How to find and access peer-reviewed studies (for free)

The peer-reviewed literature is where scientists publish their research, and it is the source for scientific information. As a result, I spend a lot of time on this blog talking about it. I have explained how the peer-review system works (also here). I have provided advice on how to evaluate studies and how not to evaluate studies. I have explained the hierarchy of evidence. I’ve explained P values and false positives. I’ve explained why many studies are unreliable and why it is important not to cherry-pick studies. I have provided worked examples of how to dissect studies (e.g., here, here and here), and I do my best to cite studies to back up all the claims I make on this blog. Nevertheless, it was recently pointed out to me that I have utterly failed to explain something important and fundamental: how and where to find peer-reviewed studies. So I am going to remedy that by providing a brief primer on how to go about finding articles on topics you are interested in, and how to get free copies of them.

Where to look

Let’s start with where to look. You can try simply doing a standard Google search, but odds are that you will get flooded with tons of blogs and websites, and it is a pretty inefficient way to find what you are after. A much better option is to use a database specifically tailored to peer-reviewed literature. There are two major ones that are freely available that I’m going to talk about: Google Scholar and PubMed (there are many others that are behind paywalls, but I am going to assume that most people reading this are not academics and don’t have access to those).

Let’s start with Google Scholar. First, I need to make it absolutely clear that this is not the same thing as a regular Google search. Literally anyone can get a blog, write an article, and it will show up in a Google search. In contrast, Google Scholar is tailored for academic articles, and you cannot manually add articles to it*. Instead, Scholar pulls from several academic databases (e.g., JSTORE) and employs bots to scour the internet for DOIs, abstracts, titles, etc., which it uses to identify peer-reviewed articles and add them to its repository. It’s not a perfect system; some articles get missed by the bots, and occasionally they pick up a non-peer-reviewed article that has the trappings of a peer-reviewed article (e.g., a non-reviewed report). Nevertheless, it is an extremely useful tool. It is a massive database that is very easy to use (more on that later) and even though I have access to more well-curated databases, Scholar is usually what I default to for quick searches.

Scholar also has the advantage of being a generalist database. In other words, it is not topic specific, and articles on medicine, zoology, climate change, GMOs, evolution, physics, chemistry, archeology, etc. can all be found within its digital walls. Sometimes though, it is useful to use a more focused database, and that is where PubMed comes in.

As its name suggests, PubMed is a repository for medical papers. It gets its papers both directly from journals and from author submissions. These submissions are checked to ensure that they are scientific papers. As a result, it tends to be more curated than Scholar, and you don’t get as many results that aren’t actually peer-reviewed papers.

There are lots of other databases out there, and if someone reading this has one that they love, feel free to mention it in the comments, but these are the two I’m going to focus on. I will quickly mention though that Mendeley’s database is often a good place to find more obscure articles. It is another generalist, but it allows author submissions, and on multiple occasions I have found papers there that didn’t show up elsewhere. So, while I wouldn’t use it as a primary database, it can be useful (you have to make an account, but it is free).

*If you are a research and have an account, you can manually add the bibliographic information for an article to Scholar, which may help Scholar to locate it if it hadn’t done so already, but you cannot simply upload an article.

How to search

Now let’s move on to how to actually find the papers you are after. For both PubMed and Scholar you can use them like a standard internet search and type in “vaccines autism,” for example, but that is going to return a ton of studies, so it is usually best to be as specific as possible. For example, if you specifically want to see results from randomized controlled trials, include that in your search terms.

Both databases also have very helpful advanced search settings. For PubMed, there is an “advanced” tab under the main search bar, and this returns a screen with a bunch of pretty self-explanatory options. For example, you can limit results to a specific author, specific journal, specific date range, specific word in the title, etc. Google Scholar is similar, but with fewer options (to get to it, click on the three lines on the left-hand side indicating a drop-down menu, then select “Advanced search”).

It can also be useful to either include or exclude specific words or phrases. PubMed and Scholar both let you include specific words or phrases by simply putting the word or phrase that you care about in quotes, at which point they will limit the searches to articles that contain that quote. This can be very useful if you are getting a lot of irrelevant results that include some parts of your search terms, but not exact phrases you are after. Conversely, there may be times when it is useful to eliminate a word. For example, if you are only interested in studies on humans, you might want to exclude a word like “mice” or “in vitro.” In PubMed, this has to be set in the advanced search option, but in Scholar, you can just ad a minus sign to the beginning of the word (or quoted phrase) that you want to exclude. This should be done cautiously, however, as you may inadvertently exclude relevant studies. For example, if you exclude the word “mice” you may accidentally exclude a study on humans that discussed rodent studies in the introduction or discussion, or even just cited a study with the word “mice” in the title. So, while this feature can be useful, it should be used carefully, and it is often better to put quotes around a word you care about, rather than eliminating a word. For example, you could put “human” in quotes, to force the search to give you more human trials. Having said that, quotes can bias search results and make it easier to cherry pick results (particularly when using long phrases). So, use these tools carefully.

Another really useful approach is to find one relevant study, then look both at the studies it cited and the studies cited by it. To my knowledge, PubMed does not have a “cited by” tab, but Scholar does under each article, thus allowing you to see which articles cited it. Also, both databases have a “related articles” or “similar articles” link under each article, which you can use to find other relevant research.

Personally, I find the citations within a paper to be the most useful. If you really want to understand a topic, then as you go through a paper, you should note the references to related studies that are worth reading. Then, you can use the literature cited section of the paper and Scholar or PubMed to look up those articles and read them. As you read them, you should find yet more articles. As you can well imagine, the number of articles you need to read balloons out pretty quickly, and it is why scientists have to spend so much time reading. This can, however, also provide a useful check for how well you have covered a topic. After reading a large number of papers, you should start to notice that the number of new, relevant papers being cited decreases. You should start to see a lot of familiar citations to papers you’ve already read. In other words, at first, the number of new citations to papers you need to read should be quite large after each paper you read, and that number will continue to grow until you start to get a good grasp on the literature. Then, it will slowly start to decrease as you read more and more of the relevant studies (i.e., it becomes harder and harder to find papers you haven’t read yet). This doesn’t mean that you are an expert and have read all relevant studies, of course, but it is a useful proxy for assessing your thoroughness.

How to get papers for free

Now comes the critical question, how do you actually get the paper without paying for it? In many cases, you can do so directly though Google Scholar or PubMed (Scholar is particularly good at finding and including links to free copies if they are available). Failing that, you have several options.

The first, is to do a standard Google search for the title of the paper. Sometimes, this brings up copies that Scholar missed. You can also check Research Gate and Mendelely, but usually Scholar picks those up. For papers on “physics, mathematics, computer science, quantitative biology, quantitative finance, statistics, electrical engineering and systems science, and economics,” you can also try, which is run by Cornell and offers free, legal, open access to many papers in those fields.

The second option (which is often the best) is simply to contact the author and ask for a copy. In almost every case, they will be more than happy to send one to you. I want to pause here for a moment to make a brief point. Scientists do not get paid for their publications. Those fees to access papers go directly and entirely to the publishers. Scientists do not get one cent from them. So, don’t feel bad about asking a scientist for their research, because you aren’t costing them anything, and they will be thrilled to know that someone is interested in their work.

To actually get a hold of an author, email is usually the best option. At least one author always includes an email address on the paper. If that address doesn’t work, they may have switched universities, but a Google search will usually bring up their current position with their current email. Failing that, you can try to contact them via Research Gate, but at least for me personally, I find that to be an inefficient way for people to get in touch with me. I don’t get notifications from my Research Gate (because they were obnoxious) nor do I check it often, so when people ask me for my papers via Research Gate, it often takes me a long time to respond. In contrast, emailing me usually results in a response is a few hours. I think this is probably true for most academics, so I’d start with email.

One final note about emailing scientists, sometimes people feel like they are inconveniencing scientists by asking for a paper (particularly people who are not academics or students) so they write them a lengthy story about what they are interested in and why they want the paper. Don’t do that. You don’t need to justify your desire for knowledge and you are just wasting their time. All you need to say is, “Dear Dr X, could you please send me a copy of your paper titled, “Y.” Thank you very much, Your Name” or something to that effect. It doesn’t have to be quite that terse, but academics often get hundreds of emails a day, so keeping your message short is appreciated.

If all of that has failed, you can go old school and drive to a University, go to the periodical room of their library, and read the actual physical journal. It sounds antiquated, but periodical rooms are pretty neat, and some older papers haven’t been digitalized.

