How And Why Lying About Plagiarism Is Bad – A Response To Fareed Zakaria And Fred Hiatt

Yesterday, Our Bad Media posted an article about a pattern of lifted text in the work of Fareed Zakaria in TIME, the Washington Post, and CNN that came after all three outlets claimed to have reviewed Zakaria’s work for plagiarism. While the outlets are all standing by their previous reviews of his work, TIME’s statement is notable for highlighting the fact that Zakaria no longer works for them. Politico’s Dylan Byers received a response from Zakaria himself that defended his work by outright lying and obfuscating about what we posted. We understand that the global brand Fareed Zakaria brings to the table is considerable, but that doesn’t make his explanation any less bullshit.

We’re standing by our previous piece. Below, we’ve posted the entirety of Zakaria’s reply with our responses:

In August 2012, CNN, Time and the Washington Post separately conducted extensive reviews of my commentary. As one part of this process, they ran my columns and cover stories (which span over 20 years) through software programs designed to detect plagiarism. All three informed me that the reviews cleared me fully. Two anonymous bloggers today have alleged that there are 11 cases in my writing where I have cited a statistic that also appeared somewhere else. These are all facts, not someone else’s writing or opinions or expressions.

The bolded is patently untrue. What our investigation works to show is not that Zakaria forgot to cite simple statistics in his pieces, but rather that he lifted extensively from the work of others. We already know it’s a problem because lifting entire passages of facts from other writers is exactly the kind of behavior that got him suspended in 2012. Take the highlighted words in the example below. Zakaria told Byers that he tries to go “as close to the original source as possible” in his work. Then why would he take 44 of David Leonhardt’s words, paste them in the exact same order (minus the words “well-known”), and only remove the sentence mentioning the Capital IQ analysis that was done for the New York Times? 

Or take this example, where Zakaria happens to write 11 words in the exact same order as they appeared in a Peter Beinart article.

Or take this example. Vadim Nikitin writes a 39-word summary of a Levada Center study; Zakaria takes it, deletes the words “May/June” and “notorious,” adds the word “Russians,” and pastes it into his TIME column.

Other examples may take closer scrutiny to see wrongdoing. But the examples above clearly show what Zakaria professes they don’t: “someone else’s writing or opinions or expressions.” Just because another journalist’s phrases happen to include statistics does not mean they are free game for cribbing without attribution.

For example, in one column, I note that the national debt tripled under Ronald Reagan. The bloggers point out that this is also in Wikipedia’s Reagan entry. But it is also in hundreds of other articles, studies, and reports — just Google the phrase. Until today, I had never read the Wikipedia entry for Ronald Reagan. As it happens, it is incorrect. (There is a difference between “public debt” – Wikipedia’s words — and national debt.)

There are three problems with Zakaria’s defense. First, we didn’t merely point out that Zakaria used a certain statistic. We demonstrated that he did so using phrasing very similar to that of a Wikipedia entry. In the image below, you can see that Zakaria merely remixes same words written by Wikipedians – changing “during” to “under,” moving “1971 to 2009” around in the first sentence, etc.

Second – and more crucially – Zakaria leaves out that plagiarism isn’t narrowly defined as taking the same words from another source. It’s also using the same ideas. Basic facts obviously don’t count as “ideas,” but synthesis & comparison of facts do. Whoever wrote the Wikipedia article had the idea to take the spending under Reagan and compare it to spending from 1971 to 2009, and next, take the public debt in 1980 and compare it to the public debt in 1988. In that order. Zakaria’s article presents the exact same idea in a very similar presentation. That’s subtle, but nevertheless textbook, plagiarism.

Third, Zakaria’s implication that the difference between public debt and national debt (for the uninitiated, national debt = public debt + intergovernmental debt) makes his piece divergent from the Wikipedia article is bizarre. If he thought there was a difference, why did he cite the same numbers ($712 billion to $2 trillion) for the national debt that Wikipedia uses for the public debt? Here’s the kicker: by changing “public debt” to “national debt,” Zakaria is the one who ends up with an “incorrect” article. According to the CBO, public debt went from $712 billion in 1980 to $2 trillion in 1988. According to the Fed, national debt went from $909 billion in 1980 to $2.6 trillion in 1988.

