will be enough to shift the math in its favor**.

In that scenario, the swollen abdomen might not have been a very good mimic at first, but that was fine because it was being selected for pheromone production, not tactile mimicry. Then, later on, with the right mutation, evolution shifted course and started evolving a tactile mimic.

To be clear, I don’t know for sure that what I described is that path evolution took, and you can no doubt think of other equally plausible paths. My point is simply that these plausible paths exist, and it is entirely possible to get to this “final” stage of a highly complex mimic one simple step at a time.

For some more examples and discussions of this type of mimicry I recommend Parmentier 2000. Guests of social insects. In Encyclopedia of Social Insects. Springer.

 For more details on evolution being blind see this post.

**Note: there technically are cases where other factors do override mutations that are only minorly beneficial. These include things like large amounts of gene flow from other populations and small population sizes, which allow genetic drift to override selection. However, for most large populations, even a tiny benefit gets selected because that is how math works (more offspring = more copies of the mutation).

Species co-evolve

Finally, you may still be thinking that everything I have said is all well and good, except that there is no way for the process to get started because a half-formed mimic would surely be detected. Put another way, if evolution was the cause of this superb mimic, then it is inherently true that even a slightly less superb mimic would not do as well. So how could my hypothetical example of a mutation that causes a rove beetle with a swollen abdomen to get a slight constriction possibly be advantages? Wouldn’t the termites detect that right away?

The answer is simply that we are seeing the late stage of an evolutionary arms race that has been going on for a very long time. It’s like looking at a modern fighter jet and asking how a biplane could ever have been useful. The answer, of course, is that when biplanes were in use, they were going up against other biplanes, and we only got to modern fighter jets by different militaries constantly trying to outdo each other (here again, evolution is not consciously trying to do anything, but the same sort of arms race still occurs due to the math).

So, let’s back the clock up several million years and return to my example of a rove beetle who has a swollen abdomen for pheromone production, then gets a mutation for a constriction on the abdomen. At that stage, the termites would be naïve to that sort of trickery. They would not have the ability to distinguish even a poor mimic because they’ve never faced that problem. This would give our rove beetle a slight advantage, but it would also create a selection pressure on the termites to do a better job identifying mimics. So, termite colonies that have genes that make them slightly less trusting of the rove beetles with constrictions will do better, produce more offspring, which means more genes, etc. This puts the evolutionary pressure back on the beetles, resulting in a selection pressure for any further mutations to make them more termite-like, but every time the beetle adapts, the pressure flips back to the termites. Every time the beetles evolves better mimicry, the termites evolves better mimicry detection, and every time the termites evolves better mimicry detection, the beetles evolves better mimicry. Back and forth the two go for millions of years, each evolving in response to the other’s adaptations.

Isn’t that neat? I find it absolutely fascinating, and I have very distinct memories of learning about these arms races for the first time as an undergraduate (bats and moths were the key example there). Learning about this topic really opened my eyes and helped me to understand the true power and flexibility of evolution. I hope it has helped you as well.

Conclusion

I hope you can now see that evolution provides a rational, internally consistent, and compelling explanation for how something like a rove beetle that mimics termites could come into existence. Creationism, in contrast, lacks a compelling explanation. All creationists can do is shrug their shoulders and say, “God wanted it that way,” but that’s a cop-out, not an explanation. To those of us who want to understand the natural world, it is completely unsatisfactory (further it runs into all sorts of issues if you believe in things like Noah’s flood: how did that beetle and its termite host survive the flood, then find each other afterwards and make it all the way to Australia?). Evolution by natural selection is the only reasonable explanation, and you simply cannot understand biology without it.