Finally, there’s always Sci-Hub. Sci-Hub has often been called the “Pirate’s Bay of academia,” and that is pretty apt. I don’t pretend to know all the details of how it works, but basically, the people who run it got access to a bunch of log-in credentials for journals and have used them to make those journals available to everyone. So, you can go to the site, drop a URL, title, or DOI for a paper, and 99% of the time, a free PDF will open. Is it legal? That is questionable. It has been sued several times, and it has had to switch domain names more than once. In my opinion, however, a more relevant question is, “is it ethical?” and as far as I’m concerned, the answer is, “yes.”

For obvious reasons, I cannot tell you that you should be using Sci-Hub, but I will tell you my personal view on the situation. I think that information should be available to anyone who wants it, and I think that it is wrong for data to be locked behind paywalls (particularly given how much research is publicly funded via tax dollars). I also think that the current publishing system is an unethical scam. Without going too much into the details, scientists have to pay “page charges” to publish in most journals, ostensibly to cover the cost to the journal for their editorial staff (see section later on predatory journals). Then, the journals sell the papers, and, as mentioned earlier, the scientists get no money back. Every single year, millions (probably billions) of dollars of grant money are paid by scientists for the privilege of being allowed to publish our work. Meanwhile, the journals rake in billions of dollars in profit from selling the articles, and in turn, stopping many people from having access to them.

To put all of that another way, the money flow goes like this:

  1. You pay the government via taxes
  2. The government gives a tiny portion of that to scientists to do research
  3. Scientists have to spend a good chunk of that money to publish their research
  4. Journals make billions of dollars in profit by charging you (the public) to access the results of the research that you already paid for via taxes.

It is an insane system that robs scientists of countless amounts of precious research funding that we could be using to actually test new questions, all while preventing many from reading the research that, in many cases, they funded with their taxes. Sadly, scientists are trapped in this system. We have to publish our research, and if we want good jobs, we have to publish in high-ranking journals, which means we have to publish in journals that charge us. Publishers know this and exploit it. Papers often cost $3,000 or more to publish. So, if you want to know my personal opinion about academic publishing companies and whether or not it is ethical to bypass their fees via Sci-Hub, I say screw them. It’s a stupid, unethical system that should be overthrown. Read up me hearties, it’s a pirate’s life for me (here endeth my rant).

Organizing your papers

This is somewhat tangential, but I think it is important. As you read papers, you should be taking notes and organizing your papers in a way that makes it easy for you to find the papers again in the future. There are several reference organizing programs specifically for this purpose, with Mendeley and Endnote being the two front runners. I started using Mendeley years ago (before it was bought by one of the massive publishers I just ranted about) and moving to a new system now would be too difficult to be worth it. Having said that, I’m really happy with Mendeley. It is free unless you need to store an ungodly number of pdfs, and it lets you organize papers in a lot of useful ways. You can create folders in the program to store different categories of papers, highlight the text, and write notes. Most usefully of all (IMO) you can “tag” papers with custom tags, then subset within a folder (or your whole collection) by those tags. For example, you could have a folder called “climate change” and tags such as: models, hurricanes, and heat waves. Then, if you need to look at a paper on hurricanes, for example, you can just subset by that tag. On top of that, you can then sort by title, author, journal, etc., or do a search for text in your notes or the papers themselves. Additionally, Mendeley backs up to the cloud, so you can access your files from any computer with an internet connection. It is very useful, and I highly recommend it (or EndNote or some other program) if you plan on reading lots of papers.

Predatory journals and reading critically

Finally, I need to make an important point about critically assessing the results you get from your searches. First, as mentioned earlier, databases like Scholar may return results other than peer-reviewed articles. So just because it showed up in the results, doesn’t automatically make it valid research.

Second, there are, unfortunately, a large number of “predatory journals.” These are, to a large extent, “pay-to-publish” journals that lack an actual peer-review system. I need to explain what I mean by this carefully, because this is not the same thing as the page charges I mentioned earlier. For real journals, you submit your paper for review with the acknowledgement that you are willing to pay the charges if the paper is accepted. Then, the paper goes out for review by other scientists, and if it is accepted you have to pay the charges. These journals care greatly about their reputation and at least try to keep shoddy research from being published (though see the next two paragraphs). In contrast, predatory journals are not real journals. They don’t actually do proper peer-review. You pay them just to publish any junk paper without critically assessing it. They are frauds and should not be treated as if they are real journals. Sometimes proper scientists get duped by them, but an awful lot of the papers in them are there because no legitimate journals would take them. Spotting predatory journals can be hard, but Beale’s List has a pretty good collection of journals and publishers to watch out for.

Beyond predatory journals, there is a wide range in quality for journals. Some journals aren’t technically predatory, but also aren’t really legitimate. To give a really extreme example, a while ago, a Bigfoot “researcher” was tired of actual journals rejecting their nonsense paper, so they started their own journal (de Novo) and published their “paper” there. I’m sure they reviewed their own paper with the highest of standards (sarcasm). That’s obviously the far end of the spectrum, but there are many journals out there that appear reputable, but actually have a strong bias towards fringe positions and tend to have pretty lax standards for review (looking at journals’ editorial boards, their scope, and their impact factor can be helpful for evaluate them).

Further, even really good journals sometimes publish bad papers. As I have said repeatedly on this blog, the peer-review system is good, but it is far from perfect, so you always have to read critically and look for a consensus of studies. The fact that a study found X doesn’t mean that X is automatically true. Scrutinize the study. Ask questions like, was this published in a reputable journal? Was the sample size large enough? Did they control confounding factors? Did they use appropriate statistics? Then, look at what other studies have found. Look at the entire body of literature rather than cherry-picking a handful of studies that agree with you. If there actually is good evidence that X is true, then you should find multiple large studies that used good methodologies and were published in reputable journals, and you should find few studies that disagree (or the dissenting studies should have small sample sizes, be published in questionable journals, etc.).

In short, databases like Google Scholar and PubMed are wonderful, powerful tools, but with great power comes great responsibility. It is extremely easy to do a quick search, find a paper that confirms your biases, then ignore all other studies and claim that you are right and everyone else is wrong, but it is your responsibility to avoid that temptation. It is your responsibility to be intellectually honest, read papers critically, and carefully examine the entire body of research, not just the studies that confirm your biases.

Key points

  • Google Scholar and PubMed are great databases for scientific research
  • Their advanced search options are very useful for wading through a mountain of literature
  • Citations within papers are also very useful for finding other relevant research
  • Papers that are behind paywalls can be obtained for free by either contacting authors (totally legal) or using Sci-Hub (questionably legal)
  • Some journals are “predatory” and do not conduct a proper peer-review
  • Journals and papers range widely in quality and you should avoid blindly believing the first study that agrees with you. Read critically and look at the entire body of literature.
Posted in Uncategorized | Tagged , , , | 14 Comments

Can we blame climate change for storms like Dorian?

Last week, the world watched in horror as Hurricane Dorian ripped though the Atlantic, leaving devastation in its path, and as usual, debates online have swirled around the same question that arises every time there is a major weather event, “is this climate change?” Many people point to storms like Dorian as evidence of climate change and the very real dangers it poses, but others cry foul and insist that climatologists and science-advocates are being hypocritical and inconsistent. “Whenever we point to a blizzard or cold front as evidence that the climate isn’t changing,” they say, “you all accuse us of cherry picking and shout that ‘weather isn’t climate,’ but when there is a hurricane or a heatwave, suddenly you’re convinced it’s global warming.” This accusation of a double standard deserves some discussion, because there is a small kernel of truth to it, but, as usual (always?) there are key pieces that climate change contrarians are ignoring. So, let’s look at this for a few minutes.

Let’s start by dealing with the part that contrarians get right. If someone’s entire argument is, “it was hot today, therefore climate change is real” or “there was a big heatwave, hurricane, drought, etc., therefore climate change is real” then they are, in fact, conflating weather with climate. I will admit that this line of reasoning is no different than saying, “it was cold today, therefore climate change isn’t real.” However, and I can’t stress the importance of this enough, that is almost never what I see people doing. Rather, the argument usually points to a current weather event as the latest in a trend of changing weather patterns. This is a key distinction for several reasons.

