33 Comments
Jan 10·edited Jan 12Liked by Dan Williams

In the legal field we already make a distinction between libel & slander, parody humour & entertainment, sincerely making a mistake when you attempt to report the facts, negligently ignoring the facts when you write a story, and having an opinion that other people disagree with. A great deal of what happens in lawsuits revolves around the intent of the person who wrote or said the thing that is the subject of the lawsuit. And where it was said -- what is libel when written by the NY Times isn't slander when it is a conversation you are having with your mates at the pub. You are evaluating the speaker, and not just the speech. If you could just evaluate the speech, then our legal system would be quite different.

You do not get around these problems by inventing a field of misinformation. It used to be that journalists spent much of their time analysing what politicians and public officials said, to see if they were lying, or at best making claims that did not hold up under scrutiny. These days, 'misinformation specialists' analyse what the general public has to say to see if it contradicts or calls into question what politicians and public officials say. Instead of the press speaking truth to power, we end up with the censors speaking power to truth.

Expand full comment
author

Beautifully put.

Expand full comment
Jan 31Liked by Dan Williams

We can definitely establish a fixed category of 'misinformation'... once we know absolutely everything and have settled all questions (including those relating to values).

Expand full comment
author

Ha - yes.

Expand full comment
Jan 14Liked by Dan Williams

The issue with media coverage often lies not in the falseness of what is reported, but in the selectivity of their reporting. This was humorously illustrated when someone joked that a photographer covering the civil unrest following George Floyd's death would send footage of the riots to Fox News and clips of the peaceful protests to MSNBC.

Expand full comment
author

Ha - yes, good example, and perfectly illustrates how people/sources with different agendas can select, omit, frame, and package facts in ways that support radically different worldviews and evaluations.

Expand full comment
Jan 13Liked by Dan Williams

This was a very interesting article, it changed my thinking. A few thoughts and perhaps a point of disagreement:

>"The concept of misinformation was intended to pick out an aberration in the informational ecosystem. However, the kinds of tactics that underlie misleading content - cherry-picking, removing context, subtle framing devices, selectively consulting congenial experts, hyperbole, and so on - are so pervasive that any attempt to distinguish between misleading content and non-misleading content will end up looking hopelessly arbitrary"

Why is it hopelessly arbitrary? I will grant that it's hard, but if you identify the "fingerprints" of misleading information (whether by accident or on purpose), and you learn to spot those sorts of tactics and to recognize the biases in yourself too, then whenever you come across something that may be misleading information (misinformation) a flag goes off in your head, and you try to fact check the claims. Granted this is very hard to to objectively and consistently, but it doesn't strike me as impossible. More over, pointing out content of this nature and teaching people/institutions about how to deal with it, doesn't necessarily entail making factual claims about, say, the degree to which vacciens work or have sideeffects.

It's also a matter of degree. You're right that by stretching the category far enough, any form of communication can be seen as misinformation, but we don't need perfect definitions to be able to study something. The boundary between different species of animals, is a sliding scale. In some ways, every mammal is part of the same family, but we can distinguish between them pretty well most of the time, and it makes communication and science simpler. Why can't we do this for "classes of information"? Surely, some people will disagree about where exactly to draw the line for ideological reasons, but that doesn't stop most researchers from creating good demarkations that can be applied across ideological lines.

>"Finally, it is difficult to see how assessments of which acts of communication are misleading could ever be unbiased. Unlike the identification of clear-cut examples of misinformation, determining which content is misleading – problematically selective, stripped of relevant context, and so on – seems highly vulnerable to the kinds of biases, prejudices, and partiality that constitute ineradicable features of human cognition."

The comparison between researchers' categorization of the amplification of vaccine-related deaths as misinformation vis-a-vis the mainstream medias selective reporting of police shootings of black unarmed men, is very apt. And here, I would say that both of these kinds of reporting, and partisan content created around them, can be viewed either as factual reports or as misinformation depending on how and where they are presented. Intentions matter, and the style of communication matters. As you know, classicly misleading but true content often works by having the reader fill in the gaps and making faulty conclusions. The extent to which a piece of content lends itself to faulty conclusions, by virtue of the degree to which it's based on selective, emotionally charged, cherry-picked facts or framing, can surely be a part of an objective assessment of whether or not to classify something as misinformation.

