Last week, the Times columnist David Aaronovitch wrote “Conspiracy theorists make monkeys of us all”. In the article he discusses the danger of denying scientific evidence and how some popular conspiracy theories, such as “the MMR vaccine causes autism” or “evolution is a myth”, have been promoted by populist politicians such as India´s Minister for Education Dr Satyapal Singh, Italy´s Deputy Prime Minister Matteo Salvini, and even Donald Trump. Aaronovitch´s article ends with: “Perhaps in light of this, we should alter the way we teach and talk about science and risk. Evaluating who and what you should trust (their methodology, their peer review process, etc) is arguably more important than learning the life cycle of the mayfly.” So I´m taking up David Aaronovitch´s challenge and today´s article will be about fact-checking. Apologies if any of this feels too obvious.
Firstly I´d like to point at a book written by Carl Sagan: “The Demon-Haunted World: Science as a candle in the dark”. Written towards the end of his life Sagan mounts a spirited defence of the scientific method – his basic argument is that the universe is a wondrous and amazing place, and “making stuff up” is simply not required. I particularly recommend chapter 12 “The fine art of baloney detection” in which Sagan describes a “Baloney detection toolkit”. This toolkit consists of a set of things you should do when trying to prove a hypothesis (things like an independent confirmation of the facts, a repeatable test etc…), and also a list of things you shouldn’t do – a list of “the most common and perilous fallacies of logic and rhetoric”. Some examples are:
- “ad hominem” — Latin for “to the man” – attacking the arguer and not the argument (e.g. The Reverend Dr. Smith is a known Biblical fundamentalist, so her objections to evolution need not be taken seriously)
- argument from authority (e.g., President Richard Nixon should be re-elected because he has a secret plan to end the war in Southeast Asia — but because it was secret, there was no way for the electorate to evaluate it on its merits; the argument amounted to trusting him because he was President: a mistake, as it turned out)
- “begging the question”, also called “assuming the answer” – (e.g. The stock market fell yesterday because of a technical adjustment and profit-taking by investors — but is there any independent evidence for the causal role of “adjustment” and profit-taking; have we learned anything at all from this purported explanation?)
I won´t go through the whole toolkit in this article, but I highly recommend reading it. You can find a good summary here which should take less than 20 minutes to get through. I´ve written about the Baloney Detection toolkit before, here, in an article for thrive.com which discusses how the toolkit can also provide a good framework for how to conduct constructive debate.
Sagan´s book is 20 years old , so doesn´t discuss social media or internet-related fake news, but nevertheless the principles described are just as applicable today. However, many people I meet at workshops often talk about internet fake news or the rumours or propaganda spread using social media so it´s worth spending some time on internet information sources. The internet makes fact-checking easier than ever before, however, perhaps because of the huge (some would say daunting) number of news and reference sources available and the deluge of advertising, propaganda and spoof information that each of us are presented with everyday, it can sometimes seem hard to see the wood for the trees. So I would offer the following simple advice: always take note of the origin. As Sagan states “independent confirmation of the facts” should be possible. So unattributed stories or facts should raise alarms immediately. A true fact or story should have a source. Once you have the source you should be able to double-check with other independent sources. Independent sources should mean sources that do not share common interests. A story that is only reported by press all supporting the same political agenda should give cause for concern. If you see a story in one news site double-check it against other news sites. Do they all use the same source? Is there corroboration from different witnesses? Are there different pictures or videos of the same event? TV news tends to be more tightly regulated than print media which tends to be more tightly regulated than internet-only so if the story is only appearing in one internet-only news source then you should start to “smell the baloney”. Note this doesn´t mean it isn´t true, it just means you can´t verify it and should be wary of repeating it, or using it as the justification for any future action. It´s also worth looking at newspaper websites in different countries. For example, the UK has stronger libel laws than the US, but the US has more “freedom of information” – a story reported across borders is likely to have had more “fact-checking” applied by not just the journalists but also editors, lawyers etc…
There are also some useful rules of thumb that can be applied. Internet-only businesses, especially social media organizations, survive through advertising, so attracting volumes of users is vital to their business. Dramatic and interesting headlines (so-called clickbait) attract more users. This is why you get these 1-20, or even 1-50, lists appearing on websites, with an attention-getting headline then requiring you to click through interminable pages of adverts with one, normally fairly dull, point or picture on each page. Other methods used to attract users will be photos or stories of popular or controversial celebrities or of weird or unusual phenomena – and less reputable content providers may sacrifice truthfulness for clicks. So “dull” is more likely to be true than “interesting” – especially if this headline is not appearing elsewhere. Sagan mentions “Occam´s razor” which states that: “When presented with competing hypotheses to solve a problem, one should select the solution with the fewest assumptions.” My own version of this is: “cock-up is more likely than conspiracy”. In other words, when something bad happens it is more likely the result of someone making an innocent or accidental mistake rather than a malevolent conspiracy. Note that this doesn´t preclude the possibility of a complex reality, or, indeed, a conspiracy, it´s just more likely that the simple explanation is true.
Finally, some websites that can help. Using any of the internet search engines (google, bing etc…) to search for quotes, sources, journalists, headlines and so on can be quite revealing and give you an indication of likely veracity. This can also help reveal the origins of unattributed facts or stories that have appeared in social media timelines. Using image search (google image search works well) will also often reveal faked or photoshopped pictures. Good old Wikipedia is also very helpful in providing background and context, although this is worth caveating. At any given time, Wikipedia is bound to have a large number of inaccuracies as the site is constantly being updated by the very large group of volunteer content providers. However, the same group are also constantly moderating each others work, and, indeed, anybody is free to report any inaccuracies spotted on the site. Wikipedia goes to great lengths to urge contributors to add references to sources for the information included on the site. It is often worth checking out the references listed as this can give you a good idea of how well researched this particular Wikipedia article is. A missing reference or broken link is often an indication that an update of the article is overdue. fullfact.org is very good for checking information in the UK economic and political realm. snopes.com is good for those urban legend type stories – this is where I got the giant grasshopper story. This is fake news, by the way, just in case anyone was wondering. The repetition of the same faked photo is a clue, and then it´s not so hard to find out from elsewhere that the picture dates from the 1930s rather conflicting with the 1991 date of the newspaper. snopes.com is also a very good place to check those social media clickbait type stories that often do the rounds.
Good journalists and researchers do all of this and more. As they are often trying to be the first to get to a story, their confirmation of facts has to come by interviewing witnesses directly, cross-referencing to check and by bearing witness themselves. They should be clear about how they obtained their information, and provide caveats on information that can be challenged. If you don’t see evidence of this in the stories they provide then this is a warning sign as to their truthfulness.
Aaronovitch finishes his article with the warning that unless we up our game with regard to “who and what we should trust” then “the Dr Singhs of this world may make monkeys of us all”. I worry he may be right, but I also know I´m not the only person determined to evolve in the opposite direction. I hope my article above is a small step towards that goal.