First, we need to consider why it is problematic to conflate weather and climate. Weather is what happens over a short period of time, whereas climate refers to long term trends and patterns. Thus, an individual storm, cold front, etc. is weather, but a pattern of increasing storms, heatwaves, etc. over time is climate. So, when people use a storm like Dorian as an example of the latest event in a trend of changes, they aren’t conflating weather and climate, rather they are talking about climate. When we look at the trends, we see that the average intensity and the proportion of hurricanes/cyclones that are very powerful has increased over time, just as climate models predicted (see note 1: Emanuel 2005; Elsner et al. 2008; Holland and Bruyere 2014; Walsh et al. 2016). There is also a trend of hurricanes hitting further from the equator (Kossin et al. 2014), and a pattern of storms increasingly “stalling” (just as Dorian did; Kossin 2018; Hall and Kossin 2019; see note 2). So, it is completely valid to talk about a storm like Dorian in that context. By the same token, it is completely right and proper to bring up climate change when a heatwave occurs, because there are strong, decades-long trends of heatwaves increasing in frequency, duration, intensity, and the extent of area they affect (Klein Tank and Konnen 2003; Della-Marta et al. 2007; Russo et al. 2014; Tanarhte et al. 2015; Perkins et al. 2012; Habeeb et al. 2015). In contrast, saying “this part of the world experienced a record cold this week” and using that as evidence against climate change is using an isolated weather event, rather than a climate trend. There is no trend of increasingly cold winters. Rather, there is a global trend of increasingly warm temperatures. Indeed, as of the writing of this post (2019), all five of the five hottest years on record occurred in the past five years, and 18 of the 20 hottest years happened in the past 20 years (based in NASA’s data; the remaining two hottest years occurred in 1997 and 1998). There is a strong trend of increasing temperatures, which is why we are justified in bringing up climate change when we experience record-breaking temperatures (no, the warming hasn’t paused, that’s a myth).

This brings me to my second, and closely related, point: the nature of cherry picking. Cherry picking occurs when you isolate and cling to any pieces of evidence that conform to your beliefs while ignoring a (usually larger) body of evidence that disagrees with you. Hopefully from the paragraph above you can see where I am going with this. Citing a particular cold front, blizzard, etc. as evidence against climate change is, by definition, cherry picking, because it is using isolated events rather than trends, and it is ignoring the big picture of increasing global mean temperatures. Conversely, talking about heatwaves, hurricanes, etc. in the context of their trends is not cherry picking. It is literally the opposite of cherry picking because it is about trends and the big picture. On that note, I want to go down a brief side tangent to point out that using regional data to argue against climate change is also cherry picking. No one ever said that every part of the planet will be warmer all the time. Rather, we are talking about global patterns and global temperatures. Citing a few cherry-picked locations ignores that big picture.

By way of example, it would be crazy to isolate one person who smoked their whole life and never got cancer and say, “see, smoking doesn’t cause cancer because they smoked and are fine.” That would obviously be cherry picking, and it would be nuts because it would ignore the overarching picture of the strong trend of smoking increasing cancer risk. That is, however, exactly what climate change deniers are doing when they cherry pick a particular cold front, blizzard, ice sheet, etc.

Now, at this point, someone is inevitably thinking, “yeah, but you can never actually say that a particular cyclone, heatwave, etc. was caused by climate change, so it’s still deceptive to try to blame climate change when one happens.” My response to this is, again, two-fold. First, in many cases, we can actually use statistical techniques to show that certain weather events were so extraordinary that they were unlikely to have occurred naturally. For example, see the 2003 heatwave that caused over 70,000 deaths in Europe (Schar et al. 2004; Stott et al. 2004; Robine 2008). Similarly, studies showed that climate change very likely contributed to the extreme rainfall observed during Hurricane Harvey in 2017 (Risser and Wehner 2017; van Oldenborgh et al. 2017; Wang et al. 2018). There are many storms like this where we can, in fact, say with a high degree of confidence that climate change contributed to them (note: science never deals in 100% confidence).

The second consideration (again, related to the first), is that although we can never say with 100% certainty that a given event was caused by climate change, we can say that climate change is increasing them or their intensity, so we are justified in using them as examples of the dangers of climate change, and whether or not climate change caused any one particular storm is irrelevant, because we know the effect climate change is having in general.

Let’s go back to smoking as an example. Smoking causes cancer. This is well-established. However, that does not mean that everyone who smokes will get cancer, nor does it mean that everyone who gets lung cancer smoked. As a result, we can never say with 100% certainty that any particular case of cancer was caused by smoking, but that is really beside the point. The point is that there is a general trend of smoking increasing cancer risks. Thus there is a high probability that a smoker’s cancer was connected to their smoking, and it makes perfect sense to show people pictures of cancerous lungs and stories of people who suffered from lung cancer after a history of smoking. To put that another way, we can never point to an individual and say with 100% certainty that smoking caused their cancer, but we can point to them as an example in a general trend that we should take seriously. The same is true with climate change. Whether or not a particular storm was definitely caused by climate change is beside the point. The point is that climate change is making intense hurricanes more common, heatwaves more intense, frequent, and long, droughts more common in some areas while floods increase in others, etc. These are serious issues that need to be treated accordingly, and it is completely right and proper to talk about climate change when these extreme weather events occur.

Note 1: It is important to clarify that it is the average intensity and proportion of hurricanes/cyclones that are high intensity (e.g., category 4 and 5) that is increasing, not the total number of hurricanes/cyclones. This is consistent with model predictions.

 Note 2: There is a general pattern of increasing hurricane stalling, but there is some disagreement among scientists about the cause. Several models did predict a slowdown in hurricane transition speed due to climate change, but others predicted no change. So not all scientists agree that climate change is the cause (despite what many think, scientists love to argue and don’t automatically assume that everything is being caused by climate change). I personally think that the evidence more compelling suggests that climate change is the cause, but I will wait for more data before reaching a verdict, and I wanted to include this note since the evidence does not point to climate change as conclusively as it does on other issues (despite what many think, I really do care about accurately portraying evidence and the scientific literature).

Related posts

Literature cited

  • Della-Marta et al. 2007. Doubled length of Western European summer heatwaves since 1880. Atmospheres 112:D15103.
  • Elsner et al. 2008. The increasing intensity of the strongest tropical cyclones. Nature 455:92–95.
  • Emanuel 2005. Increasing destructiveness of tropical cyclones over the past 30 years. Nature 436:686–688.
  • Habeeb et al. 2015. Rising heatwave trends in large US cities. Natural Hazards 46:1651–1655.
  • Hall and Kossin 2019. Hurricane stalling along the North American coast and implications for rainfall. Climate and Atmospheric Science 2
  • Holland and Bruyere 2014. Recent intense hurricane response to global climate change. Climate Dynamics 42:617–627.
  • Kossin et al. 2014. The poleward migration of the location of tropical cyclone maximum intensity. Nature 509:349–352.
  • Kossin 2018. A global slowdown of tropical-cyclone translation speed. Nature Letters 558: 104-108.
  • Klein Tank and Konnen 2003. Trends in indices of daily temperature and precipitation extremes in Europe, 1946–99. Journal of Climate 16:3665­–3680.
  • Perkins et al. 2012. Increasing frequency, intensity and duration of observed global heatwaves and warm spells. Geophysical Research Letters 39:L20714.
  • Risser and Wehner 2017. Attributable Human‐Induced Changes in the Likelihood and Magnitude of the Observed Extreme Precipitation during Hurricane Harvey. Geophysical Research Letters 44:12457–12464.
  • Robine et al. 2008. Death toll exceeded 70,000 in Europe during the summer of 2003. Epidemiology 331:171–181.
  • Schar et al. 2004. The role of increasing temperature variability in European summer heatwaves. Nature 427:332–336.
  • Stott et al. 2004. Human contribution to the European heatwave of 2003. Nature 432:610–614.
  • Tanarhte et al. 2015. Heatwave characteristics in the eastern Mediterranean and middle East using extreme value theory. Climate Research 63:99–113.
  • van Oldenborgh et al. 2017. Attribution of extreme rainfall from Hurricane Harvey, August 2017. Environmental Research Letters 13:019501.
  • Walsh et al. 2016. Tropical cyclones and climate change. Climate Change 7:65–89.
  • Wang et al. 2018. Quantitative attribution of climate effects on Hurricane Harvey’s extreme rainfall in Texas. Environmental Research Letters 13:054014
Posted in Global Warming | Tagged | 1 Comment

Don’t cherry pick your experts

The appeal to authority fallacy is one of the most common logical fallacies in internet debates. It is a favorite tactic among climate change deniers, anti-vaccers, young earth creationists, and pretty much anyone else who rejects “mainstream” science. I previously wrote about this at length and explained when it is and is not a fallacy to appeal to authority, as well as discussing some of the different forms that the fallacy can take. In this post, however, I want to focus on one particular variant of the fallacy and explain why it is actually just a special case of cherry picking and is the result of strong cognitive dissonance and motivated reasoning. This variant occurs when you cite one particular expert as if they are infallible while ignoring a much larger body of experts who disagree with them.

As I have frequently said on this blog, no matter what crackpot proposition you believe, you can find someone, somewhere, with an advanced degree who thinks you’re right. That does not, however, automatically make you correct. Simply having an advanced degree doesn’t guarantee that someone knows what they’re talking about, nor does it make them infallible. On literally any scientific topic, you can find a handful of people who disagree with the rest of the scientific community, but blindly assuming that those people are right is problematic for a number of reasons. Most importantly, science is determined by evidence, not authority. We always need to look at the evidence that has been published in support of a position, rather than the list of names associated with it.

