Russell's Paradox of ghostwriters | Hacker News Reader

Russell's Paradox of ghostwriters

statmodeling.stat.columbia.edu

41 pointsluu28 comments

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, so much so that when 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.”

And then what? That's a subordinate clause, not a sentence. There's supposed to be a verb that tells us what happened.

Maybe this guy needs a better ghostwriter.

The new "Tom Clancy Command and Control" is by someone who almost gets Clancy's style right. But his work is still inferior to the actual books written by Tom Clancy, who's been dead for a decade.

Writing well is really difficult, so it's not at all surprising that someone who is not a professional writer gets some help. But there's a huge gap between "the concepts were all mine but I got hep with the wording" and "I signed my name twice, once on the book and once on a check".

> In contrast, Malcolm Gladwell deserves credit for producing readable prose while having his own interesting style. I doubt he uses a ghostwriter.

He could use a ghostresearcher. Most of his claims are rather dubious.

There's no shame in it: academics have grad students, who at least sometimes get a mention in the acknowledgements.

> There's no shame in it: academics have grad students, who at least sometimes get a mention in the acknowledgements.

In what field????? In fields I'm familiar, whoever does the research (i.e. the grad student or the postdoc) is the first author, helpers are co-authors and the PI i.e. the main academic who leads the lab is the last author. In what field PI is the first author and grad students "sometimes get acknowledgement"? That sounds like academic fraud.

It used to happen much more in the past. Many decades ago, AFAIK, at least in the journals / fields / areas I'm familiar with.

I'm not sure where that might happen or be tolerated, today. If you go back and look at research into antibiotics, DNA, etc. back in the 1940s and 1950s ... and even research in the 1960s and 1970s, decades later, there ended up being serious questions around ethics, authorship, etc. for some really seminal papers, for example.

I agree. Grad students must appear in the paper because most universities have an explicit or implicit requirement of a few papers to get the Ph.D.

It's very field dependent. The first author=person who did most of the work and last author=advisor is pretty much just how biomedical science works. Other fields have different traditions. I'm a computational biologist who has worked in both biology and computer science, and in CS (at least in the 1990s-early 2000s) the tradition was simple alphabetical order, for example, with no privilege afforded to either the first or last place. Along with other quirks like how presenting a result at a conference was considered equal to publishing it as opposed to in biology where conferences are more social networking events.

In my experience for computer science (systems field) it’s usually the same, first author=most work and last author=advisor.

In theoretical CS, the convention is still alphabetic order. In most other CS, advisor tends to be last.

I seem to recall that some of the Stratemeyer[1] series had a bit more of the ghostwriter's personality leak through in the earlier books. I have also been told (but not confirmed myself) that they filed off most of those edges when they revised the books to remove some of the casual racism that hadn't aged well.

1: https://en.wikipedia.org/wiki/Stratemeyer_Syndicate

> “Nearly all experts and celebrities use ghostwriters,”

Funny enough, my mind first went to Bernie Sanders' My Revolution.

Either he didn't use a ghostwriter, or the ghostwriter was Larry David doing a Bernie Sanders impression.

The Wikipedia entry for the book doesn't mention a ghostwriter[1]. But after reading this article, I think Wikipedia should have a policy of explicitly stating so-and-so did not use a ghostwriter for what are apparently edge cases.

Hey Wikipedians-- can you make this so?

1: https://en.wikipedia.org/wiki/Our_Revolution

> I think Wikipedia should have a policy of explicitly stating so-and-so did not use a ghostwriter

How could anyone ever verify that was the case? Most people who use a ghostwriter won't admit to it, so anyone who claims that they didn't can't be trusted.

> Either he didn't use a ghostwriter, or the ghostwriter was Larry David doing a Bernie Sanders impression.

I'm not a big memoir guy, but your description is enticing.

Where is the paradox?

In this town there are two kind of people. Those who write their biographies themselves and those who use the help of the town Ghostwriter. Who writes the Ghostwriter biography?

Thanks for phrasing it this way!

The Ghostwriter, but he is of no help.

10,000 monkeys with typewriters!

Now we have 70B LLM models.

Russell himself said that construction isn't a version of his paradox, and it depends on there being only one ghostwriter in addition to depending on claims nobody made.

So I don't know what the author actually intended. The train of thought in the article never gets particularly close to anything that resembles a paradox.

Those two categories are vague and non-exhaustive (and non-exclusive). Also, if the Ghostwriter writes his biography himself (entirely), in most senses of the word, that doesn't mean he "helped" himself to do it. So, no paradox at all there, and certainly not a Russellian one.

This piece does seem incomplete, starting off with a big grammar issue

> [...], so much so that when 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.”

"so much so that when ..." what?!

meta commentary?

> "so much so that when ..."

This is a common enough turn of phrase, at least in my corner of the world. What's your problem with it?

https://www.collinsdictionary.com/sentences/english/so-much-...

The conclusion is missing. Normally the formula is:

"So much so that when {occurrence} {outcome}"

This has the occurrence (the inclusion of the topic) but not the consequent outcome - seems like some anecdote about that author not recollecting the clearly memorable topic got mangled.

There's a missing word later in the piece as well.

Yes exactly, thank you!

> There's a missing word later in the piece as well

You mean where the author, Andrew Gelman, "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," and just completely skips over the verb?

I think that part was done by the ghostwriter.

Did anyone else read this as "Russell's Paradox of Ghostbusters" the first time?

Reply on news.ycombinator.com