Toward a taxonomy of "fake news" types
Wednesday, October 11, 2017 at 4:24AM
Dan Kahan

Likely this has occurred to others, but as I was putting together my umpteenth conference paper (Kahan 2017b) on this topic it occurred to me that the phrase “fake news” conjures different pictures in the minds of different people. To avoid misunderstanding, then, it is essential, I now realize, for someone addressing this topic to be really clear about what sort of “fake news” he or she has in mind.

Just to get things started, I’m going to describe four distinct kinds of communications that are typically conflated when people talk of “fake news”:

1. “Fake news” proper

2. Counterfeit news

3. Mistaken news

4. Propaganda

1. What I principally had in mind as “fake news” when I wrote my conference papers was the sort of goofy “Pope endorses Trump,” “Hillary linked to sexual slavery trade” stuff.  My argument (Kahan 2017a) was that this sort of “fake news” likely has no impact on election outcomes because only those already predisposed—predestined even—to vote for Trump were involved in meaningful trafficking of such things.  (Most of the bogus news reports were pro-Trump).

These forms of fake news were being put out by a group of clever Macedonians, who were paid commissions for clicks on the commercial advertisements that ringed their made-up stories. Rather than causing people to support Trump, support for Trump was causing people to get value from reading bogus materials that either trumped up Trump or defamed Hilary.   Because support for Trump was in this sense emotionally and cognitively prior to enjoyment and distribution of these stories, the result in the election would have been no different had the stories not existed.

2. But there are additional species of “fake news” out there.  Consider the fake advertisements purchased by Russia on Facebook, Twitter, Google etc. These were no doubt designed in a manner to avoid giving away their provenance, and no doubt were professionally crafted to affect the election outcome.  I’m inclined to think they didn’t but all I have to go on are my priors; I haven’t seen any studies that disentangle the impact of these forms of “fake news” from the Macedonian specials.

I would call this class “counterfeit news” based on its attempt to purchase the attention and evaluation of real news.

3. Next we should have a category for what might be called “mistaken news.”  The category consists of stories that are produced by legitimate news sources but that happen to contain a material misstatement.

Consider, e.g., the report by Dan Rather near the end of the 2000 presidential campaign that he was in possession of a letter that suggested candidate George W. Bush had arranged for a draft deferment to avoid military service in the Vietnam War. Rather had been played by an election dirty trickster.  This error (for which Rather was exiled to retirement) was likely a result of sloppy reporting x wishful thinking.  At least when they are promptly corrected, instances of “mistaken news” like this, I’m guessing, are unlikely to have any real impact (but see Capon & Hulbert 1973; Hovland & Weiss 1951-52; Nyhan & Reifler 2011).

4. Finally, there is out and out propaganda. The aim of this practice is not merely to falsify the news of the day but to utterly annihilate citizens’ capacity to know what is true and what is not about their collective life (cf. Stanley, J. 2015).  If Trump hasn’t reached this point yet, he is certainly well on his way.

So this is my proposal: that we use “fake news,” “counterfeit news, “mistaken news,” and “propaganda” to refer, respectively, to the four types of deception that I’ve canvassed.

 If someone comes up with a better set of names or even a better way to divide these forms of misleading types of news, that’s great.

The only point I’m trying to make is that we do need to draw these kinds of distinctions. We need them, in part, to enable empirical researchers to figure out what they want to measure and to communicate the same to others.

Just as important, we need distinctions like these to help citizens recognize what species of non-news they are encountering, and to deliberate about the appropriate government response to each.


Capon, N. & Hulbert, J. The sleeper effect: an awakening. Public Opin Quart 37, 333-358 (1973).

Hovland, C.I. & Weiss, W. The Influence of Source Credibility on Communication Effectiveness. Public Opin Quart 15, 635-650 (1951-52).

Kahan, D. M. Misconceptions, Misinformation & the Logic of Identity Protective Cognition. CCP Working paper  No. 164. (2017a), available at

Kahan, D. M. & Peter E. Misinformation and Identity protective cognition. CCP Working Paper No. (2017a). Available at

Nyhan, B. & Reifler, J. When corrections fail: The persistence of political misperceptions. Polit Behav 32, 303-330 (2010).

Stanley, J. How Propaganda Works (Princeton Press. 2015).

Article originally appeared on cultural cognition project (
See website for complete article licensing information.