The second problem is a logical contradiction that is inherent to this variant of the fallacy. Namely, this argument implicitly assumes that someone must be correct just because they are a scientist, however, in so doing, it also implicitly assumes that thousands of other scientists are wrong despite the fact that they are scientists. Do you see the inherent problem here? It posits that having the support of a scientist is sufficient evidence that a position is correct, while simultaneously ignoring a much larger group of scientists that don’t support the positions. This is why it is a special case of cherry-picking. This fallacy cherry-picks which people to trust based entirely on personal biases and ideology rather than actual expertise.

Let me use a few comments that were recently left on my Facebook page to illustrate. These came from climate change discussions, so before I discuss them, I want to be clear that the vast, vast majority of scientists agree that we are causing climate change, and there are very few climatologists who disagree. The exact number varies depending on which survey and methodology we’re talking about, but it is consistently in the high 90s (close to 100%; I talked about all of this in more detail here, and I discussed the fraudulent “Oregon petition” that purports to have signatures from thousands of scientists here). The point is that very few climatologists dispute the evidence, so using one of the handful of scientists who disagee as evidence is inherently cherry-picking.

To illustrate, let’s look at the comments. The first of these (orange above) was fairly mild and pointed out that Dr. Patrick J. Michaels disagrees with the consensus on climate change. The second took a more forceful approach by insisting we aren’t causing climate change because Dr. Richard Lindzen (the MIT professor) says we aren’t. Indeed, the commenter mocked the rest of us for saying that a 30-year MIT professor is wrong. According to this commenter, Lindzen’s academic status and experience must make him correct. If we stop and think about this for five seconds, however, it becomes obvious that the commenter was himself laughing at and trying to discredit literally thousands of professors from respected universities from all over the world. Thus, mocking everyone else for disagreeing with one professor makes no sense given that the commenter was disagreeing with thousands of professors. It is an inherently hypocritical and disingenuous argument. Similarly, the first commenter was focusing on the fact that Michaels disagrees with the consensus, while ignoring the fact that thousands of other scientists disagree with him. Do you see the point? If we are going to play this game of appealing to authority, then surely it makes more sense to trust the vast majority of experts rather than a handful of cherry-picked individuals. Or, to put this as a question, why should you place blind, unwavering faith in people like Lindzen, while totally ignoring the vast majority of experts who say he’s wrong?

The answer is simple: confirmation biases. People don’t cling to the words of people like Lindzen because he actually knows more than every other scientist. Rather, they follow him because he says what they want him to say. He gives confirmation to their pre-established views; therefore, they blindly trust him and use his degrees to give their position a false sense of credibility. Indeed, motivated reasoning is so powerful that most of these people don’t even realize the inherent logical contradiction in their views. They don’t see why it is inherently contradictory to say that one person must be right because of their degrees/experience while also saying that thousands of other people with the same degrees and experience are wrong.

There is also another issue here that is worth mentioning. In many cases, the handful of experts who disagree with a “mainstream” position have conflicts of interest or other issues that make them untrustworthy. To be clear, I’m not engaging in baseless speculation here. This is well-established. Case in point, Lindzen has been on the payroll of fossil fuel companies and interest groups for quite a while, including receiving $30,000 from Peabody Coal and $25,000 a year (starting in 2013) from the Cato Institute, a conservative think tank was started by one of the Koch brothers (this was revealed in 2018 court documents that you can read here). Patrick Michaels has also has a lengthy string of connections with the Koch brothers, Cato Institute, etc. from which he has received hundreds of thousands of dollars. This type of situation is the norm for the handful of climatologists who deny anthropogenic climate change. Meanwhile, many climatologists struggle to adequately fund their research, and they make very little money. There are exceptions, of course, but most research comes from independent scientists (see note).

To be clear here, I’m not saying that people like Lindzen and Michaels are automatically wrong because of their conflicts of interest, but those conflicts do mean that we should scrutinize them more closely, and we certainly shouldn’t be placing blind faith in them and assuming that they are actually smarter than the entire rest of the scientific community. Further, when we apply that scrutiny, we find that they both have a long history of making false statements about climate change (examples for Lindzen; examples for Michaels). Nevertheless, the people who appeal to their authority generally have an inflated sense of their credibility (both comments I posted illustrate this).

So, which makes more sense, trusting a handful of scientists most of whom have enormous conflicts of interest, or trusting the vast majority of scientists, most of whom don’t have conflicts of interest? To put that another way, Occam’s razor states that solution that makes the fewest assumptions is usually the correct one, and there are clearly fewer assumptions involved in assuming that a handful of scientists are wrong as opposed to assuming that nearly the entire scientific community is wrong.

Before I end this post, I want to state again that I am not saying that climate change is true because of all the scientists who say it is true. Rather, we know that climate change is true because of the thousands of studies showing that it is happening, we are causing it, and it is dangerous. The consensus among scientists exists because of that consistent body of evidence. That’s how a consensus works in science. First a consistent body of evidence is accumulated, then that body of evidence results in a consensus among the scientists themselves. So, my point is not that we should blindly follow an expert consensus. Rather, my primary point is that we should not blindly follow cherry-picked experts, and my secondary point is that if we are going to appeal to authority rather than basing our views only on the evidence, the consensus among experts is nearly always based on a consensus of evidence, thus making it fairly reliable, and it makes far more sense to trust a view that is shared by the vast majority of experts, rather than blindly following a handful of dissenting voices.

In summary, arguing that a contrarian must be right because of their credentials is inherently logically contradictory, because that argument implicitly assumes that thousands of other scientists are wrong despite having the same credentials. Science is about evidence, not authority, but if we are going to appeal to authority, then surely it makes sense to trust the majority of experts, rather than a few fringe scientists. Finally, I want to make it clear that although I have focused on climate change for this post, the same thing happens on other topics, and it is just as flawed there. Anti-vaccers, for example, love to cite the handful of doctors and scientists who oppose vaccines as if they are infallible, but doing so is illogical and foolhardy. The fact that someone has an advanced degree does not automatically make them right.

Related posts

 Note: Governments are by far the biggest funding source for climate change research, with much of that money going to independent scientists working out of universities. Many people seem to think that receiving a grant from the government is a serious conflict of interest, which has always baffled me (particularly when talking about countries like the USA). Politicians have, historically, been very opposed to the concept of anthropogenic climate change, and in countries like the US it is still hardly a popular political position, and many administrations have refused to accept it or take it seriously. So why would anyone think that money originating from those governments is a conflict that would bias research towards showing that we are causing climate change? Why would a government that doesn’t acknowledge climate change want to fund research showing that we are causing it? You really don’t think that politicians would LOVE a study saying that we aren’t causing it? My point is simple, grants from the government are usually (and correctly) considered to be neutral sources of funding, rather than conflicts of interest, but if we wanted to say that they are biased, surely that bias would be in favor of the fossil fuel companies who spend billions lobbying politicians.

Posted in Global Warming, Rules of Logic | Tagged , , , , , , | 7 Comments

Windows into Science: Scientific Conferences

I am writing this on my way home from attending two scientific conferences. These meetings are critical components of the modern scientific enterprise, and they provide a lot of insights into how science works, so I thought it would be a good idea to briefly explain what they are, why they are valuable, and what they show us about modern science.

Before I go any further, I want to quickly mention that you don’t have to be a scientist to attend these meetings. Anyone is welcome to come. Most scientific organizations have annual conferences, so please find a scientific society that studies something you are interested in and attend one of their meetings. We’d love to have you participate in science with us.

Presenting current research

At the most fundamental level, scientific conferences are a place for scientists to present their current work to the rest of their community, and the bulk of the time at conferences is devoted to talk sessions (usually split into 15 minute slots, where each researcher has 12 minutes to present, ideally leaving three minutes for questions) and poster sessions (where researchers stand beside posters displaying their research and other scientists talk to them and ask questions). Generally (though not always), the research being presented at conferences is from recent or ongoing studies. Many talks include preliminary data or data from studies that aren’t fully completed. Thus, they are a good opportunity to showcase what a lab is doing right now and draw attention to research that hasn’t been published yet.

How many talks are given during a conference depends entirely on the size of the meeting. I’ve been to small conferences that have only a few dozen people, last one day, and have just a single session. These are usually for either regional scientific societies or a very narrow research focus (e.g., research on a single species). Others are huge, with thousands of presenters, several days of talks, and numerous sessions running simultaneously. These usually include researchers on a wider range of topics who come from all over the world to attend.