My usual procedure with a piece of data that I encounter is to check it out, going as close to the original source as is possible. If the data is government generated (GDP, spending on pensions, tax rates, defense spending, etc) then I often don’t cite a source since it is in the public domain.

If Zakaria had done his research, “going as close to the original source as possible” (instead of changing a word to try and cover his tracks), he would not have made the error he did.

If it is a study or survey produced by a think tank, then I usually cite the institution that conducted the survey. In many of these cases, there was a link in my column to the source. This was not always possible, however, because Time magazine, for example, did not always allow for links.My columns are often data-heavy, so I try to use common sense, putting a source into the text when it was necessary.

Two problems here. First, Zakaria laying the blame for failure to attribute at the feet of TIME “not always allow[ing] for links” is about as bad as it gets. Any journalist knows that medium is no excuse for citation problems. If you can’t link, simply put “according to.”

Second, even if we allow all the instances of errors in TIME to fly because of a problem with the medium (which we shouldn’t), then an error like the one below is still inexcusable because it appeared in one of his online Washington Post columns, which are usually chock-full of hyperlinks. As you can see below, none of the information lifted from Korb et. al. is hyperlinked back to their original study.

 

Even when Zakaria does link to Korb, it does not go back to the Korb et. al. CAP study he cribbed from, but rather a CNNMoney op-ed by Korb that has none of the information Zakaria appeared to lift from the CAP report.

In many of the columns cited by the bloggers, I found the data they refer to in a primary source not the secondary one that they highlight. For example, in my column that mentions Greece’s debt, I noted that “one estimate” suggests that Greece has been in default for half of its existence since its independence. The bloggers found a Businessweek article that had the same fact. But I didn’t get it from there; I never read that article. The “estimate” I refer to is in the scholarly book by Carmen Reinhart and Kenneth Rogoff, “This Time It’s Different” – which I did read. The Businessweek article, incidentally, does not cite the Rogoff book or any source at all for the same fact Does that mean that the Businessweek author was claiming that he was the source for it? No, of course not, because it was a fact in wide circulation at the time.

First, Zakaria says that we’re accusing him of using the same primary source as Businessweek for the information about Greece’s debt. We’re not. What we show is that Zakaria takes the same information as Peter Coy, changes a few key words, and puts it in his article. “Since winning independence from the Ottoman Empire in 1832, the nation has” is changed to “Since it gained its independence from the Ottoman Empire in 1832, Greece has” while “spent half its time in various stages of default or restructuring” is chopped in half and reversed, leaving, “has been default or restricting for half this period.” 

 

Second Zakaria is- well, there’s no other way to say it – straight up lying here when he says Coy “does not cite the Rogoff book or any source at all for the same fact.” See for yourself:

There is one additional case – the 12th — that involves, not a piece of data, but a quote from Richard Holbrooke that also appeared in a George Packer essay in the New Yorker. I got that from a direct conversation with Holbrooke in person several months before he died. He had made that particular comment to me many times. I asked him in this case if I could quote him. He agreed. I put it into my notebook, marked, “for attribution.”

There’s not much to add here other than the two possible conclusions that can be reached: 1. that Holbrooke regularly faced this observation, Zakaria would somehow overlook Clinton’s role in it, and then hold on to that quote for about a year, or 2. Zakaria plagiarized from the New Yorker. There’s public precedent for one of these.

 

 

Last, but not least, is a response to not just Zakaria but Mr. Fred Hiatt of the Washington Post, the paper’s editorial page editor who popped approximately eight monocles yesterday while scoffing that it was “reckless even to suggest this is plagiarism.” The use of the word “reckless” shows an incredible lack of self-awareness for someone who headed the Post’s editorial board when it was cheerleading the country into war in Iraq.

“Reckless” is also a funny choice of words for someone who just assumed that this was all we had to post. Our Bad Media will have more extensive examples of plagiarism by Fareed Zakaria later this week.