L’Affaire Mukherjee: the last word

Barring unforseen circumstances, this will be the last post I put up about Siddhartha Mukherjee’s misleading article about epigenetics and The New Yorker (see my posts on it here and here).

First, on the website of the Scripps Translational Science Institute, Mukherjee has written a rebuttal, which is not really directed at the posts on this site but to a bunch of emails sent to the New Yorker that criticized his piece.  As far as I know, those critical emails were sent directly to the magazine rather than the author, and then the NYer passed them on to Mukherjee. He then wrote a big email response and sent it to the critics (I was not among them, though I eventually saw his response). That email  is what Scripps published, so you should consider his rebuttal as a response to the criticisms sent to the New Yorker and not to what I posted.

One example of this difference: in point 8 of his rebuttal, Mukherjee asserts that he did indeed criticize the “epigenetic-Lamarckian” process as being unlikely to cause adaptive evolution. Yes, that’s true—and I said in my original post that I agreed with him, but thought he could have made this point a bit more strongly in the New Yorker piece. Perhaps some other critics faulted him in the emails sent to the New Yorker. So, for the record, let me say this: all of us, including Mukherjee, agree on the gist of what follows (though I don’t know if he’d sign off on this wording):

There is absolutely no evidence for any Lamarckian form of evolution based on “epigenetic” markers on the DNA produced by the environment. Further speculations about this—and claims that it shows that the modern theory of evolution is wrong—are misguided and should be ignored pending some real evidence. 

Although I don’t think Mukherjee’s response on the Scripps site is very convincing, nor rebuts the faults of the New Yorker piece enumerated by the various scientists on my site  (see especially Ptashne and Greally’s criticisms, and the letters by Madespacher and Henikoff), I’m not going to deal with this further. I will let readers and other scientists judge the entire exchange. The only future posts I will put up about this matter will be notices and comments on any press coverage (see below).

But I do fault the New Yorker for failing to print any criticisms of Mukherjee’s piece, for that piece will then stand unsullied, forever, in its pages. Only those who have read my posts or the press coverage to come (see below) will know of the problem.  The New Yorker really needs to look seriously at how it vets its science pieces, and, as I noted before, its entire attitude towards science. But their arrogance suggests that they won’t do this.

Finally, there will be several press pieces about the controversy, as I (and several others, surely including Mukherjee) have talked to reporters. The first press piece, “Right but wrong” (subtitle: “The field of epigenetics is poorly understood by non-scientists. Did a recent New Yorker magazine article help matters?”), was published yesterday by Aleszu Bajak in Undark, an online organ of MIT’s Knight Science Journalism Program. There you can read Mukherjee’s “explanation” of why his New Yorker piece wasn’t a thorough (and I’d add “accurate”) summary of the field. (Hint: not enough space!)

One plaint about the Undark piece (and about the Vox piece that I just saw and will dissect tomorrow): you cannot excuse inaccuracies or misleading information on the grounds that you didn’t have enough space to tell the truth. From Bajak’s conclusions:

At the same time, it’s worth asking if such a thing [providing “an honest explanation of the nuances of gene regulation”] can be done at all, by anyone, in a popular magazine — the goal of which, in any case, is not to mimic a scientific journal, but to communicate to the general public the alluring frontiers of science — in this case, epigenetics (however that might be defined). It’s possible that Mukherjee fell short, but his critics would do well to consider the exceedingly difficult challenge he and his editors sought to undertake.

Sorry, but I’m not sympathatic to the problems of journalists who decide to tell a distorted story because they are either too lazy or don’t have enough space to tell the right story.

In fact, Mukherjee could have told the true story of gene regulation, rather than the cute but incorrect story, in the same amount of space that he had. But of course if he described the real state of the field, he wouldn’t have a novel story to tell.

So listen up, journalists: it’s better to tell the truth than mislead readers with fine words and a false story. Not all scientific explanations are equal, and the first job of a science journalist is to get the facts right. After that you can dress it up with fine words.


  1. Posted May 7, 2016 at 10:47 am | Permalink

    Vox.com has used a lot of your work in gathering comments to write a piece about the article as well.

    What’s funny (sad, not haha) is that the vox.com article gets the definition of epigenetics wrong.

  2. Richard Bond
    Posted May 7, 2016 at 11:34 am | Permalink

    As someone with a great interest in evolution, but no formal expertise (my degree is in physics), I would have been taken in by Mukherjee’s article, until, that is, the mention of Lamarck. To an even more casual reader, I expect that it would sound pretty convincing. All the more important, therefore, that editors of news media that propagate this sort of stuff check it before publication.

    • Torbjörn Larsson
      Posted May 7, 2016 at 1:28 pm | Permalink

      Yes. And I think the abject failure of Mukerjee’s pomo (?) epigenetics to be a vehicle for lamarckian evolution shows that it is powerless. If it can’t ride on modern, 4 billion years old evolutionary mechanisms, what effect could it ever have?