Regardless of the size and scope, conferences are a great place to draw attention to your current research and keep up with the current research that other labs are conducting. Science is not done in a vacuum, and as a scientist, you have to constantly stay abreast of the current research and update your knowledge as new studies come out. As a result, attending conferences is really important for active scientists.

Maintaining and creating collaborations

Although talks consume the most time at conferences, they aren’t actually the most important component or reason for attending. That honor goes to making and maintaining connections within your field. People often think about scientists as lone mavericks working in isolation, but that simply isn’t how modern science operates. Science is an immensely collaborative process, and the best studies usually involve numerous collaborators from all over the world, each of whom contributes different expertise and skills. So how do you build these collaborative teams? Talking to people at conferences is often the best bet.

This is really the most critical function of conferences. They bring together the majority of researchers in a particular field, thus facilitating collaborations and dialogue. If you take me up on my suggestion that you should attend a conference (which I hope you will), pay attention to the conversations that happen during breaks and over meals. Watch closely, and you will continuously see scientists bouncing ideas off each other, sharing proposals, giving advice, and planning new projects. Sometimes, this is entirely organic. For example, one researcher may see a proposal by another scientist who is doing research that is similar to their own and decide that they should collaborate on a joint project. This happened to me during one of the conferences I just attended. Another scientist presented on a disease ecology system that was similar to mine, and after hearing their talk, I invited them to join me for lunch to discuss a joint project comparing our systems.

In other cases, discussions and meetings may be pre-arranged. For example, two long-term collaborators who live on other sides of the planet will often use meetings as an opportunity to sit down together and discuss the next steps of their joint research.

Finally, meetings are a good place for students to find potential advisers. For example, MS students at conferences are often scouting for PhD advisers, and conferences give them the opportunity to both showcase their own research and talk to potential mentors.

Scientists love to argue

The final thing I want to talk about is a common misconception about science that conferences help to dispel. Many people seem to think that scientists go around constantly agreeing with each other and blindly following the “dogma” of their fields. Nothing could be further from the truth. Scientists are extremely argumentative and critical, and when we read papers or watch talks, we carefully evaluate the methods and results rather than blindly accepting them. Conferences are a great place to see this on display in real time.

Talks are generally followed by a question session, and it is extremely common for audience members (other scientists) to use that time to try to shoot holes in the presentation they just saw. On multiple occasions I have seen a presentation crash and burn during the question session, sometimes with the presenter being brought to tears as the problems with their research were brought to light.

That method of argumentation is, of course, the rude way to do things. The more polite way, which also happens frequently, is to talk to the present privately later in the conference, ask them additional questions, and point out potential problems they may have missed. Ideally, this exchange will benefit the presenter, because, given that conference presentations often include preliminary results, there may still be time to fix the project before it is too late.

Finally, in addition to directly talking to the presenter, scientists talk among themselves about the presentations. I have been to dozens of conferences, and at some point, during every single one of them, I have found myself sitting around with my colleagues dissecting talks, discussing their pros and cons, and talking about issues we had with the methods and/or data presentation. Pay attention at conferences and you will see this happening constantly.

This brings me to my closing point. If you are someone who questions the value and reliability of science, who ignores studies you don’t like, who thinks that scientists blindly agree with the “dogma” of their fields, who thinks that scientists are engaged in some form of conspiracy, etc. then I strongly encourage you to attend some conferences. Meet scientists. Talk to them. Observe how they think and process information. Listen to the questions that they ask and watch how they respond to talks. Scientists don’t absorb information passively. Rather, we actively consume it and engage with it, and seeing that process in action will change how you view science and scientists.

Posted in Nature of Science | Tagged | 6 Comments

Stop using Al Gore as an excuse for rejecting science

There are a lot of bad reasons for rejecting the science of climate change (indeed, there are no good reasons), but some arguments are more well thought out than others. This is not going to be an article about one of those arguments. Rather, I am going to write a brief, cathartic rant about a particularly stupid argument that arises so frequently that it deserves to be its own internet law (much like a climate change version of Godwin’s law).

Before I go any further, I do want to clarify that I am calling the argument stupid, rather than the people who make the argument. It is entirely possible for otherwise rational people to be so blinded by their biases that they make utterly idiotic arguments, and I think that is exactly what is happening here, at least I hope it is, because the argument is breathtakingly stupid. Indeed, it is so moronic that anyone making it has just demonstrated beyond a shadow of a doubt that they don’t understand anything about climate change and are refusing to believe it because of political biases rather than an examination of the facts. So, if you ever find yourself being tempted to use this argument, please reflect carefully on it, and re-evaluate your biases and views. Really ask yourself, “is this actually a good argument?” and if you conclude that it isn’t (as you should) then ask yourself why you were tempted to believe/use it.

The argument in question involves invoking Al Gore and it can take one of two forms, both of which are childish and inane. The first takes the form of, “Al Gore said X, and X wasn’t true, therefore global warming isn’t real.” The second version occurs when someone arbitrarily accuses someone of believing in climate change because Al Gore said its real. See, for example, the screen shot from a recent thread on my Facebook page where someone did exactly that. As usually happens with this argument, at no point did I bring up Gore or anything political, but out of the blue came an accusation that I just blindly follow whatever Gore said.

Both forms of this argument are wrong for the same blatantly obvious reason. Namely, Al Gore is 100% irrelevant to debates on global warming. He is not a scientist, and no scientists care what he thinks. No one accepts climate change because he said so. We accept it because of the overwhelming body of evidence showing that it is real. Similarly, climate change deniers act as if any mistakes that he made indicate that the science itself is flawed, but that is absurd. He doesn’t speak for scientists. He is not the arbiter of scientific consensus.

I care about actual climate studies, not Gore’s reporting of those studies, and the actual studies show that climate models have been very accurate, regardless of what Gore may or may not have said about them (I honestly have never even seen “An Inconvenient Truth”). If Al Gore said that the moon was made of green cheese, that wouldn’t mean that scientists are wrong about its real composition, but that is exactly the reasoning of this insane argument. Or, if we want to use the other variant of this argument, Gore believes that we are breathing oxygen, but it would be absurd to accuse someone of only accepting that we breathe oxygen because Gore says so.

Now, I should pause for a second and clarify that if someone says something like, “climate change is true because Gore said…” then and only then could the climate change denier attack Gore without committing a straw man fallacy (depending on how they structured the argument), but here’s the thing, in all the literally hundreds of climate debates I have been in over the years, I have never once seen someone invoke Gore as evidence for their position. Maybe it has happened at some point, but I have never seen it. Nevertheless, in a huge portion of those debates, at some point, the denier has brought up Gore as if that somehow helps their position. This is what I meant earlier when I said that it was like Godwin’s law. Just as Godwin’s law states that as any internet debate continues, the probability of someone bringing up Hitler/the NAZIs approaches 1 (i.e., 100%), even so, as any climate debate continues, the probability of the denier bringing up Gore approaches 1.

So why is this argument so common? Well, I am neither a psychologist nor a social scientist, so I am poorly equipped to answer that, but here is my best guess from the patterns I have seen in these debates. Although science is inherently apolitical, people often try to make it a political issue, and there is a pretty deep divide between republicans and democrats on climate change (though that divide has lessened in recent years). Thus, even though climate change is not a political topic, many people refuse to believe it for political reasons. As a result, Republicans often associate it with Al Gore, even though Gore is irrelevant to those of us who accept the science. Thus, this argument actually demonstrates their own biases and faulty way of thinking about science, rather than being a valid criticism of the evidence.

In short, there is no reason to bring up Al Gore in a climate debate. We accept that climate change is happening because of the evidence, not because of Gore, and if you actually think that we accept it because of Gore, then you need to carefully re-examine your views, because you are simply displaying your own biases and cognitive errors.

Note: After posting this, someone pointed out that others have had similar thoughts before me and already coined the term “Gore’s law.”

Related posts 

Posted in Global Warming | Tagged , | 12 Comments

DNA is DNA: The anti-GMO movement ignores basic genetics

Genetic engineering (GE) is simultaneously one of the most misunderstood technological marvels we have invented. The internet is full of articles and videos denouncing the supposed evils of genetically modified organisms (GMOs) with one of the most common arguments claiming that GMOs are inherently dangerous because they alter organisms’ genetic codes. It certainly is true the GE works by altering genomes, but what this argument ignores is the undeniable fact that all of our breeding methods work by altering organisms’ genomes. We’ve been genetically modifying organisms for thousands of years, GE simply lets us do it faster and more precisely.

At this point, I can already hear people screaming at their screens, “SELECTIVE BREEDING AND GENETIC ENGINEERING AREN’T THE SAME THING!!!!!!” This is the response I get ever time, often accompanied by statements about the horrors of moving genes from one organism to another. If you are tempted to have this response, then all I ask is that you actually hear me out, because this response ignores very basic concepts in genetics, and I want to talk about those concepts. If you lay aside your biases for a second and walk through this with me, I will show you how the genetics actually works and why GMOs are not fundamentally different from other crops.

