A few months ago we discussed the repulsive story of a UCLA USC professor who took full credit for a series of books that were ghostwritten. It turned out that one of the books had “at least 95 separate passages” of plagiarism, including “long sections of a chapter on the cardiac health of giraffes.”
You’d think you’d remember a chapter on the cardiac health of giraffes. Indeed, if I hired someone to write a chapter under my name on the cardiac health of giraffes, I think I’d read it, just out of curiosity! But I guess this guy has no actual curiosity. He just wants another bestselling book so he can go on TV some more and mingle with rich and famous people.
OK, I’ve ranted enough about this guy. What I wanted to share today is a fascinating story from a magazine article about the affair, where the author, Joel Stein, “Nearly all experts and celebrities use ghostwriters,” and then links to an amusing magazine article from 2009 subtitled, “If Sarah Palin can write a memoir in four months, can I write my life story in an afternoon?”:
When I heard that Sarah Palin wrote her upcoming 400-page autobiography, Going Rogue: An American Life, in four months, I thought, What took her so long? To prove that introspection doesn’t need to be time-consuming, I decided to try to write my memoir in one day. Since Palin had a ghostwriter, I figured it was only fair that I have help too, so I called Neil Strauss, who co-wrote the best-selling memoirs of Marilyn Manson, Mötley Crüe, Dave Navarro and Jenna Jameson. . . .
The whole article is fun. They wrote a whole memoir in an afternoon!
That particular memoir-book was a gag, but it got me thinking of this general idea of recursive writing. A writer hiring a ghostwriter . . . what a great idea! Of course this happens all the time when the writer is a brand name, as with James Patterson. But then what if Patterson’s ghostwriter is busy and hires a ghostwriter of his own . . .
Perhaps the most famous ghostwritten book is The Autobiography of Malcolm X, by Alex Haley. After Roots came out, the Malcom X autobiography was promoted heavily based on the Haley authorship. On the other hand, parts of Roots were plagiarized, which is kind of like a ghostwriter hiring a ghostwriter.
A writer hiring a writer to do his writing . . . that sounds so funny! But should it? I’m a professional writer and I call upon collaborators all the time. Collaborative writing is very rare in literary writing; it sometimes happens in nonliterary writing (for example here, or for a less successful example, here), but usually there it follows a model of asymmetric collaboration, as with Freakonomics where Levitt supplied the material, Dubner supplied the writing, but I assume that both the content and the writing benefited from conversations between the authors.
One of the common effects of ghostwriting is to give a book a homogenized style. Writers of their own books will have their original styles—most of us cannot approach the caliber of Mark Twain, Virginia Woolf, or Jim Thompson, but style is part of how you express yourself—and nonprofessional writers can have charming idiosyncratic styles of their own. The homogenized, airport-biography style comes from writers who are talented enough to produce this sort of thing on demand, while having some financial motivation not to express originality. In contrast, Malcolm Gladwell deserves credit for producing readable prose while having his own interesting style. I doubt he uses a ghostwriter.
Every once in awhile, though, there will be a ghostwriter who adds compelling writing of his own. One example is the aforementioned Alex Haley; another is the great Leonard Shecter. I’d say Stephen Dubner too, but I see him as more of a collaborator than a hired gun. Also Ralph Leighton: much of the charm in the Feynman memoirs is that voice, and you gotta give the ghostwriter some of the credit here, even if only to keep that voice as is and not replace it with generic prose.
There must be some other ghostwriters who added style rather than blandness, although I can’t think of any examples right now.
More generally, I remain interested in the idea that collaboration is so standard in academic writing (even when we are writing fiction) and for Hollywood/TV scripts (as discussed in comments) and so unusual elsewhere, with the exception of ghostwriting.
If ghostwriters keep hiring ghostwriters, would the book converge to the most plain-vanilla book ever? Or would it diverge wildly, with each additional ghostwriter adding diverging elements? A boring research question that nobody should pursue because who cares: can we use LLMs to see which characteristics make the message diverge and when does it converge?
Or just use the LLM to ghost write to begin with. I see no reason you can’t ask an LLM to write a story in the style of X or in the style of a ghost writer writing on behalf of X, or….. In fact, many people have done just that, at least for short pieces of writing. Luckily it is still possible to recognize many of these attempts – the more glaring ones are actually somewhat entertaining. I wonder how many professional ghost writers are already relying on Chat GPT.
“make the message diverge and when does it converge?”