The paper in question is: Zilberman and Pires-Silva. 2023. A new species and morphological notes on the remarkable termitophilous genus Austrospirachtha Watson from Australia (Coleoptera: Staphylinidae: Aleocharinae). Zootaxa 5336

Related posts

Posted in Science of Evolution | Tagged , , | 1 Comment

Vaccines are tested against placebos

Posted on December 2, 2025 by Fallacy Man

Lately, my Facebook page has been flooded with people insisting that  “no vaccine is tested against a placebo” (sometimes stated with additional qualifiers like “double-blind” or “saline placebo”). This claim, like so many anti-vaccer claims, is blatantly false. Nevertheless, I think it is worth looking at it more closely to better understand the tools available to scientists, the information they provide, and which tools to use for which applications. People often act as if randomized placebo-controlled trials (RCTs) are the one and only valid scientific approach and no other study design is even worth considering. In reality, placebo-controlled trials are great for certain applications, but they are far from the only useful approach, and in many cases, they are actually a highly inappropriate method.

saline placebo vaccine comment resposne

Vaccines are tested against placebos (including saline placebos)

It is simply not true that vaccines aren’t tested against placebos (including inert saline placebos). Anyone who says otherwise is either lying or willfully ignorant. Don’t take my word for it, go to Google Scholar or PubMed or any other scientific database with medical research and do searches like, “saline placebo double blind vaccine” and you will find literally thousands of papers. This sort of testing is standard as part of Phase II and Phase III clinical trials. Vaccines routinely go through RCTs before being released to the public (including COVID vaccines, btw, e.g., this RCT on Pfizer’s vaccine [BNT162b2; Thomas et al. 2020] or this RCT on Moderna’s vaccine [mRNA-1273; Baden et al. 2021], both of which used saline placebos).

It’s also worth briefly noting that among those studies you will find numerous trials where the vaccine either did not work or had serious side effects and, therefore, never went to market. Vaccines are stringently tested and ones that don’t pass those tests are either scrapped or modified then retested.

Other types of controls

Although clinical trials do generally use inert, saline placebos, there certainly are some studies that either use an older version of the vaccine or the vaccine’s adjuvants as the control (“placebo”). There are usually very good reasons for this, so let’s talk about them.

Let’s start with why scientists might use an older version of a vaccine as a control rather than an inert saline placebo. I’d like to begin with an analogy. Imagine you are a seat belt manufacturer, and you have developed a new seat belt design that you want to test. What would be the best control for those tests: no seat belt at all or the current standard seat belt design? Obviously comparing it against the current standard design makes more sense, right? You already know that the current design works. You already know its safety profile. You already know the level of protection it provides and how often it ends up injuring someone (seat belts aren’t 100% safe or effective after all). So, the relevant question is not, “how safe and effective is the new design compared to nothing?” The question is, “how safe and effective is the new design compared to the current standard design?”

The same situation is true for vaccines. If you develop a new version of a given vaccine, it may make more sense to simply compare it to the current standard of that vaccine which already has a known safety profile and level of effectiveness, rather than comparing it to a placebo (this is especially relevant given the ethical issues that we’ll get to in a minute).

Using adjuvants as the control group often occurs for similar reasons. Namely, it lets scientists isolate the effects of the active components of the vaccine compared to the adjuvants. This is useful both because the adjuvants have generally already been tested and have known safety profiles and because it means that if problems arise in the vaccine group, then scientists can narrow down which part of the vaccine is causing the problem and make appropriate changes.

I say this a lot on this blog, but scientists aren’t stupid. They spend a lot of time thinking about experimental designs and their designs have to pass the approval of ethics committees before proceeding. They try very hard to ensure they are using the best design possible to answer the question at hand. So, when you see something like a trial that used an adjuvant rather than an inert placebo, don’t throw up your hands and assert that the research is all nonsense. Rather, take time to carefully look at the study. Read the authors’ justification for the design, and look at how it fits into the broader literature. There’s likely a good reason for the study’s design.

Ethical issues with placebo-controlled trials

Despite everything that I have said so far, it is true that after the initial clinical trials, most follow-up studies on vaccines are not placebo-controlled, and there are several important reasons for this. One of the biggest is simply ethics.

Once a vaccine has been demonstrated to be safe and effective, it becomes highly unethical to do a double-blinded placebo-controlled trial where you are leaving half the participants (often children) vulnerable to diseases that you know you could prevent. It is totally unethical to play Russian Roulette with their lives. Despite what anti-vaccers like to claim, the benefits of vaccines are so extraordinary and so well documented that it is unethical to withhold them for an experiment.

Fortunately, scientists have other tools they can use, many of which are actually better than RCTs when it comes to detecting the conditions that people are often concerned about (e.g., autism).