I’m going to focus here on transgenic GMOs (i.e., ones that move DNA from one species to another) because they are the ones most people freak out over, but please realize that they are not the only type of GMOs (others simply alter existing genes) and everything that I’m going to say applies to the other types of GMOs as well.

Note for clarity: Obviously GE and selective breeding are not the same thing. No one thinks they are. Rather, the point is that the products they produce are not fundamentally different from each other. They both are the result of modifying genomes.

 Note: I am in no way shape or form affiliated with or paid by any GE companies. In fact, I pay to maintain this blog out of my own pocket.

 DNA is DNA is DNA

 The first major problem with this anti-GMO argument is that it inherently assumes that the DNA of different organisms is somehow fundamentally different. It isn’t. DNA is just a code. It’s a blueprint for constructing and operating organisms, with different genes coding for different proteins, and in all organisms, it is made of the same four bases A, T, C, and G. How you arrange those bases determines which proteins an organism will make and how those proteins will be arranged. That’s it. That is all that it is.

There is no such thing as a “tangerine gene” or a “lizard gene.” There are just genes. We may use taxonomic names to describe where a gene evolved, but there is nothing inherently lizard-like about a gene in a lizard. It is just the sequence that happened to arise and be selected in lizards.

To put that another way, given enough generations, any organism can, in concept, evolve to have any DNA sequence. In other words, a lizard could independently evolve the exact same sequence of DNA that is present in a tangerine. Would that make the lizard some sort of bizarre, mutant tangerine/lizard hybrid? Of course not, because DNA is DNA regardless of what organism it is in.

Think about DNA like an alphabet. In this analogy, the alphabet is the DNA bases and the words are amino acids (strings of bases) that join together to form sentences (proteins). Different books, blogs, etc. have different strings of words, but those strings are not intrinsically tied to a given work. They are just the arrangement that a given author chose. The sentences on my blog, for example, are not some special just because they are on my blog. They are strings of letters, just like every other sentence, and you can shuffle those strings around to make new strings or just copy them and directly paste them into another work (that would be plagiarism, of course, but the point is that you could do it). It’s the same with DNA. Just as all English books are operating off the same alphabet, all organisms are operating off the same four bases, which can be arranged in an infinite number of ways, even allowing different organisms to converge on similar strings of bases.

Mutations and selective breeding

 Now that I have hopefully cleared up some misconceptions about DNA, let’s talk about genetic variation and breeding. First, we need to understand some basics of mutations. I talked about this in detail here, but to be brief, every time and organism reproduces, it copies its genetic code and passes that on to its offspring. However, those copies are rarely perfect, and chance mutations (modifications) arise. There are many different types of mutations such as insertions (where a new base is inserted into the DNA), deletions (where a base is removed), duplications (where a section of DNA is duplicated), inversions (where a section of DNA is flipped around so it is “backwards”), etc. These random mutations are extremely important because they give organisms variability, and both natural and artificial selection require variability.

This illustrates a hypothetical example of how mutations can change a genome, by taking our starting crop and mutating it to have the sequence we want (highlighted in green)

So how does this fit into breeding? To explain that, I want to use the illustration to the right. Imagine that we have a crop, we’ll say it’s tangerines, and we want to breed them to be larger. We’ve already been selecting the biggest crops for a few generations, but now we are running out of variation and things are slowing down. Are we screwed? Well, not necessarily. Let’s imagine for sake of example, that currently our crops have the sequence GCGCTA, but the sequence GCGGTCGTACTA would produce larger crops, with the GTCGTA being the critical section (I’ve highlighted this green in the illustration). Can we get there without using genetic engineering? Well, if we are lucky enough to get the right mutations, and each step of that mutation process has a benefit (like slightly larger crops), then yes, we can get there.

I’ve illustrated one way this could work. During one replication, a duplication could occur and give us another GCG. Then, if we are lucky enough to get a T inserted between the GC, all we need is a deletion of an extra C and we have our GTCGTA. We have, by careful breeding and fortunate random mutations, managed to produce the sequence we want, with some caveats. There’s actually a lot that I need to elaborate on and clarify here, so bear with me.

First, you’ll notice that in my example, there are unintended consequences. We got an extra A in the second base position, and we lost the CTA on the end. This very often happens with selective breeding. Genomes are getting shuffled around each generation, so while you are selecting for one trait, another part of the genome will likely change, and that change may not be advantageous. Indeed, as I’ve written about before, this process once resulted in toxic Lenape potato that produced potentially dangerous levels of solanine. The point being that artificial selection can have lots of unintended consequences.

The second point I want to make is that farmers are not usually aware of what is happening at the genetic level. For the tens of thousands of years that it took us to make our modern crops, farmers weren’t going out each year and collecting DNA samples. They were simply selecting the best plants each year and breeding those. Nevertheless, each generation, this was happening. New mutations were arising, and genomes were getting shuffled around. Every generation, the genome was being modified. Indeed, if you look at our modern crops and compare them to their wild varieties, the difference is staggering. We have used this slow, gradual process to make countless changes; we just didn’t know exactly what changes were happening from one generation to the next (note: pay attention to both the scale of these changes and our lack of knowledge about them, because that will be important later).

The final point here is that this process is inherently very slow. It takes many, many generations, but what if we wanted to speed it up? Can we get new mutations more quickly? Well that is where mutation breeding (aka mutagenesis) comes in. For this form of breeding, we take crops and expose them to radiation or various chemicals to increase their mutation rate. Thus, we can make this process happen much more rapidly, by quickly making many mutations.

The catch here is obviously that it is extremely imprecise. In fact, it is random. We make countless random mutations to a plant’s genome and hope that a good variety gets produced, but somehow, that doesn’t seem to bother anyone. I very rarely meet an anti-GMO activist who also rails against these haphazard genetic mutations. Odd, it’s almost like the anti-GMO movement is based on emotions rather than facts.

Crosses, hybrids, and GMOs

 So far, I have been talking about modifying the genomes of highly related organisms (i.e., a single crop variety), but it is often faster and easier to simply supply a crop with genes from elsewhere, rather than waiting and hopping that they randomly arise (indeed, crosses have been prevalent throughout agricultural history). So, let’s take a look at those.

This hypothetical example shows that crossbreeding varieties, hybridizing species, and using genetic engineering can all produce a product with the desired target sequence (highlighted green), but methods other than GE often involve unintended changes. Note: these are fictitious sequences that I made for the sake of example.

I’m going to use the same example as before. We have a tangerine crop with the sequence GCGCTA, and we want the sequence GCGGTCGTACTA, with the GTCGTA in the middle being the most important (again, highlighted green in the illustration).

Now, if we are lucky, there will be another crop of the same species that already has the GTCGTA sequence. If that is the case, we can simply crossbreed the two varieties. This results in the gene we want from the other variety being inserted into our variety, and, with a few generations of breeding for the trait it produces, we get the crop we are after.

However, just as before, there will likely be unintended consequences. You see, when we crossed those crops, we didn’t just move the gene we were after. Rather, we recombined the entire genome, which means that multiple traits which were not originally in our crop are now there. We have made thousands of changes to the genome. Some of those may be benign, some may be helpful, but some could be harmful.

Note: Again, to be clear, farmers usually aren’t aware of the specific gene sequence they want. Rather, throughout history, we have selected based on visible traits (what is known as the phenotype), but by selecting phenotypic traits like crop size, we were actually modifying the genomes (genotype).

Now, let’s suppose that no other tangerine varieties have what we want, but grapefruits do. As it turns out, tangerines and grapefruits can actually hybridize, despite being different species. Once again, this moves DNA from one variety to another (in this case actually jumping species), and, as before, the entire genome gets screwed with, not just the gene we are after. Further, in this case, the changes are likely to be fairly dramatic since we are swapping thousands of genes between species.

Finally, we get to genetic engineering, and for the sake of this example, let’s say that grapefruits don’t have the gene we want, but lizards do. Thanks to GE, we can take that gene and put it into our tangerine. This accomplishes exactly the same thing as every other method I have talked about, except, unlike all of the other methods, GE very precisely moves just the gene we are after without moving anything else! In other words, we can make one very precise change to the genome rather than making hundreds or even thousands of unintended changes.

This is what boggles my mind about the anti-GMO movement. If the gene was in a grapefruit (as in the second example) no one would have any problems with me swapping entire genomes around via hybridization to get that gene into my tangerine. Even though I would have just moved thousands of genes around with countless unintended consequences, no one would care, but if I very carefully and precisely take just that one gene and move it, suddenly there are riots in the street.

