Curiouser and curiouser

I have just finished reading for review one of Ian Stewart's popular maths books from 2008 (I missed it first time around), and it certainly lives up to its name Professor Stewart's Cabinet of Mathematical Curiosities as it contains the most curious thing I've ever seen in a review copy of a book in all the years I've been reviewing.

On four of the pages are hand-written corrections.

When I came across the first one, a simple change from McMahon to MacMahon, I assumed the change was actually printed in the book, to be quirky, but no, these are pencil corrections. To understand why it's so strange, you need to be familiar with the production process for a book like this.

Early editing, including copy editing, is done on bog standard Word documents (whether on a computer or printed out first). Then the book is typeset. Of course this no longer involves setting metal blocks of type in frames, but the text is imported into software that lays it out exactly how it will be on the final printed page. The result is then sent out for proof reading, both to the author and to one or more professional readers. Again this may be as a file (by now a PDF) or printed out on a laser printer.

Also around this time, there may be bound proofs produced. These look like rather scruffy paperbacks, bound up from the uncorrected proofs and are intended for early reviewers. They are easily distinguished from the final book - the cover makes it very clear what they are. Authors often get one or two of these, but no one, in my experience, uses them for proof correction, as they come out rather later than the unbound proofs.

Then, after considerable delay, the actual, final book is produced. Authors get copies of these, but in my experience they just look at the overall thing and flip through it. They certainly don't read it page by page looking for errors. Apart from anything else, they are usually fed up with the thing by now - the last thing they are going to do is read the final printed book.

And yet what I have here is a final printed hardback - not even a paperback, so certainly not a bound proof, not even one with the wrong cover on - in which someone has added pencilled corrections. It's an enigma, or, as Prof. Stewart might put it, a curiosity, that leaves me with a number of unanswered questions.
  • Whodunnit? Are these corrections in the very hand of the mighty Prof Stewart himself? Or someone at the publishing house? They clearly aren't from a disgruntled reader who sent the book back in, as the instructions for correction imply an insider.
  • What was intended with that first error? There's a big difference between period 2 and period 8.
  • Do all hardback copies carry the same mistakes? Or was this a short run that was then corrected? Do I have a strange one off copy?
  • Was it fixed in the paperback? If you have a copy of Professor Stewart's Cabinet of Mathematical Curiosities, do take a look at page 226 and compare it with my version below. Let me know what you see!

POSTSCRIPT: Ian Stewart has kindly got in touch to clarify what probably happened. This is not his handwriting, so was done by the publisher. It seems likely this was a copy the publisher kept to accumulate details of errors they were notified of to put in a future reprint and was accidentally sent out. 

I still think there are a couple of mysteries. First, why would you note errors in a copy of the book? They would be very hard to find. It would be much easier simply to note them as changes (or on an electronic copy). Secondly, this appears to be the 14th printing, but Ian implied that by the 12th this should have been fixed.

If you'd like to buy a copy yourself and make the comparison, you can get the book at and (These are the paperback versions, by the way, with different covers.)