Other useful study designs

Beyond the ethical issues, placebo-controlled trials are also problematic because they are expensive to run, difficult to maintain adequate controls over long periods, and are data hungry when it comes to rare events. That last point is the one that I really want to focus on here.

Sample size is obviously important in research, and the rarer the effect, the larger the sample size needs to be, but that quickly becomes a problem with RCTs. Imagine, for example, that an outcome in question has a background rate of 1 in 1000 and the drug being tested increases that rate to 2 in 1000. How big of a sample size do you need to detect that increase that with an RCT? Even without getting into the mathematical details, you should easily be able to convince yourself that it is going to require a rather massive data set to detect such a rare event. Even at 10,000 participants in each group (a truly enormous RCT) you’d only expect 5 positive cases in the control group and 10 in the drug group. That’s not a very big difference.

So, in those cases, a much better design is often what is called a case-controlled design. I explained this in more detail here, but essentially, what it does is flip the study design so that you start with a group of people with the outcome of interest (e.g., autism), match them with a group of people that is similar except that they lack that outcome of interest (e.g., don’t have autism), then you work backwards to look for the potential cause of interest (e.g., difference in vaccination rates between the two groups). This study design is great for rare conditions because you start with that rare outcome, rather than starting with a huge number of people and waiting to see if some eventually develop it. So even if the outcome only occurs in 1 in 100,000 people, you may be able to find enough medical records to let you do a meaningful study. Even a few hundred participants can be quite powerful for this design, whereas a few hundred people would be utterly meaningless in an RCT for rare conditions.

Although not the point of this post, I’ll note that this design has repeatedly been used to study vaccines and autism, often with large samples sizes, and the result has consistently been that vaccines do not cause autism (e.g., Destefano et al. 2004; Smeeth et al. 2004; DeStefano et al. 2013; Uno et al. 2015)

Now, you may be thinking, “but vaccines cause wide-spread harm, not rare occurrences!” In which case, you’re wrong, but, more importantly, we have a better design for more common conditions. Namely, a retrospective cohort study. Again, more details here, but in brief, this design generally follows more of the “standard” experimental approach, but it uses medical records, rather than actually giving patients anything (thus avoiding the ethical issues). So, you can look through medical records to, for example, categorize people into groups that did or did not receive a given vaccine (but are otherwise similar) then look at the proportions in each group that developed an outcome of interest (e.g., autism). The big advantages here are that you don’t have the ethical issues, the cost is lower than RCTs, and you don’t have to personally follow patients for years. As a result, you can achieve massive sample sizes of tens or even hundreds of thousands of participants, something that is almost never possible in RCTs. So, this design can be much more powerful than an RCT (for certain questions) simply because of the enormous sample sizes it allows.

Once again, several massive cohort studies have looked at vaccines and autism and consistently found that there is no association (e.g., Hviid et al. 2019; Madsen et al. 2002; Anders et al. 2004; Jain et al. 2015)

A lack of correlation is generally a lack of causation/you don’t always need placebos

There is one more critical topic that needs to be discussed here. Namely, study designs like cohort studies and case-controlled studies are much better at showing a lack of effect than they are at showing causation.

Causation is a hard thing to demonstrate (something anti-vaccers don’t seem to grasp). To confirm it, you need to not simply show that two things change together (i.e., are correlated), but rather that nothing other than causation can explain that correlation. This is the beauty of a randomized placebo-controlled design and why it is the “gold standard” for testing effectiveness. By randomizing the treatment across patients and administering a placebo (plus careful statistical analysis), you can control confounding factors so that you can be confident that the changes you see in the treatment group are being caused by the experimental factor rather than simply being associated with it. RCTs are, in most cases, the best way to establish causation (at least for topics like medicine).

Because they do not randomize and do not include placebo controls, case-controlled studies and cohort studies generally struggle to demonstrate causation. If they were able to very carefully match their participants and include appropriate covariates in their models, they may be able to strongly suggest a causal relationship, but there are serious limits to their interpretation. They are, however, great at establishing safety, because while correlation does not automatically indicate causation, a lack of correlation does generally indicate a lack of causation.