Similarly, if I couldn’t get that gene from a grapefruit, I could repeatedly blast my tangerines with radiation, randomly mutating hundreds of genes until I eventually got the sequence I was after, then I could grow that crop on an organic farm, and almost no one would throw up a fuss about it. No one would be accusing me of “playing God” or making some unnatural “frankenfood.” Yet if I use GE to make the exact same thing, just with far fewer unintended genetic changes, suddenly everyone loses their minds!

Let me put this one final way. Imagine that we are using GE to move the gene, but we are moving it from the grapefruit into the tangerine. Would you have any problems with that? If you would, then why would you be fine with hybridizing those crops? Conversely, if you’d be fine with moving that gene from a grapefruit to a tangerine, then why would you have issues if the gene was coming from a lizard?

The cognitive dissonance here is just unfathomable to me. Again, look at the difference between our crops and the wild varieties they started out as. Countless genes have been changed, and no one cares. Why should carefully and precisely modifying one more make such a big difference?

The reality is, of course, that this is no rational justification for these logical inconsistencies, because the anti-GMO position is not based on facts or logic. It is based on appeal to nature fallacies and appeal to emotion fallacies. Eating something that was designed in a lab feels wrong to many people, whereas eating something that was bred in a field feels natural and normal, even if the lab crop involved only one very precise change, while the field crop involved thousands of imprecise changes as entire genomes were mixed from separate species.

Note: Throughout this post, I have very carefully avoided using the fish gene in a tomato example because I often hear people use that as an example of how bad GMOs are when, in reality, no such tomato is on the market. Therefore, I refuse to perpetuate that misconception.

Note: In all my examples, I have been using a simplistic short sequence of DNA for illustrative purposes. In reality, a gene is typically made of much longer strands, but everything that I have said fully applies to those longer strands.

Natural selection won’t help you

At this point, I often find that people try to wriggle out of the inherent logical blunders of their position by invoking natural selection and claiming that there are checks and balances in nature that just don’t exist in a lab. That argument is, however, utter nonsense. I’ve previously devoted an entire post to this argument, so I’ll be brief here.

First, nature doesn’t care about you. Natural selection does not have checks and balances for your protection. It simply selects whatever traits allow a given organism to pass on the most genes. In other words, it helps the organism in question, regardless of whether or not that is beneficial for you. So, the entire premise of this argument is wrong.

Second, our crops came from artificial selection, not natural selection, and as I explained previously, artificial selection often has unintended (even dangerous) consequences.

Third, GMOs are far more stringently tested than other crops. So not only are they more precise, but they are more rigorously tested.

Finally, this argument hinges on GMOs being novel crops, but, again, all of the methods described produce novel crops. When someone makes a new crop by bombarding it with radiation, that is a novel crop when farmers first start using it. When someone hybridizes two plant species that have never been hybridized before, that is a novel crop when farmers first start using it, etc. Why should GMOs be held to a different standard than any other type of crop? Please explain to me exactly what checks and balances are in place for new hybrids and mutants that aren’t in place for the GMOs?

Different methods, same result

joker meme, GMO mutation breedingThe point that I am trying to get you to grasp is that all of our agricultural methods modify genetic codes and produce genomes that aren’t found in nature. All of them can achieve the same result. However, only genetic engineering achieves the result precisely. DNA is DNA. It doesn’t matter where it came from or how it was moved. Really think about this. If you are going to insist that GMOs are somehow fundamentally different from the products of other methods, then I want you to justify that. Explain to me why it would be ok to introduce the sequence GTCGTA by randomly mutating a crop, but not ok to introduce it by GE? Similarly, why is it ok to introduce that sequence by hybridizing two species, but not ok to do so via GE? Why is the former totally acceptable while the latter is considered mad science?

The simple reality is that we have been modifying organisms’ genetic codes for millennia. Almost none of our food is “natural.” It didn’t evolve naturally in balance with its ecosystem. Rather, we cleared forests and grasslands to make agriculture fields and carefully bred the crops with the deliberate intention of making them unnatural. We didn’t want the small crops nature had to offer, so we modified them. We improved them by altering their genomes to suit our needs. Genetic engineering is simply the most recent development in a process that has been playing out for thousands of years, and it only differs from its predecessors in that it is far faster and more precise.

Related posts

Suggested further reading


Posted in GMO | Tagged , , , | 6 Comments

Climate change denier vs Attenborough: Why did those walruses die?

Image credit: AU Department of Foreign Affairs and Trade

Sir David Attenborough recently lent his voice to a new Netflix series, “Our Planet,” that documents human impacts on the planet, with a particular focus on climate change. Unsurprisingly, climate change contrarians were unhappy about this series and seized any opportunity to criticize it. The criticism that has gained the most traction revolves around a scene of walruses falling off of cliffs. According to the documentary, they were forced onto those cliffs because of a lack of sea ice and their deaths are, therefore, a result of climate change (i.e., a lack of sea ice forced them onto land, where there was little space, resulting in some climbing the cliff and ultimately falling to their deaths [walruses are not well adapted to cliffs]). Climate change deniers took issue with this and insisted that the documentary was fraudulent and it was actually polar bears that killed the walruses. I was tempted to leave this situation alone (and I’m admittedly late to this controversy), but it perfectly displays so many different flawed lines of reasoning and deceptive tactics that I think dissecting it will be instructive (also, no one impugns the honor of Sit Attenborough without me having something to say about it).

The person behind the controversy

To begin with, we should meet the person who started this fuss: Dr. Susan Crockford. She (as far as I can tell) started this on her blog “Polar Bear Science.” Crockford is a self-proclaimed polar bear expert. She is also a climate change denier who has stated that temperatures are not rising significantly and frequently argues that polar bears are doing just fine and will adapt to any decrease in arctic ice. As a result, she is a favorite source for those who deny climate change, and she is frequently cited as an expert who disagrees with the consensus. She also frequently writes articles for the Global Warming Policy Foundation (a pseudoscientific think tank) about how well polar bears are doing. Additionally, she was on the payroll of the Heartland Institute (another climate change denying group).

Here’s the thing though, she’s not a polar bear expert. Indeed, she isn’t an expert on anything even remotely related to climate change. She is an adjunct professor of anthropology (not someone who holds a research position), and, as of this writing, she has published exactly zero peer-reviewed research articles on polar bears. She is not a polar bear researcher. She doesn’t work with them.

On her blog, she states, “I am a different kind of polar bear expert than those that study bears in the field but having a different background means I know things they do not and this makes my contribution valuable and valid.” That is, however, utter nonsense. You can’t just decide that you are an expert on something. That’s not how expertise works. By way of example, I do a lot of research on turtles. I have years of research experience and multiple publications on them. As a result, I get to call myself a turtle expert. In contrast, someone who has never studied turtles and never published research on turtles does not get to call themselves a turtle expert, and they don’t get to assert that not doing actual research has somehow allowed them to know things that I (and other actual experts) missed while doing all of our actual research. You can’t just decide that you are an expert on something. That’s not how this works. Think about it this way, every fact we know about polar bears was found by someone actually studying polar bears. Thus, she cannot magically know things that those people don’t, because they are the ones who discovered everything we know.

Indeed, an actual polar bear expert, Ian Stirling (who, amusingly, she compares herself to) said that she has “zero” authority on polar bears.

I’m taking the time to explain all of this for three reasons. First, this strategy of claiming that a non-expert is an actual expert is an extremely common tactic among climate change deniers and science deniers more generally. They present people like Crockford as bonafide experts so that you will take them seriously when they say something like, (paraphrasing) “polar bears aren’t declining,” but Crockford is not a polar bear expert and claiming that she is one is a lie. She is a quack and a fraud who is mascaraing as an expert.

Second, this arrogant notion that not doing actual research actually makes someone more knowledgeable/qualified is pervasive among pseudoscientists, and it is absurd for all of the reasons I have just explained. You can’t just magically pull previously unknown information out of thin air, and this concept that not being a real researcher is somehow beneficial is utterly insane.

Third, the source of a claim or accusation can help you judge the level of skepticism that you should apply. It is not, in and of itself, enough to determine what is true and what is false, but it is a useful starting point. If, for example, a claim is coming from someone who is an actual expert or who, at the very least, has a well-established record of veracity (like, say, Sir David Attenborough) then you have pretty good reason to think that the claim is likely true. You should still fact check, but the burden of proof is fairly low. In contrast, if a claim comes from someone with a long history of being untrustworthy, then the burden of proof is quite high, and you should be very dubious until you are presented with really clear evidence. In this particular case, the accusation is coming from a non-expert with a known agenda, known connections to science-denying groups, and a long history of making false claims about her expertise. In other words, not a trustworthy source.

