Not again? Photographer: Jemal Countess/Getty Images
Not again? Photographer: Jemal Countess/Getty Images

Journalism has only a few “capital crimes” -- offenses where a single instance can kill your career. In fact, plagiarizing and making up quotes or facts are the only two I can think of. They often get lumped in together, but as I’ve been reading and thinking about the accusations of plagiarism against Fareed Zakaria, the differences between them have loomed large.

Making up facts or quotes is pretty clearly a crime against the reader, and possibly against the subject, depending on whether the quotes make them sound bad. The journalist is using the credibility of his or her position to get people to believe something that isn’t true. That untrue thing may be trivial -- what’s possibly most remarkable about Jonah Lehrer’s fabrications is how minor and unnecessary they often were -- but that doesn’t alter the fact that you have deliberately gotten someone to believe a lie.

Plagiarism is a completely different sort of crime, but we often treat it as similar. When someone is caught plagiarizing, the editors often write solemn notes, apologizing to the reader. And yet, in most of these cases, the overwhelming majority of readers don’t care. Other journalists care. But unless you’re a famous columnist or something, the average reader doesn’t even look for the author’s name -- I can’t tell you how many lifelong readers of The Economist have assured me that they remember my work from my time there, even though regular stories in The Economist have never within living memory had any bylines. (They do byline the massive, 10,000-plus word special reports that were called “surveys,” but I never wrote one.) They don’t care if the description of Steve Jobs’ pancreatic cancer was written by someone working for the paper they’re reading, or someone else halfway across the country. They care whether it’s a) accurate and b) good reading.

Plagiarism is something that other journalists care about. And they have a right to. It’s no fun to put a lot of work into a story, and then have some other reporter rip it off. Your boss, especially, has the right to care: He’s paying you to report, not to copy someone else. If he wanted a story written by someone at another publication, it would be cheaper to pay a reprint fee.

And yet consider a practice that is much-lamented among reporters at smaller magazines and newspapers. You put months of work into a big story, developing sources, figuring out how some arcane system works, and carefully writing it all up. You run the story, to much acclaim from your editors and readers. Then a few days or a week later, you stumble onto the website of one of our nation’s major newspapers, and there is your story: same sources, same facts, different byline. One of its reporters has gone through your article and pulled the research reports it took you weeks to find, called all the sources you had to painstakingly develop, and fished pretty much the exact same quotes that you got through strategic questioning, then written it up as an “original story” for their much larger readership, without ever mentioning your name. They’re basically getting even more credit than you, for taking a few days to duplicate work that would have taken them months to do from scratch.

Yet that's “legal” in journalism (though apt to get you sour looks and some passive aggressive questioning if you ever meet your victim face to face) even as plagiarism is verboten, and likely to get you fired. And unlike conning people, plagiarism is perfectly acceptable in other contexts. When I was in tech consulting, it was quite common to see people type out the material from manufacturer spec sheets into proposals. I suppose people could have rewritten the specs in original language, but why on earth would they? Was the client going to care? How about the manufacturer? If you’d raised the issue with anyone in this loop, I suspect that you’d have gotten a puzzled stare.

For my book on failure, I thought a lot about what constitutes failure. One of the most interesting interviews I did was with Charles Bosk, a sociologist who has spent his career studying medical errors. Bosk did his first work with surgical residents, and his book divides the errors into different categories: technical errors (failures of skill or knowledge), judgment errors (failing to make the right decision in a difficult case), and normative errors. The last category includes not being prepared to discuss every facet of your patient’s case, and interestingly, trying to cover up one of the other kinds of error.

Surgeons, he said, view the first two kinds of errors as acceptable, indeed inevitable, during residency. You learn to do surgery by doing surgery, and in the early days, you’re going to make some mistakes. Of course, if you just can’t seem to acquire the manual skills needed to do surgery, then you may have to leave the program for another branch of medicine, but some level of technical and judgment error is expected from everyone. Normative error is different; it immediately raises the suspicion that you shouldn’t be a surgeon.

That’s because normative error means that you aren’t committed to the process of getting it right. That’s why Jonah Lehrer’s fabrications were so troubling, even though most of what has been uncovered didn’t substantively affect the stories he was writing; it just made them punchier, an easier read. But someone who is willing to tell little lies in his writing is demonstrating that he doesn’t care about the journalist’s fundamental duty to be as accurate as possible, which means you can’t trust them not to make up bigger lies as it suits them.

That’s not exactly the problem with plagiarism. Of course, the person you are copying may not have gotten it right. But I frequently refer to stories from other publications (with links and quotations and proper attribution, of course), and that’s always a risk. Plagiarism makes this problem only slightly worse, by muddying the source of any errors that are identified.

Plagiarism might actually fall into Bosk’s fourth category of error, the one I find most interesting: quasi-normative error. That’s when a resident does something that might be acceptable under the supervision of a different attending physician, but is forbidden by the attending physician he reports to. In the program he studied, if your attending physician did a procedure one way, that’s the way you had to do it, even if you thought some other surgeon’s way was better.

In other words, quasi-normative error is contextual. So with plagiarism. In college and in journalism, it’s absolutely wrong, because “don’t plagiarize” is -- for good reason -- in your job description. In most of the rest of corporate America, lifting copy from somewhere else might be illegal if the material is copyrighted, but in many situations, maybe even most situations, no one, including the folks from whom you are lifting the copy, will care. They certainly won’t care if you “self-plagiarize” (as Jonah Lehrer was also accused of doing), and I’m very thankful for that, because I wrote a lot of proposals for my company, and there are only so many original ways to describe a computer network. Yet I’d never copy and paste my own writing for Bloomberg without a link, a block quote and attribution.

The question that I can’t even begin to answer is why people who are obviously forbidden to plagiarize do so anyway. The Internet has made plagiarism a lot easier, but it’s also made it very easy to get caught. Before, if someone wanted to catch you plagiarizing, they’d have to perform superhuman feats of cross-referencing; you were likely to be uncovered only if you plagiarized something very famous, or the original author happened to read your work and recognize their own words.

Nowadays, it’s often as easy as typing any moderately recognizable sentence fragment into Google. A while back, Terry Teachout, the Wall Street Journal’s drama critic, pointed out something fascinating to me: If you type even a small fragment of your own work into Google, as few as seven words, with quotation marks around the fragment to force Google to only search on those words in that order, then you are likely to find that you are the only person on the Internet who has ever produced that exact combination of words. Obviously this doesn’t work with boilerplate like “GE rose four and a quarter points on stronger earnings”, or “I love dogs,” but in general, it's surprisingly true. Given that this is the case, plagiarism seems desperately foolhardy, in addition to being wrong.

But that’s a mystery for another blog post.

To contact the author of this article: Megan McArdle at

To contact the editor responsible for this article: James Gibney at