To understand what I mean, we need to talk about causation just a little bit further. Imagine that you show that X is positively correlated with Y. So, when X goes up, Y also goes up. Does X cause Y? Maybe, but it could also be that Y causes X or that some other variable is causing both X and Y. You’d need an RCT to tease that out.

However, suppose that X and Y are not correlated. So changes in X do not correspond to changes in Y. Does X cause Y? No. It’s an easy answer. At least within the statistical limits of the study in question, if changes in X don’t result in changes in Y, then X doesn’t cause Y.

What this means is that when large cohort and case-controlled studies find a total lack of association between vaccines and something like autism, that is actually really good evidence that vaccines don’t cause autism. Think of it this way, how could vaccines be causing autism if the group that received that vaccines doesn’t have higher autism rates than the group that did not receive the vaccine? (again assuming appropriate statistical design, case-matching, and within the confidence limits of the study)

To put that another way, if you are one of the people who likes to harp on the supposed lack of placebo-controlled vaccine trials (again, the do actually exist), then I want you to look at the large cohort studies and tell me exactly how you think a placebo would have made a difference. Look at studies like Madsen et al. (2002), which looked at records for 440,655 children who received an MMR vaccine and 96,648 children who did not receive an NMR vaccine, then compared their rates of autism (they were not statistically different). Exactly how do you think things would have been different if the people in the unvaccinated group had received a placebo instead of simply not being vaccinated. Do you think the placebo would magically have prevented them from developing autism? What would that have changed? A lack of placebo is simply not a valid criticism for this sort of study. If people who received the vaccine didn’t experience higher rates of autism, then that is good evidence that the vaccine doesn’t cause autism.

“But what about the entire vaccine schedule!?”

Finally, one last specific criticism I sometimes here is, “sure, individual vaccines were tested against a placebo, but no one has ever done a placebo-controlled trial on the entire vaccine schedule.” For once, that claim is at least true (to the best of my knowledge), but it is also a meaningless demand for an impossible test. That test would be completely unethical and also extremely difficult to pull off. It’s just not a plausible experiment.

There are, however, other approaches that scientists have used. For example, there is this massive cohort study of almost a million children that looked at numerous combinations of vaccines, comparing them to single vaccines (Bauwens et al. 2022). There are also studies that looked at the effects of antigen load (i.e., the number of antigens children are exposed to from vaccines; DeStefano et al. 2013; Iqbal et al. 2013; DeStefano et al. 2013). Other studies have looked at adding vaccines to the schedule or the effects of vaccines given together or spaced out (Olivier et al. 2008; Arguedas et al. 2010; Vesikari et al. 2010). All of these are different ways to examine combinations of vaccines.

Additionally, based on everything we know about vaccines (including studies like the ones cited above), there is simply no reason to expect serious harm from the routine schedule, and no matter how thoroughly something has been tested, there will always be things that haven’t been tested.

Imagine, for example, that someone actually did manage to do an RCT on the whole schedule and found a lack of significant side effects, anti-vaccers could then respond (and likely would respond) with things like, “well what about the whole schedule + GMOs, or the whole schedule while taking aspirin, or the whole schedule while watching 10 hours a week of TV” etc. There are endless possibilities, most of which are utter nonsense.

Don’t get me wrong, when actual serious, plausible concerns arise, scientists should (and do) take that seriously and test accordingly, but that’s simply not the case with these arguments about vaccines and placebo-controlled trials. Likewise, just to be 100% clear, there are side effects from vaccines (just like all real medicines). No one is saying that they are 100% safe, but they are very well-tested and serious side effects are extremely rare. Nevertheless, I (and all scientists) welcome improvements in vaccines to make them even safer. Unfortunately, instead of perusing those improvements, we are left doing things like looking at vaccines and autism for the 100th time as if one more study will somehow make a difference.

Statistical note: Throughout, you may have noticed that I use phrases like “within the statistical limits.” This is an important, albeit somewhat technical, caveat. In brief, it is never possible to prove a negative. There are several reasons for this, but the most important for the topic at hand is that it is always possible that there is an effect, but it occurs rarely enough that you were not able to detect it with the current sample size. So even if you have a sample size of several million participants, you aren’t going to detect something that occurs once every billion patients. So you can never, for any treatment, say with 100% certainty that there is no effect. However, if with large enough studies and proper statistics and designs (as has been done for vaccines and autism), you can confidently state that if there is an effect, it is so extraordinarily rare that for all intents and purposes, there is no effect. Stated another way, for something like vaccines and autism, we are as close to being able to state that “vaccines do not cause autism” as we are ever going to be. We have used so many giant studies, that we can state that if vaccines cause autism, it is such an extremely rare side effect that it is statistically undetectable.