To be fair to Crockford, several newspapers did report that two people with more relevant expertise (Lori Quakenbush and Lori Polasek) had questions about the sequence, seemingly largely because such events have been known for decades (more on that in a minute). However, neither of them appear to be making the type of bold accusations that Crockford is making, and their comments are sparse enough that it is hard to say much about them, so I’m going to focus on Crockford’s assertions (note also that other experts have supported the description of events in the documentary).

Note 1: You can read more about when it is ok to defer to authority and when and how you should evaluate the source of a claim in my other posts (for example here and here).

Note 2: Just to be 100% clear, actual experts and actual studies strongly disagree with Crockford’s rosy assessment of polar bear populations. I talked about this with sources here.

The documentary’s claimed

Let’s look at what the documentary actually claimed. It showed a large group of walruses in what is known as a “haul-out,” where they come on land to rest. It explained that this would usually happen on ice shelves, but as the arctic ice melts, walruses are being forced to use land. It then asserted that limited space sometimes forces walruses to climb high cliffs to get away from the crowds. Then came the really disturbing footage. It showed walruses falling off the cliff to their death and said that this was happening as they attempted to return to sea (it mentioned that walruses have poor eyesight). It attributed these deaths to climate change by arguing that the walruses were only up there because of a lack of sea ice which was the result of climate change.

Cockford’s accusations

So what is Cockford’s issue with the sequence? The underlying issue is simply that she is a climate change denier and is, therefore, unhappy about a documentary highlighting the very real damage that climate change is doing. As a result, she and her supporters have been citing the fact that walruses haul-outs have been reported for decades and there are previous reports of them falling from cliffs. Thus, according to them, this is not a result of climate change.

More specifically, she argues that the walrus deaths in the documentary were actually the result of polar bears driving the walruses off the cliffs. Indeed, she insists that the crew was filming a previously reported event where polar bears drove walruses off a cliff. The crew of the documentary has denied that polar bears were responsible for the deaths that they filmed, but Cockford insists that they are lying.

Evidence and bad arguments

There is a lot to unpack with Cockford’s claims, and they illustrate quite a few logical blunders and flawed tactics, so let’s go through this.

First, let’s get some facts straight. We have known for a long time that walruses will haul-out onto land but prefer to use ice (Fay 1982). We have also known for a long time that these haul-out events can result in mortalities from trampling, falling off cliffs, etc. It is true that this is not a new phenomenon. That does not, however, automatically mean that climate change is not playing a role in current haul-outs and subsequent deaths. Indeed, the argument being used by Cockford is simply a variation of the classic (and flawed) climate change denier argument that if something happened naturally in the past, it must be happening naturally now. That logic is a non sequitur. It’s not logically valid. The fact that something happened naturally in the past does not mean that the current occurrence is natural, and what scientists are actually reporting is that climate change has increased the frequency and size of haul-outs. In other words, there would naturally be a low background rate of these events, but climate change is increasing them.

So, where is the evidence for that? Well, we know that terrestrial haul-outs primarily occur when there is little sea ice (Chadwick et al. 2008), we know that climate change is greatly reducing the amount of sea ice available (Meier et al. 2007; Stroeve et al. 2012), and we know that haul-outs can be dangerous for walruses (Chadwick et al. 2008; Fischbach et al. 2009). The logical conclusion is, therefore, that climate change will result in larger and more frequent haul-outs, ultimately resulting in more deaths, and that is what seems to be occurring (Jay et al. 2011; Jay et al. 2012). Indeed, Maccracken (2012) reported that since 2000, the number of walruses using terrestrial haul-outs in the Chukchi Sea (the area where the documentary was filmed) has increased by an order of magnitude.

Side note: “Maccracken” is the best name for a marine biologist ever.

So, where does this leave us regarding the accuracy of the documentary? Attenborough’s facts were correct. Global warming is reducing sea ice, which is causing walruses to increase their use of haulouts, which does result in mortalities. To be completely fair, it is impossible to say with 100% certainty that climate change caused the particular haul-out the documentary filmed, but it is likely that it influenced it, and it is true that climate change is making those events more frequent. So, I really don’t see anything wrong with what the documentary presented.

Moving on, what about Cockford’s assertion that it was actually polar bears that drove the walruses off the cliff? This is a great example of a straw man fallacy that totally misses the point and is designed to confuse people rather than point out a serious issue (it is another common tactic of deniers). Consider, for example, her statement,

“We know that walruses reach the top of cliffs in some locations and might fall if startled by polar bears, people or aircraft overhead, not because they are confused by shrinking sea ice cover.”

This is an absurd statement and, indeed, reveals the irrelevance of her whole argument. No one said that the walruses jumped because they were “confused by shrinking sea ice cover.” The documentary claimed that they jumped while trying to return to the sea, not because they were confused by the lack of ice. Further, the reason for the jump is actually irrelevant. The relevant question is simply, “why were they up there in the first place?” In other words, let’s assume for a minute that Crockford is right and polar bears drove them off the cliff, that wouldn’t change the fact that that walruses are increasingly climbing cliffs because of climate change! In other words, that would only change the proximate reason they jumped, not the ultimate reason they were on the cliffs in the first place (i.e., climate change). Thus, her entire argument is a huge distraction. Whether or not polar bears were present is 100% irrelevant, but by making a big deal about it, Crockford distracts people from the real problem: climate change is forcing walruses to go on land and put themselves in dangerous situations.

Finally, we need to talk about the burden of proof and Crockford’s logically invalid attempts to shift it (another classic anti-science tactic). Crockford has boldly asserted that the crew is lying and polar bears were the proximate cause of the walruses jumping. The crew has responded by saying that polar bears were not driving the walruses that they filmed off the cliff. So, who bears the burden of proof? The burden always lies with the person making the claim (or accusation). In other words, if someone accuses you of a crime, they have to prove that you are guilty, rather than you having to prove that you are innocent. Thus, Crockford must prove that the crew is lying, rather than the crew having to prove that they are telling the truth. Crockford, of course, can’t do that, which leads her into several blunders.

First, she said, “The crew and WWF can show I’m wrong by providing evidence of where the Netflix film footage was shot.” This would, according to her, either prove or disprove that the crew was filming the documented case of polar bears driving walruses off a cliff that I mentioned earlier. At a quick glance, that may seem reasonable, but it is actually an invalid attempt to shift the burden of proof. It is insisting that the crew has to prove their innocence, rather than her having to prove their guilt. That’s not how the burden of proof works.

Second, when the crew refused to reveal their filming location (as is common practice) she said, “I can only conclude, therefore, that the two incidents are indeed essentially one and the same.” This is a logical blunder known as an argument from ignorance fallacy. This occurs when you use an absence of evidence as evidence for your pet position. The fact that we don’t know where they filmed doesn’t automatically mean that it was the location Crockford says it was. Rather, it simply means that we don’t know. Similarly, it is invalid to assume that they aren’t revealing the locations for deceptive purposes (again, scientists and documentary crews frequently don’t reveal precise locations for conservation purposes).


In summary, the documentary showed walruses climbing a cliff and ultimately dying during their attempts to get down. Although these haul-out events have been known for a long time, climate change is making them more frequent and increasing the number of walruses present at them. It is, therefore, completely fair to say that climate change is contributing to walrus deaths from these events and Attenborough was correct to bring attention to that. The arguments to the contrary are bringing assumptions rather than facts, ignoring the ultimate reason the walruses were on the cliffs, and attempting to shift the burden of proof. Further, these accusations are being made by a known climate change denier who pretentiously claims to be a polar bear expert despite having never studied polar bears and having zero expertise on climate change.

Related posts


  • Chadwick et al. 2008. Pacific walrus response to arctic sea ice losses. USGS 2008-3041
  • 1982. Ecology and biology of the Pacific walrus, Odobenus rosmarus divergens Illiger. North American Fauna 74: 1–279.
  • Fischbach et al. 2009. Enumeration of Pacific walrus carcasses on beaches of the Chukchi Sea in Alaska following a mortality event, September 2009. USGS 2009-1291.
  • Jay et al. 2011. Projected status of the Pacific walrus (Odobenus rosmarus divergens) in the twenty-first century. Polar Biology 34:1065–1084.
  • Jay et al. 2012. Wlarus areas of use in the Chukchi Sea during sparse sea ice cover. USGS Marine Ecology Progress Series
  • Maccracken 2012. Pacific Walrus and climate change: Observations and predictions. Ecology and Evolution 8: 2072–2090.
  • Meier et al. 2007. Wither Arctic sea ice? A clear signal of decline regionally, seasonally and extending beyond the satellite record. Annals of Glaciology 46: 428–434.
  • Stroeve et al 2012. The Arctic’s rapidly shrinking sea ice cover: a research synthesis. Climate Change 110:1005–1027.
Posted in Global Warming | Tagged , , , | 6 Comments