Why do we compulsively add that line of digits to the copyright page of all our books? So we can tell at a glance which printing an individual copy comes from, of course.

But really, what use is that information? Have you ever been in a situation where it became necessary to know the printing number? I always thought we did it so that, if there was a problem in the manufacture of a book, we’d be able to say which printer did the book, and when. However, though I’ve lived through lots of manufacturing problems, I have never needed to know the printing number to solve the problem, and I find it hard to think of a situation in which that line of digits would be important evidence in any investigation. If there’s something wrong with a printing of a book we will probably have noticed it when we got our advance copy. Whenever I’ve had to do with printing a book, I have always felt it my obligation to be the first to detect any flaws. You are always in a better position apologizing to the rest of the company that such and so is wrong with this book, rather than having people point it out to you. (Fessing up to a problem is always the best policy. If you admit you did something wrong, people will tend to try to comfort you; whereas if you sweep it under the carpet, you’ll be in the firing line when it is finally detected.) If there’s an error in a book requiring stock to be sorted in the warehouse to segregate good copies from bad, the people doing the work will look at the pages where the error occurs, not at the copyright page to see if the printing number is 6 or 7. It’s just as easy to open the book to page 94 as it is to open it to page iv, and doing so leaves no room for ambiguity.

Deleting the lowest digit in the printing history key line every time you print the book does cost something.  Not a lot it’s true, but something. A large publisher can easily spend $50,000 a year updating the printing history line in their books. I’d like to hear from any publisher who saved $50,000 in any year by having information gleaned from that line — indeed $50,000 over a 10 year period! And please don’t boast to me about sticking it to a printer for the error in the last six printings. If you weren’t aware of the problem, shame on you — treat your suppliers not as adversaries, but as partners. They have to make some money in order to stay in business to serve you: the bargaining relationship has become asymmetrical.

The silliest (well maybe the second silliest) argument I was ever given for needing that line of digits was when an editor argued till blue in the face, that if the author didn’t see on the copyright page that this was the 4th printing, he’d have no way of knowing that the corrections we made after the 3rd printing had in fact been done. My insistence that the author might actually discover this by looking at the page on which correction had been needed, was pooh-poohed. The discussion came up in the context of a print-on-demand book. My colleague was shocked to discover that print-on-demand books didn’t have the printing history key line updated! As if anyone could afford to update files for a book which may be ordered two times today, six times next week, and fifty times in year, each printing being a separate “printing”. We’d need to be updating our files several times a day — “OK guys. That was the 767th printing, please update to 768 immediately because we may get another order tomorrow morning.” I suggested the rather obvious expedient of printing on the copyright page “Reprinted with corrections July 2010” or some such notice. You are making corrections anyway, so making one on the copyright page as well comes as close to costless as can be.

The other classic argument for retaining the line of digits was “Book collectors like it”! So some editors are happy to spend a significant amount of money every year so that in a few years time when the book gets into the second-hand market, some dealer can easily figure out whether he’s looking at a first printing or not. Most older books do not carry information about printing number. I’ve never been a book collector myself, though I’ve “collected” a huge mass of books. I’ve always wanted a book because I wanted to read it, not because it was an investment. As far as I can see the practice of indicating a printing number came about between WWI and WWII. I have a Cambridge University Press book which tells me a complete story (as Cambridge books do): “First edition June 1930, Reprinted July 1930, Reprinted with alterations December 1930, ‘Miscellany’ edition 1937, Second edition 1950”. Of course this only tells me what Cambridge were doing in 1950 (which date also appears on the title page), not 1930. My Oxford University Press edition of the Poems of Tennyson carries 1917 on the title page and nothing else about printing beyond a note on the last page that it was printed at the University Press, Oxford. Doubleday & McClure’s edition of the Works of Rudyard Kipling has 1899 on the title page and no other mention of printing or printer. It doesn’t even say it was printed in USA, though I guess it was. Alfred Knopf, who have always been rather keen on acknowledging printers, only tell me in the 1932 edition of Sigrid Undset’s The Burning Bush that this is the First Edition. I think we can assume that this means it is the first printing, and it was printed at Vail-Ballou and bound at H. Wolff. I wonder what the second printing would have said. Dorothy Parker’s Death and Taxes (Viking MCMXXXI) tells me that I have in my hands the “Second printing before Publication” — a statement almost more promotional than bibliographical.

I don’t have enough books to draw good conclusions, and have been unable to find any references to when publishers started identifying printing numbers, so I’ll stick with my WWI-WWII timeframe till better evidence comes to light. One bibliophilic site did allow that the practice was “recent”, but to a bibliophile recent could mean anything later than 1800. The Chicago Manual of Style (unsurprisingly prescriptive) tells you to use the year/date line of digits 16  15  14  13  12            6  7  8  9 (year on left, impression number on right). Some cunning publishers stick a couple of initials in the middle indicating the printer. Most don’t bother with the year. The printing number line can go from higher to lower,  9  8  7  6  5  4  3  2  1 or from lower to higher 1  2  3  4  5  6  7  8  9. Different people do one or the other.  Every time you print you delete one digit, so the lowest remaining digit tells the printing number.  Some overly elaborate people do a sort of converging line — 1  3  5  7  9  8  6  4  2 so that whatever printing it is they will retain a neatly centered line of type.

I have a suspicion (based on zero research) that British publishers may be more inclined to use the narrative version, while American publishers stick to the bare line of digits. I remember having to go through a spell where we had to blind stamp an initial on the bottom right hand corner of a hardback’s back cover to identify the bindery! But I really think it’s all been a vast waste of time. I bet that adding the printing number came into fashion in step with that great facilitator of reprints, offset lithography. When printings were done by letterpress, and reprinting mostly involved retypesetting, the change from one printing to another would be more obvious. When we started doing lots of reprints, the idea of tracking them would naturally come to mind. We seem now to have persuaded ourselves that we cannot live without these numbers, though nowadays reprinting is a several-times a year phenomenon. We have one which is in it’s 104th printing — what meaningful information is being conveyed by the “difference” between the 103rd and 104th printings?

Although I can’t think of any reason to keep doing it, I do feel strangely reluctant to recommend that we stop. I suppose I am hoping/fearing that someone out there will come up with the “real” reason that makes it all worth while.  I’m not holding my breath though.