Related posts

Literature cited

  • Anders et al. 2004. Thimerosal exposure in infants and developmental disorders: a retrospective cohort study in the United Kingdom does not support a causal association. Pediatrics 114:584–591
  • Arguedas et al. 2010. Safety and immunogenicity of one dose of MenACWY-CRM, an investigational quadrivalent meningococcal glycoconjugate vaccine, when administered to adolescents concomitantly or sequentially with Tdap and HPV vaccines. Vaccine 28:3171-3179
  • Baden et al. 2021. Efficacy and Safety of the mRNA-1273 SARS-CoV-2 Vaccine. New England Journal of Medicine 384:403–416
  • Bauwens et al. 2022. Safety of routine childhood vaccine coadministration versus separate vaccination. BMJ 7
  • DeStefano et al. 2004. Age at first measles-mumps-rubella vaccination in children with autism and school-matched control subjects: a population-based study in metropolitan Atlanta. Pediatrics 113:259–266
  • DeStefano et al. 2013. Increasing exposure to antibody-stimulating proteins and polysaccharides in vaccines is not associated with risk of autism. J Ped 163:561–567
  • Dolhain et al. 2019. Infant vaccine co-administration: review of 18 years of experience with GSK’s hexavalent vaccine co-administered with routine childhood vaccines. Expert Review of Vaccines 19:419-443
  • Glanz et al. 2018. Association Between Estimated Cumulative Vaccine Antigen Exposure Through the First 23 Months of Life and Non–Vaccine-Targeted Infections From 24 Through 47 Months of Age. JAMA 319:906-613
  • Hviid et al. 2019. Measles, mumps, rubella vaccination and autism: A nationwide cohort study. Annals of Internal Medicine.
  • Iqbal et al. 2013. Number of antigens in early childhood vaccines and neuropsychological outcomes at age 7–10 years. Pharmacoepidemiology and Drug Safety 22:1263-1270
  • Jain et al. 2015. Autism occurrence by MMR vaccine status among US children with older siblings with and without autism. JAMA 313:1534–1540
  • Madsen et al. 2002. A population-based study of measles, mumps, and rubella vaccination and autism. New England Journal of Medicine 347:1477–1482
  • Olivier et al. 2008. Immunogenicity, reactogenicity, and safety of a seven-valent pneumococcal conjugate vaccine (PCV7) concurrently administered with a fully liquid DTPa—IPV—HBV—Hib combination vaccine in healthy infants Vaccine 26:3142-3152
  • Smeeth et al. 2004. MMR vaccination and pervasive developmental disorders: a case-control study. Lancet 364:963–969
  • Thomas et al. 2020. Safety and Efficacy of the BNT162b2 mRNA Covid-19 Vaccine through 6 Months. New England Journal of Medicine 85:1761–1773
  • Uno et al. 2015. Early exposure to the combined measles-mumps-rubella vaccine and thimerosal-containing vaccines and risk of autism spectrum disorder. Vaccine 33:2511–2516
  • Vesikari et al. 2010. Immunogenicity and safety of the human rotavirus vaccine RotarixTM co-administered with routine infant vaccines following the vaccination schedules in Europe. Vaccine 28:5272-5279
Posted in Nature of Science, Vaccines/Alternative Medicine | Tagged , , , , , , | Comments Off on Vaccines are tested against placebos

Don’t be intellectually lazy

Posted on March 25, 2025 by Fallacy Man

stick figure, read before posting, article, title, fact check blue and whiteIn recent conversations on this page, I have been struck by just how intellectually lazy science-deniers usually are. This is hardly a novel observation, but I think it bears discussion. I also want to note that this sort of lazy thinking is common in politics and countless other topics, and it is very easy to fall into these bad habits. Critical thinking is a skill, and like most skills, it requires practice. Being well-informed takes hard work. Blind adherence to biases and preconceptions is much easier than rigorous fact-checking and serious contemplation. We are all prone to cognitive biases, but if we want to have rational views based on evidence and logic, then we need to acknowledge those tendencies and fight against them. We need to be humble and acknowledge the limits of our personal knowledge and be intellectually diligent and honest. Blind denial of any information you don’t like is easy and seductive, but it is not rational or intellectually rigorous.