Does this count as a ghost in the shell speaker?
“This dev conference organizer seems addicted to making up women / Eduards Sizovs, the organizer behind the Devternity and JDKon developer conferences, admitted that at least one woman speaker’s profile was ‘auto-generated.’
https://www.theverge.com/2023/11/28/23978254/devternity-jdkon-developer-conference-fake-women-speakers
Yes.
Transcript coming soon. See sport and finance ‘news’.
And colorwashing diversity via plagiarized training set.
Ghostwriters are common enough, but how abut ghost colleagues? Cyril Burt not only invented the ANOVA data (regarding twins separated at birth) but also the women, Miss Margaret Howard and Miss Jane Conway, who supposedly did the tedious, pre-computer calculations. See the following to get a fuller picture
https://en.wikipedia.org/wiki/Cyril_Burt
He was a USC professor not UCLA! USC is much much better at scandals and fundraising than UCLA… It’s the only things they do well. (I’m a USC alum)
Daniel Lakeland: I found this regarding a photo-shopping neuroscience scandal at USC
https://www.msn.com/en-us/health/medical/usc-has-been-rocked-by-allegations-of-research-fraud-against-a-neuroscientist-heres-what-happened/ar-AA1kQrPN
Photoshopping software was invented around 1987 and is also responsible for scandals in neuroscience at Stanford (president resigns but still a member on the faculty), Ohio State, and Johns Hopkins (Nobel Prize winner implicated).
In my view, the Harvard flap regarding Francesca Gino’s behavioral science publications, seems trivial in comparison.
This scandal has been mentioned before, in a comment on the second David Agus plagiarism blog post. Posted by Daniel Lakeland:
https://statmodeling.stat.columbia.edu/2023/11/22/in-all-seriousness-whats-usc-gonna-do-about-its-plagiarizing-professor/#comment-2288099
My east coast bias that I get USC and UCLA confused.
Here’s one to add to the USC list: “University of Southern California social work students just sued the school…” for misleading claims. https://www.businessinsider.com/usc-lawsuit-misrepresenting-social-work-program-massive-student-loan-debt-2023-5 — whether it’s fraud, sleaziness, or simply taking advantage of terrible choices made by Master’s students is up for debate.
Raghu:
More background from this L.A. Times article:
Don’t worry, not all of USC’s permanent faculty are world-class!
The LA Times article continues:
And this hilarious bit:
Also I like that Kareem put his collaborator on the author list for his Sherlock Holmes fiction. They got better as they went along. I suspect Kareem had some good ideas and wrote a lot of the first book then his author collaborator polished it. In the second and third book I’m guessing she was involved from the start and so they were more polished. They’re good books, so I’m glad they got together and did them.
The thing that’s appalling about ghostwriting is that the writer’s name is absent; it’s basically plagiarism on the part of the named author. In contrast to the ghostwritten books by famous or infamous authors, there are books in which the well known person and the other person — presumably the writer doing most of the actual writing — share authorship. As an example, George Church’s and Ed Regis’ “Regenesis.” (A mediocre book; not recommended.) I’ve wondered why the authors who hire ghostwriters don’t take this approach. Do they, or the publishers, really think that adding another name to the author list will dissuade people from buying the book? Is the named author really that vain?
Yes (to the vanity). Brand dilution is real: https://pubsonline.informs.org/doi/full/10.1287/mnsc.2022.00852. Actually, all of the literature on brand dilution doesn’t exactly map into the ghostwriter phenomenon, but that opens the possibility for a publication to explore it as a new case.
I thought brand dilution is a company that does X well branching out to do Y, perhaps badly. If so, writing the book is in itself the brand dilution. Naming the ghost writer would perhaps minimize the dilution, since no one is ascribing the writing solely to the famous author.
I did say that brand dilution doesn’t map exactly to ghostwriting. But it does have some common features. By definition (at least my definition), ghostwriting is a way to increase the quantity of the brand (why use ghostwriters if not to put out more stuff?). The quality may also be impacted – if it is reduced, it is the brand that will suffer. So, it is a form of dilution, even though it doesn’t quite fit the cases that are usually discussed. Still, I do attribute it largely to vanity.
I don’t see anything appalling about ghostwriting.
I don’t don’t understand why there is an obligation to name the actual writer of a book in particular – but nothing else. Henry Ford’s name is still on all of the cars his company makes but he’s been dead for 80 years and I glance over at my computer and see the names of two dead people on it and none of the names of the people who actually made it.