    • Marilee Lovit
      Posted May 7, 2016 at 3:52 pm | Permalink

      I lack the expertise to understand this very well, but was put off on my first reading of Mukherjee’s article by notions of “a layer that hovered, ghost-like,… or levitated… above the genome…” Also, twins are always interesting, but they did not seem to have a lot to do with the central subject.

      About the need to communicate difficult scientific concepts truthfully even in a popular magazine (rather than trying to make a good story more important than the facts)–I am reminded of Stephen Hawking in a Brief History of Time saying that even difficult new concepts and views of reality gradually become part of general human knowledge. So it is important for the explanations to be as truthful as possible, and not mislead.

  3. merilee
    Posted May 7, 2016 at 11:51 am | Permalink


  4. jaxkayaker
    Posted May 7, 2016 at 12:13 pm | Permalink

    I’ve read part of the rebuttal at the Scripps site, and Mukherjee apparently doesn’t realize that at least part of what he wrote doesn’t mean what he claims he intended to convey.

    “The theme is repeated in subsequent pages:

    “When one twin breaks an ankle and acquires a gash in the skin, wound-healing and bone-repairing genes are turned on, thereby recording a scar in one body but not the other.”
    And then again:

    Chance events—injuries, infections, infatuations; the haunting trill of that particular nocturne—impinge on one twin and not on the other. Genes are turned on and off in response to these events, and epigenetic marks are gradually layered above genes, etching the genome with its own scars, calluses, and freckles.
    That the role of gene regulatory factors gets three independent mentions, each one preceding any description of epigenetic mark, begs the question of what, exactly, the readers were objecting to. Indeed, the role of cascades of gene regulation appears a fourth time, below, in a description of Yamanaka’s experiments. So actually, the number, really, is four independent times that genes turned on and off are emphasized as primary responses to the environment. By the way, the sentence above even clarifies that gene regulatory factors are primary, and epigenetic changes may be secondary.”

    By talking about genes being turned on and off without explicitly mentioning gene regulatory factors and what they are, and instead moving on to epigenetic factors leaves the reader with the impression that the epigenetic factors ARE the gene regulatory factors. It’s poor writing, even if we leave aside the misuse of “begs the question”.

    • Diane G.
      Posted May 7, 2016 at 8:53 pm | Permalink

      “It’s poor writing, even if we leave aside the misuse of “begs the question”.”

      Which must have passed through a New Yorker copy editor on its way to print. (Or electrons.) Which means either said editor was asleep at the switch, or that the mag’s style book has been changed to reflect popular usage. I suspect it’s the latter, which I find very annoying, even though I understand and basically buy the descriptivist approach to linguistics.

      (Pretty off-topic, but I needed to subscribe anyway, so why not?)

  5. Curt Nelson
    Posted May 7, 2016 at 1:08 pm | Permalink

    sub to the max

  6. jaxkayaker
    Posted May 7, 2016 at 1:36 pm | Permalink

    Jerry, at the bottom of Mukherjee’s blog (sorry) post, he links to your two earlier posts here on your website. That, and what he wrote in the post does make it appear he’s attempting to rebut what you published.

    Vox should have just paid you to let them republish your posts instead of whitewashing the situation at the end. Typically of modern journalism, they act like both sides are on equal footing wrt the evidence, and just having a difference of opinion.

    • Posted May 7, 2016 at 3:43 pm | Permalink

      There was of course substantial overlap between what was sent to the New Yorker and what appeared on my site, as some of the same people were involved! But Mukherjee’s post is not a rebuttal of what I put on my website, which was a distillation of Ptashne and Greally’s thoughts. If you look at what Mukherjee’s trying to rebut, a LOT of it is NOT stuff that appeared on my site. He’s rebutting the emails and letters to the New Yorker.

      Yes, the links are there, but that’s just for reference, since my site was the only place where some of the criticism appeared.

      • Posted May 7, 2016 at 8:08 pm | Permalink


        It is another misleading act of public (mis)communication that Mukherjee’s “rebuttal” has been left by him (and Scripps) to imply that he was responding to what Jerry posted.

  7. Chemist
    Posted May 7, 2016 at 3:08 pm | Permalink

    This whole incident sadly reflects the state of science journalism – sensational stories over reality & facts. If you can write a compelling article or convey an exciting message in short form it will draw eyeballs / an audience. Whether the article reflects accuracy is almost irrelevant to the publisher or general public (see Trump or Chopra to get what I mean.)

    It just seems the public wants easy answers or great stories over details that are complex or facts difficult to comprehend. Certainly seems that when critical thinking is required, the easy road is taken rather than putting in the hard work of thinking things through to the truth.