To illustrate what I mean by being intellectually lazy, I am going to use comments on a recent post I wrote about masks and COVID vaccines (as well as a few others from related posts), and I’m going to broadly categorize the lazy responses into three groups: blind denial, refusal (or inability) to cite evidence, assumptions and generalities.

Blind denial

The aforementioned article took me a long time to write. I wanted to be thorough and present a fair and honest representation of the scientific evidence. So, I spent many days reading the literature, fact-checking, and making sure that what I was saying was correct. I cited roughly 50 peer-reviewed articles (mostly meta-analyses and systematic reviews) and read numerous others while preparing that post.

Unsurprisingly, many of the comments were, shall we say, less rigorous. They can be summed up simply as, “I don’t believe it, and you’re a gullible idiot for believing it.” None of these people presented actual logical reasons why my arguments were wrong. None of them presented problems with the studies I was citing. Indeed, most of the comments were made by people who probably did not even bother to read my original post. They simply saw a title that disagreed with their views, so they automatically assumed I was wrong.

This is the extreme end of being intellectually lazy. Automatically dismissing any information that disagrees with your view is the very definition of being close-minded and it is the epitome of intellectual laziness.

One of the things that I say over and over again on this blog is that we must always be willing to question our views. We must always be willing to actually consider contrary evidence and carefully examine the possibility that we are wrong.

I want to be clear here, there would have been nothing wrong with someone saying, “I don’t agree with you because of the following specific problems with the studies you cited…” followed by actual issues with the studies and appropriate contrary evidence. I’m not saying that you have to blindly accept contrary information. Rather, I am simply saying that you have to give it a fair hearing. If you are being given legitimate evidence, then it behooves you to take it seriously and carefully examine it before deciding whether it is correct or incorrect.

So, to everyone who rejected my statements regarding COVID vaccines and masks, my question continues to be, why? What specifically do you disagree with? Why is the evidence that I presented unsatisfactory? Or, to put it another way, if all of those massive studies from around the world are not satisfactory, then what would be? What evidence would make you reconsider your position? Being intellectually rigorous means taking the time to ask yourself these sorts of questions. If you are rejecting something, ask yourself why, specifically do you reject it? Can you cite specific problems with it, or does your response consist of vague generalities (see section three).

Let’s take a step back from the topic of COVID and talk more generally for a second. Surely, we can all agree that being well-informed inherently involves a willingness to accept contrary evidence, right? How could you ever possibly know that you are wrong about something unless you are willing to look at opposing evidence when presented with it? Refusal to even consider contrary evidence creates a self-reinforcing view that is immune to logic. Blind denial is so much easier than a careful examination of the facts, but it is a trap that we must avoid at all costs. Be intellectually vigilant, not lazy.

Refusal (or inability) to cite sources

The next form of intellectual sloth that I want to discuss is a refusal to back up your claims. This is another topic on which I spend a lot of time on this blog. “That which can be stated without evidence can be dismissed without evidence.” Stated another way, the burden of proof is always on the person making a claim.

So, for example, when I claimed that COVID vaccines saved millions of lives, I backed up that claim with multiple large peer-reviewed studies. In contrast, the good people in the comments simply made their claims without providing supporting evidence, and when asked to provide that evidence, they refused, disappeared, or dodged. That is intellectually lazy.

At this point, many people respond with something like, “we aren’t all walking around with a stack of papers all the time.” My response to that is two-fold. First, if you are going to enter into a public debate, and especially if you are going to enter into a debate where the other side has already presented copious evidence, you do, in fact, have an obligation to have evidence for your claims. No one should take you seriously unless you can provide that evidence.