So why does anyone give a hoot who wrote what article or book?
Here’s why: so they can attack the person who wrote it rather than the content of the article, book or whatever. It’s the same reason the left demands “transparency” in campaign finance: they want to attack the source rather than the content. These days it even means physically attacking the source, as we’ve seen so vividly recently. Interestingly, when the attacked source strikes back through non-violent means, like black-listing the attackers, that’s somehow unfair, and the attackers claim to be entitled to identity protection!! So I guess we’re supposed to name the authors when it benefits the people who want them named, but not when it doesn’t.
That’s the left. It fits perfectly in academia, which poses as a meritocracy of ideas but where, in reality, “peer” – AKA “gang” – assessement is the key to success.
Bro you need to get out more
chipmunk
I’m shocked that you don’t believe in accountability. If you don’t know who wrote something, you don’t know who is responsible. But I see it did give you an opportunity to rail against the “left” even though that has nothing to do with the issue.
Neil Strauss… where have we heard that name before?
https://observationalepidemiology.blogspot.com/2017/12/merry-muskmas-to-all.html
Hi Andrew,
As painful as it is to say, but fraud is associated with USC not UCLA. It’s a typo.
Just an interesting FYI that Alexander Dumas invented a type an industrialized ghost writing for his serialized novels. He employed a studio of writers who helped him materialize his ideas. I’ve heard different versions of how this worked, but by some accounts, his stable of writers had different specializations: some wrote up nothing but scenery, others descriptions, etc— and he could produce material quickly from the raw materials his studio kept at hand. I think there is debate, though, about the extent to which he relied on ghostwriters.
Even after learning about his approach, I still love reading Dumas. And, really, it makes sense as a business mode when someone has a reach/platform but doesn’t want to actually do the hard parts. Maybe if they just listed themselves as “producers” it would be a bit more transparent and acceptable?
Dumas once joked to his head collaborator about his latest hit, “How is it? I haven’t had a chance to read it yet.”
(and yes, I too still enjoy Dumas. The fact that it was a group effort doesn’t bother me at all.)
Somewhat related: “Naked Came the Stranger” by Penelope Ashe: https://en.wikipedia.org/wiki/Naked_Came_the_Stranger
Interesting! Kinda sounds like Hollywood movie productions.
On one hand word processors and web apps make it easier to write collectively, on the other hand many novelists and academics from the middle of the last century were husband-and-wife teams with only one (usually the husband) listed as author.
Sean,
Good point. Also, back in the day, editors and publishers had much more input, right? When I write a book, I deliver it in finished form to the publisher, who then publishes it, with essentially zero intervention, just some copy editing where they want me to futz around with commas, etc. But back when authors sent in handwritten or typed manuscripts that needed to be set into print anyway, I can well imagine that editors were making all sorts of changes without getting credit.
Has anyone other than Charles Barkley ever claimed to have been misquoted in his autobiography?
Jonathan:
It’s a good thing that I didn’t put any fake Rubin quotes in the second or third editions of Bayesian Data Analysis!
I’m surprised no one mentioned Courant and Robbins “What is Mathematics” which Courant refused to put Robbins’s name on until Robbins threatened to “make a fuss” [Robbins’s interview with Warren Page]. Finally, when the paperback appeared Robbins name was added to the title page, but Courant continued to hold the copyright and received the royalties.
There’s an impressive set of physics textbooks by Lev Landau and Evgeny Lifshitz. It’s sometimes said the books contain “not a word of Landau and not a thought of Lifshitz.”
(That’s way too unkind: Lifshitz was a really good physicist and made some important discoveries, I’m sure some of the material is based on his thoughts, maybe most of it. But Landau was one of the greatest physicists of all time, so poor Lifshitz is forever in his shadow).
Someone once told me that of the two volumes on String Theory by Green, Schwartz and Witten:
“The cover is Green, the letters are Schwartz (black) and it’s written by Witten.”
I have a very intriguing idea/storyline for either a screenplay/novel or both. I was under heavy sedation for open heart surgery and went to the most vivid and life like place only to witness a lot of my side interest in current military affairs, technology, medical science and AI to work its way into the story. It was like a dream state movie that I now need to follow up on. How do I get an honest assessment to validate going further, to the point where those involved would defer some payments for services against a percentage of hopefully large back end gains. I am not particularly wealthy but am willing to spend a little to test the waters.
Hopefully Not A Potential Sucker