Nevertheless, I agree with you that the classification of misinformation, whether in research or by governments, should not be based on current trends, public opinion, or ideological beliefs.

So all this is to say that I'm more optimistic about the study of misinformation as a scientific endeavor. This doesn't mean that the scientists themselves should be the artibors what is true and what isn't, but that they can study the ways in which human communication, human biases and media contribute to the spread of erroneous beliefs. Perhaps this shouldn't be called "Misinformation science", and it should instead be part of more established sciences within social science, and psychology, say. But just because it's hard to draw the lines, doesn't make it unscientific.

Expand full comment
author

Great comments here. I mostly agree, although a few points of pusback: I am quite sceptical of the idea of simple surface-level "fingerprints" of misinformation (see, e.g., https://www.bostonreview.net/articles/the-fake-news-about-fake-news/). That's something I will be writing about this year, so I will look forward to your thoughts then.

I also share your optimism about the study of misinformation in the sense of a scientific study of the ways in which human psychology and specific social systems interact to produce biased and misleading content and generate false beliefs. (I have tried to contribute to that study in my own published research). My pessimism doesn't concern that project, but rather the project of delegating the task of deciding what constitutes misinformation to a class of misinformation experts. I should have been clearer about that distinction, because I think it's a helpful one.

Expand full comment

Thanks for the reply. And I can't really say I disagree with that. I look forward to reading more about this from you!

Expand full comment

Thank you for this piece!

I'm a journalist and I am very disturbed by this practice of fact-checkers simply defining what is true and what is not, even when the truth is so dificult to achieve.

This doesn't help journalism and journalists to be more realible.

In fact, It does more harm than good in my opinion.

I wrote something about this theme of journalism, journalists and misinformation in Brazil, but I couldn't hit the right tones as you did (obviously).

So, thank you!

Expand full comment
author

Thanks for the kind words Gabriel - glad you enjoyed it.

Expand full comment

An interesting argument. I wonder though how can you quantify the "degree of alarmism surrounding misinformation". Surely this is an exercise that will suffer from the same lack of scientific precision.

Also, this leads to some pretty clear recommendations for policy interventions. For instance, an information system that has much different incentives than social media networks.

Expand full comment
author

Good point. Difficult to quantify, but I would say that listing "disinformation/misinformation" as the number 1 threat faced by humanity over the next two years (https://www.conspicuouscognition.com/p/misinformation-and-disinformation) is a clear-cut case of alarmism. I think my main worry when it comes to expansive definitions of misinformation is not imprecision as such but the ways in which imprecision interacts with other factors to ensure that decisions about which misleading information to focus on will inevitably be problematically selective.

I'm sympathetic to your point about policy interventions, but I think the policy interventions guided by misinformation alarmism are often misguided. For example, social media networks often *reflect* pre-existing problems in society rather than cause them, which means that tring to change social media networks will often have very limited impact (and might have problematic second-order effects, such as people becoming frustrated at top-down interventions in informational ecosystem that they view as one-sided). Moreover, trying to change the incentives on social media is really hard to do - make it boring and people just won't use it. I go into some (but not all) of these issues here: https://iai.tv/articles/misinformation-is-the-symptom-not-the-disease-daniel-walliams-auid-2690

Thanks for the comment.

Expand full comment
Jan 14Liked by Dan Williams

My main complaint about e.g. The Guardian is that it's full of opinion pieces, and I don't find a random Guardian journos opinion at all enlightening.

But they rarely lie about actual facts, in the cases where the article.contains something that looks like a verifiable fact. (Cherry picking of facts, on the other hand, they do). "I hate Jeremy Corbyn" is an opinion, not a fact. And, probably, it is true that the journalist hates Jeremy Corbyn.