Second, (and more seriously) at least in my experience, that response is usually an excuse made by someone who doesn’t actually have real evidence. Multiple times now I have had a one-on-one conversation with someone where I kept pressing them for sources before they finally admitted that, “well, it’s just something I heard somewhere.” That’s a big problem, especially when the thing they “just heard” is something like, “all scientists are fundamentally wrong about basic facts in their field.” Likewise, people often try to dodge a request for evidence with something like, “just google it.” This is a cop-out response that, at least in my experience, almost always signifies a position built on sand.

So, if you find yourself unable to produce evidence for your claims, really ask yourself, “why do I believe this?” Have you actually seen reputable evidence to back up your claim, or is it just something you saw/heard on the news, facebook, youtube, etc. Have you verified that claim? Have you fact-checked it and traced it back to its original source, or is it just something that you believe because it sounded correct to you? If you can’t provide the actual evidence, then why do you believe it?

This is what I mean by being intellectually vigilant. You owe it to yourself to make sure there is actually a logical reason you hold the views that you hold, and if someone asks you for evidence, and you can’t produce it, take that seriously. Don’t be lazy and shrug that off. Be introspective about your views and get to the root of why you think a given thing is true or false. Does your view trace back to verifiable facts from legitimate sources? If not, why do you believe it? Why would you want to hold a view that isn’t based on actual facts and evidence?

Once again, this applies to far more than just science, and, in my experience, an inability to cite specifics is usually a sign of intellectual laziness where someone holds a view simply because it feels correct to them, rather than because they have carefully examined the evidence.

For example, a few months ago I had a discussion with a relative who insisted that a particular politician was “destroying the economy with their socialist policies.” I responded simply by asking them which policies specifically were destroying the economy. I asked them to name the pieces of legislation. If their view was actually based on evidence, that should have been a simple task. If they were actually basing that strongly held view on a careful examination of this politician’s policies and their economic impacts, it should have been quite easy to direct me to some specifics, but they could not give me any species. Instead, they hemmed and hawed and made excuses and cited their “personal experience.” The sad reality is that they were actually highly ignorant of this politician’s economic policies and were basing their views on misinformation and fearmongering. They had been fed misinformation by biased sources, and rather than fact-checking and testing the validity of those claims, they blindly believed them because they fit with their world view. That is what I mean by “intellectually lazy” and the dangers of that approach should be obvious.

Here again, it is worth being introspective. If, continuing the example above, you think that a politician has caused X, but can’t give any actually specifics of what they did to cause X and instead have to rely on vague generalities, then why do you have such a strong conviction on this issue? Where is your information coming from? Is it reliable?

Assumptions and generalities

This final category is quite broad but essentially consists of dismissing the evidence because of sweeping generalities that are based on assumptions.

By far the most common expression of this form of intellectual laziness is the “shill argument.” Many people dismissed the large body of evidence I presented because they assumed that all the studies were influenced by money from pharmaceutical companies. To be clear, conflicts of interest should be considered when evaluating studies, but that is not what these people were doing. Rather than actually checking the studies for conflicts of interest (studies always declare them) and carefully considering the evidence in light of those conflicts (when they existed), they were simply assuming that all of the studies were hopelessly compromised and all of the worlds hundreds of thousands of scientists had been bought off. In reality of course, many of the studies were not funded by pharmaceutical companies and had no conflicts of interest, and even when a conflict of interest is present, it does not automatically nullify the study. Rather it is simply another piece of information that has to be considered (see details here).

Conspiracy theories more generally also fall into this category of lazy thinking. They make sweeping generalizations based on assumptions that they cannot verify and often haven’t even tried to verify. They try to wave a magic wand and dismiss any and all evidence they don’t like by boldly proclaiming unverifiable statements rather than actually looking at the evidence. This is, of course, so much easier than actually seriously engaging in a topic, but only one of those paths will lead to an accurate, evidence-based view of the world.

This post has become something of a rant, so I will end it simply by reiterating the title: don’t be intellectually lazy. You owe it to yourself to examine your views and ensure that you are basing your positions on a careful consideration of the evidence. Being well-informed is hard work. It takes effort to fact check, verify, and examine contrary evidence, but it is vital if you want to have a realistic view of the world rather the going through life blindly.

Posted in Rules of Logic, Vaccines/Alternative Medicine | Tagged , | Comments Off on Don’t be intellectually lazy