    The fact that The New Yorker isn’t stepping up to correct the errors in Mukerjee’s article is indicative that there isn’t a high price to pay for propagating mistakes. Comparing the attention the article has received versus what any correction has gained speaks volumes.

  8. Posted May 7, 2016 at 3:43 pm | Permalink

    Bam! Well said sir!

  9. Joe Dickinson
    Posted May 7, 2016 at 3:53 pm | Permalink

    I think Mukherjee’s response dodges the main criticism. He does indeed write of genes being regulated, but I can’t find any specific mention of protein (or RNA) transcription factors. So, I think the lay reader would draw the conclusion that the story is about some clever scientists who figured out how gene regulation actually works while everyone else remains mystified.

    He justifies the omission by saying it would take thousands of additional words to provide that background. Not so. Here it is in about 130 words (by a less gifted writer):

    “Beginning in the late 19th century, embryologists devised methods to remove, isolate and transplant individual embryonic cells or small groups of cells (sometimes even just nuclei) and were able thereby to demonstrate that most cells retain a complete set of information (genes) throughout development. So, it must be that there are mechanisms to regulate the expression of that information, just as you can have many programs on your computer’s hard drive but select just one to run at any given time. Beginning in the 1960s, a large number of proteins have been identified that bind DNA at specific target sequences and turn nearby genes on or off. This almost certainly is the main basis for cell specialization. However, my heroes were interested in other mechanisms that might also play a role (segue to main article).

    • Posted May 7, 2016 at 4:38 pm | Permalink

      Yes, you’ve hit on the main problem with his rebuttal, and corrected it. Unfortunately, if he had told your version, it wouldn’t have been novel!

      • Posted May 10, 2016 at 1:00 pm | Permalink

        There are surely novel things in this field that he could describe for lay readers without misleading them.

    • jaxkayaker
      Posted May 7, 2016 at 7:17 pm | Permalink

      Your first paragraph, Joe, makes the same point I did in my last paragraph at 12:13 above (comment 4).

    • Posted May 7, 2016 at 8:10 pm | Permalink


  10. Gordon
    Posted May 7, 2016 at 5:12 pm | Permalink

    As was said a couple of days ago a discussion such as this makes you site worth every cent for we non-experts who may not comment on such stories but do read it.I am much better informed and perhaps a little bit wiser.

    • Diane G
      Posted May 7, 2016 at 9:10 pm | Permalink

      Cents? Does Jerry have a tip jar? 😉

      (Annoyingly, Mukherjee’s probably the only one who’s made any cents in this whole debacle. Tho not sense.)

      • Filippo
        Posted May 8, 2016 at 2:05 am | Permalink

        One trusts that it will not take on too much of the skunk’s scents.

        • Diane G.
          Posted May 8, 2016 at 4:06 am | Permalink


          Too late!

      • Merilee
        Posted May 8, 2016 at 9:49 am | Permalink

        Does Jerry maybe also have a swear jar for the site?

        • Diane G.
          Posted May 12, 2016 at 7:57 pm | Permalink

          He’d be smarter to have one at Pharyngula.

  11. John Harshman
    Posted May 7, 2016 at 9:22 pm | Permalink

    Biggest disaster in science journalism and biggest setback in public understanding of biology since ENCODE.

  12. Joe Dickinson
    Posted May 7, 2016 at 11:33 pm | Permalink

    As with others who commented, I enjoyed and admired his cancer book. But this episode gives pause that I have experienced before. I am reading a book that I enjoy and from which I feel I am learning a lot. Then I hit a topic within which I feel a certain degree of competence and find major errors. So now, I wonder, how far should I trust the stuff in areas that I do not know so well?

    • Diane G.
      Posted May 8, 2016 at 4:02 am | Permalink

      Ay, there’s the rub.

      • Marilee Lovit
        Posted May 8, 2016 at 6:20 am | Permalink

        True of the New Yorker too. I have much enjoyed Elizabeth Kolbert’s writing on climate in that magazine. But should I have more doubts than I do?

        Sometimes Harpers has science articles. I wonder if they are better reviewed prior to publication.

        • Filippo
          Posted May 8, 2016 at 6:24 am | Permalink

          I should hope and expect so, at least under former editor Lewis Lapham’s watch.

          • Merilee
            Posted May 8, 2016 at 9:20 am | Permalink

            Lapham was great, but I also expected much better from Remnick at the NYer. When Tina Brown was editor for a ahort while, not so much…

        • Merilee
          Posted May 8, 2016 at 9:17 am | Permalink

          Same with The Atlantic. Caveat emptor, I guess, but we emptors are clearly not always well-enough informed about every topic we approach.

          • Posted May 10, 2016 at 1:12 pm | Permalink

            If we were, we would not need such articles in the first place.

One Trackback/Pingback

  1. […] writes again, L’Affaire Mukherjee: the last word , excerpt from the […]

%d bloggers like this: