Sock puppet reviews is the charming name for artificial reviews posted on-line (e.g. on in a log-rolling effort to get the sales of your perhaps less than brilliant book up into the double or even triple figures. These tend to be paid for by authors in an effort to make their books seem more popular. Writers have always taken in each other’s laundry: if you review me (well), I’ll review you (well). Never of course an explicit contract, but certainly an implied one in many cases. When review media were thinner on the ground this game was only playable by a few — those who were reviewers for the print review media. Now with Amazon and the likes it becomes a game open to any and all. If you Google “paid book reviews” you will find several businesses willing to review your book for a fee. This New York Times article gives an account of one such operation. It’s amazing how much authors will pay, and how little the business has to pay people to actually write the reviews! Amazon doesn’t like these sock puppet reviews as you can see, and as this piece from Techdirt tells.

Here’s another account from a British perspective, from The Bookseller blog Futurebook:
“Amazon has retrospectively deleted a number of reviews from its website that it deems fall foul of its revised guidelines. This has irked writer/reviewers such as the ever vociferous Joe Konrath, who describes it as a major fail on Amazon’s part, but blames the company’s actions on those authors who put their names to the ‘No Sock Puppets Here Please’ letter. Amazon’s new guidelines prohibit authors and publishers from posting reviews against products where there is a conflict of interest, either because of a direct financial interest (i.e the publisher of the book), or where there is a ‘directly competing product’ (i.e. a crime author reviewing a fellow crime-writer’s book). This at least is the theory, it is not clear how Amazon is policing it. On Konrath’s blog there is plenty of argument over unintended consequences, but the truth is that the move shows how precarious a world dominated by Amazon really is, a detail that seems lost on some. Amazon can shift this landscape with a tweak of its computer code. And it doesn’t always get this right. As Lee Child notes in the comments section on Konrath’s blog: ‘So what has happened here is that a problem was perceived (much more likely via the NYT than anywhere else) and Amazon tried to solve it via robotic automation, and failed to hit the spot. They’ll try a few more times, and eventually they’ll get it a little closer to right.'”

Self published authors puffing their friends’ books on Amazon is one thing. But here’s a report of someone hacking into the website of an academic publisher and faking peer reviews. This strikes at the basis of the trust which underlies the publication of academic books and articles: we publishing staff can’t possibly be expected to understand some of the research papers and book manuscripts we pass gate-watcher judgement on. We have to trust our trusted advisors. Well maybe we now have to deal with them the old-fashioned way, by mail and by telephone.

In the TLS of 10 April 2013, Eric Naiman has written a long and fascinating piece which exposes a web of fictitious reviews, publications, academics, all brought to light by some checking into the planted story (repeated in several recent biographies of Dickens, though in some cases later withdrawn) that in 1862 Dostoevsky on a visit to London, had a meeting and a revealing conversation with Dickens. Naiman set out to discover the source of this canard, but along the way his detective work lead him to a number of scholars who share a readiness to quote one another as well as their putative creator, in a circularity that makes the head spin. Made-up reviews, written by made-up people, and sometimes even published in made-up publications is sock puppetry of a higher order. Naiman writes: “There is, it seems to me, a fundamental difference between posting partisan, anonymous reviews on Amazon, where there is no assumption of proper evaluative standards or impartiality, and placing similar reviews or hoaxing articles in academic journals, which are still the most hallowed sites for the development and transmission of humanistic ideas. The former is a cheap act of virtual graffiti; the latter may be the closest a secular scholar can come to desecration.”

Not sure I’m willing to dismiss the Amazon problem as virtual graffiti — not sure whether virtual graffiti is something I’d want to encourage anyway, even if I had a clear idea what it was. As long as you’re not having to pay a lot for a self published ebook, I guess the amount of harm done by being mislead into buying it by made-up reviews isn’t huge in the scheme of things. But it’s still wrong. Personally I don’t make buying decisions based on what people I’ve never heard of have said about a book, but I have looked at Amazon reviews (even written one or two). To the extent that they can tell you something factual, they are useful. If Mr Bing Liu is right (see the NY Times article) and fully 1/3 of on-line reviews are not genuine, we should consult on-line reviews with “caveat emptor” ringing in our ears.