Expand full comment

You don't specifically touch on the information/misinformation dimension of 'news'/'current affairs' journalism but it is something very relevant to your thesis here. 'The News' has always been one of the great delusions of modern times. It has always been a fraud on account of the inevitable Editorial Selectivity (whereby for instance, some murders warrant months of agonising and outrage whereas others don't even get a mention). Hence the very CONCEPT is a fraud........the illusion that you can know what the most important thing going on in the world is (without any effort on your part) just by checking 'The News'. There has only ever been one way to be genuinely informed about the world beyond one's own direct experience.....putting in the time and effort to read widely.

Expand full comment
author

Completely agree that following the news is a terrible way of trying to build an accurate model of reality. That's one reason why I find it funny/problematic that many misinformation researchers often simply define mainstream news as reliable and non-misinformation, despite the fact that its audience often ends up very misinformed about the world. In fact, on a very broad definition of misinformation, one would have to include mainstream news and current affairs journalism precisely for this reason.

Expand full comment
Jan 11Liked by Dan Williams

But then, how *can* science help?  If science is the systematic study of nature, then in the view from Mars, the anthropic information ecosystem is a part of nature on Earth that is subject to scientific study.  Scientific understanding enables technologies and means for evaluating design principles that *could* improve the human epistemic condition, as it has our physical and biological conditions.

Some of the names for this value-neutral nascent scientific discipline are, "collective social cognition", "network epistemics", "cognitive security", "social physics".

Absolutely, *use* of scientific understanding is value-laden.  One way that value differences can be partially bridged is by neutralizing their ideological components by replacing mere assertions about policy outcomes, with predictive models.  *If* such-and-such policy is implemented (e.g. content moderation by reputational scoring), *then* measurable outcomes will be thus-and-so (e.g. containment of aggressive & trolling behavior on social media.)

When it comes to physical phenomena, the scientific method succeeds because the mental domain of theories and models lies in close proximity to hard reality.  The challenge of the social sciences is much greater because so much of the reality at stake consists of the relatively arbitrary constructions of attitudes and beliefs that reside in people's heads.  Hence, one person's misleading narrative is another's speaking truth to power.

Looking forward to reading more of your thoughts on this. It is important for scientifically-minded people to make positive, constructive contributions.  Reflections and critiques on the positioning of scientific methods with respect to the subject of study are a vital part of the process.

Expand full comment
author

Great comment. Completely agree that scientifically-minded people should strive to make positive, constructive contributions, and I think that some forms of scientific investigation into our collective epistemic systems and institutions are not just possible but very important. (I research this general topic myself). What I am sceptical of is delegating the task of classifying what constitutes misinformation on a very expansive definition to a class of misinformation experts. I will try to make this clearer in my follow-up to this week's post.

Expand full comment
Jan 11·edited Jan 11Liked by Dan Williams

The irony of the 9/11 Truther movement is that the truth about what Bush/Cheney did in the aftermath of 9/11 is so much worse than the crazy conspiracy theories people concocted about the events of 9/11. Bush/Cheney invaded a country while torturing detainees all the while selling America out to China. Bush’s brother and son of HW Bush has been on China’s payroll for over 20 years…NOTHING TO SEE HERE!! PLEASE MOVE ALONG!

Expand full comment
author

It is noteworthy that the lie that Iraq had WMDs was far more consequential and damaging than conspiracy theories about 911, and the former was supported by elite establishment institutions...

Expand full comment
Jan 10Liked by Dan Williams

If conspiracy theories are the symptom, not cause, of a loss of trust in government - then I think the loss of trust is a big, big problem. Like, for example, in the case of covid you need to be able to convince people that the vaccine is safe and effective - because the FDA says so, and the FDA is trusted by the public.

Here is case where lack of institutional trust can directly lead to mass deaths, which I would argue is sufficient justification for declaring it to be a serious problem. (If anything else killed that many people, we'd think it a serious problem too - well, except maybe road traffic accidents).

But, possibly, misinformation researchers are just looking at the wrong thing. The question of why people are loosing trust in the government is curiously underexamined by misinformation researchers.

Expand full comment
author

Yes exactly. Distrust of establishment institutions is an incredibly big problem, adn the vaccine case illustrates that well. But in my view looking to *misinformation* as the cause of that distrust is often misguided and simplistic.

Expand full comment

The right wing echo chamber didn’t just turn against the vaccines they also turned against masking and social distancing and lockdown. So the southeast could have had tens of thousands of fewer deaths had the Republican governors behaved like Cooper did in NC and simply allowed the public health authorities to do their jobs. And we’re talking about a few more months of masking because Omicron wave was really the point where the virus became much more transmissible while becoming less severe.

Expand full comment

However, we now know that masking, lockdown, social distancing and even the vaccines were ineffective. I think that “misinformation” is often we don’t yet have enough information to know. I am not an anti-vaxxer but since the covid vaccine debacle, I would hesitate to accept any medical intervention as soon as introduced. Lying to the public on a large scale has created enormous problems going forward.

Expand full comment

We literally have reams of data that shows masking and social distancing and lockdowns were effective at slowing the spread…it’s why Hawaii has the lowest Covid death rate and Arizona which did the fewest public health mitigation measures in 2020 has the highest Covid death rate.

Expand full comment

Let’s see that “reams of evidence.” The Cochrane review of masking was pretty convincing.

Expand full comment
Jan 10Liked by Dan Williams

In place of a science of misinformation that you’re critiquing is the public policy rx that follows to double down on educating people, especially young people, to be discerning, skeptical and epistemically humble information gatherers who are, for example, habitualized to assess the agenda and other incentives (another potential area of misinformation) motivating sources and to cross-check with other reliable sources? Is this subject getting the attention it deserves in the typical education?

Expand full comment
author

Yes I agree that's very important - although I'm inclined to think the more important challenge is not just to improve individual information processing but to build better, more trustworthy epistemic institutions and to win trust in them.

Expand full comment

Misinformation research is a solution looking for a problem.

The solution is censorship.

The real problem is believing in an ideology that does not conform to reality.

https://frompovertytoprogress.substack.com/p/why-do-ideologies-fail

Expand full comment

You run into a problem right off the bat. You write "Misinformation...is simply false information that’s being spread. This is a common definition. Charitably, what it means is something like “demonstrably false information.”

This is pretty much a useless definition. For example, can you demonstrate that the idea that the earth is flat is false? This is a hard thing to actually demonstrate to an unbeliever.

A more useful definition of misinformation would be asserting extraordinary claims without evidence as truths. Example, Ancient or modern-day Aliens visiting Earth, Creationism, Donald Trump won the 2020 election, Unchecked global warming will make humans extinct, Covid vaccines are more dangerous than Covid, genetically engineered crops are dangerous, etc.

An extraordinary claim is something that is unusual, newsworthy. It's the man bites dog as opposed to dog bites man. For example, vaccines. The purpose for vaccination is to reduce the harm from getting the disease. For developing a vax to make financial sense, the harm from the vax must be less than the disease. Thus, one would expect anyone making a vax must have good reason to believe that it causes less harm than it prevents. Because if they are wrong, they could go bankrupt or worse. So, a claim that the vax actually does cause more harm than the disease would be an extraordinary claim, in the man bites dog category and requires extraordinary evidence.

Expand full comment

It’s good you added the widespread publicity about the rare police shooting, too bad you’re not quite willing to point out the two year set of NYT headlines and articles about 2016 Russian Collusion, which was a Clinton campaign hoax against Trump, supported by a Dem Deep State FBI.

Nor, 3 months later, point out the false NYT headline about Israel bombing a hospital in Gaza.

There is also the issue of T Truth. What IS a woman, and who defines it? An XX female who identifies as a woman is one, but can an XY male become one? I claim no, but others claim yes. What is false info when definitions of what words actually mean are in dispute?

And where is the burden of proof? For legal issues, it is on the prosecution, which may fail to prove its case with a Not Guilty verdict. That’s taken as, yet is not really the same as, Innocent.

Expand full comment
author

I'm willing to call out bias, misperceptions, misleading communication, etc., on any side of political divide. The point about rare police shootings was just one example - certainly not the only one - and I agree lots of the propaganda about 2016 Russian collusion was also bad and highly misleading.

Expand full comment

Lol, RussiaGate was pushed by Bush Republicans Trump foolishly surrounded himself with. The partisan media just behaved like they have been behaving since around 2000. So the government official that appointed Mueller was appointed by Trump—Rosenstein.

Expand full comment