How to Identify a First Edition

There is a gap between a book that says first edition and a book that is one, and the auction data measures it precisely.

Across the 592,936 priced book, map and print lots in this database, the typical sale is €60. Take the 10,212 lots whose description claims a first edition in any language — first edition, eerste druk, Erstausgabe, prima edizione, édition originale — and the typical sale is €66. That is a premium of ten per cent for the most desirable phrase in bookselling. Compare the same author's first-edition-labelled lots against their other lots and the premium disappears entirely: 1.00×, no difference at all.

The label is nearly worthless. A verified first printing is not. The rest of this guide is about closing that gap yourself, because nobody else is going to do it for you.

Edition is not printing, and the difference is the money

A publisher sets the type once. Everything printed from that setting is the first edition. But they may go back to press five times in a year, and each of those runs is a separate printing — also called an impression.

Collectors want the first printing of the first edition. That is the batch that existed before anyone knew whether the book would matter.

The auction listings show people getting this exactly right, and exactly wrong. One lot in the database is described as "1st Ed / 4th Pr" — an honest seller telling you it is a first edition and a fourth printing. Another calls itself a "RARE First edition 3rd print". Both are truthful; neither is what a collector means by a first.

You can see it in the prices. Lots that claim a plain "first edition" — 8,968 of them — sell for a typical €65. The 30 lots that specify both, "first edition, first impression" or "1st/1st", sell for a typical €135.50, more than double the market. Thirty lots is a thin sample and I would not build a valuation on it, but the direction is exactly what you would expect: the sellers who can name the printing are the ones who checked.

Number lines

From roughly the 1960s onwards, most English-language publishers print a row of digits on the copyright page. It looks like nonsense and it is the fastest check you have.

10 9 8 7 6 5 4 3 2 1

The rule is one line long: find the lowest number in the row. That is the printing. A line containing a 1 is a first printing. When the publisher goes back to press they strike the 1 off the plate, so the second printing reads 10 9 8 7 6 5 4 3 2 — lowest digit 2.

Common variants, all of which mean the same thing:

• Centred lines. 1 3 5 7 9 10 8 6 4 2 — printed from the middle outwards so the type wears evenly. Still a first: the 1 is present.
• Year and printing together. 89 90 91 92 10 9 8 7 6 5 4 3 2 1 — the left group is the year of printing, the right group the impression. First printing, and printed in 1989.
• Letters. a b c d e f g h i j or A B C D E. Lowest letter present is the printing.

And two traps:

• Publishers who omit the 1. Random House's classic house style stated "FIRST EDITION" in words and ran a number line whose lowest digit was 2, because the words did the work of the
  1. Under that style a line reading 2 3 4 5 6 7 8 9 alongside the words "First Edition" is a first printing. If you strip the words and read only the line you will misread it as a second. Farrar Straus and several others have used similar schemes.
• No line at all. Most books before about 1950 have none. Absence of a number line is not evidence of anything.

The copyright page, decade by decade

The number line is a recent convenience. Older books make you work.

Before about 1900. Usually no copyright page in the modern sense. What you have is the title page date, the publisher's imprint, and sometimes a printer's note at the back. The practical test is negative: a genuine first says nothing about being a later edition. Any phrase like second edition, new edition, nouvelle édition, tweede druk or revised and corrected rules it out. Publisher's advertisements bound at the rear are often dated, and a catalogue dated later than the title page means a later issue.

Roughly 1900 to 1950. Copyright notices appear, and with them the useful convention that the copyright year should match the title page year. A title page reading 1931 over a copyright notice reading 1928 is a reprint. Some houses added their own marks — Scribner's used a capital A on the copyright page for firsts through the 1930s, later with their seal.

1950 onwards. Printed statements become common, and from the late 1960s the number line takes over. The ISBN arrives around 1970; a book with an ISBN on a jacket that its 1965 title page would not have had is a later issue or a later jacket.

Publishers are not consistent with each other and are not consistent with themselves across decades. When a book matters, look up that specific publisher's practice in that specific decade rather than applying a general rule.

Stated first editions: when the words can be trusted

"First Edition", "First Printing", "First Published in 1954", Erstausgabe on a half-title — whether these mean anything depends entirely on who wrote them.

Trust them when they are the publisher's own house style, printed on the copyright page as part of the book's manufacture. That is a statement the publisher made at press time and it is usually reliable, subject to the omitted-1 quirk above.

Do not trust them when:

• They appear in the listing rather than in the book. This is most of the 10,212 lots in the data, and it is why the label is worth only ten per cent.
• The book says "First published in 1954" but nothing else. That sentence describes the title, not the copy in your hands. Every printing of that book will carry it, forever. It is the most-misread line on any copyright page.
• The phrase is "first edition thus", "first illustrated edition", "first paperback edition" or "first edition in English". All of these are honest and none of them is the first edition.
• The book is a facsimile or an anniversary reissue, which frequently reproduce the original copyright page in full.

Book club editions: the trap that catches everyone

This is the single most common way people are misled, and it is worth reading twice.

Book club editions reproduce the trade first edition's copyright page exactly, number line and all. Everything you have just learned to check will tell you the book is a first. It is not, and its value is typically a small fraction of the real thing.

They give themselves away physically, not textually:

• The blind stamp. A small blind-embossed dot, square, circle, maple leaf or similar, pressed into the rear board near the lower corner, usually under the jacket. Tilt the book to the light and look across the surface. This is the classic tell.
• No price on the jacket. Trade jackets carry a printed price at the top of the front flap. Book club jackets have it removed, replaced with "Book Club Edition", or blank. Beware: publishers also clip prices from remainders, and sellers clip them from gifts — so a missing price is a prompt to check further, not a verdict.
• Smaller trim size. Hold it against another book from the same publisher and era. Book club printings are commonly a few millimetres shorter and narrower.
• It feels wrong. Lighter than it should be, thinner and greyer paper, boards that are paper-covered rather than cloth, glued rather than sewn, dull and flat-looking cloth.

Here is the honest data point: in a database of 800,385 lots, only 19 mention a book club at all, and most of those are not what we are talking about. Book club editions do not announce themselves. That is precisely the problem — if you are buying, assume nothing about a modern hardback until you have looked at the rear board.

Issue points

Sometimes even the true first printing exists in more than one state, because a fault was caught and corrected mid-run. A typo, a misnumbered page, a plate printed the wrong way round, a colour changed on the jacket. The uncorrected copies came first, so they are the earliest state — and they are the ones that carry the premium.

These are called issue points, and they are bibliography rather than common sense: you cannot deduce them, you have to look them up for the specific book.

The auction data is full of sellers who did:

• Ian Fleming's Dr No, 1958, described as "first edition, first state" — €610.
• Steinbeck's The Grapes of Wrath, 1939, "with first issue dust jacket" — €750. The jacket state, not the book, is the point.
• Jerome K. Jerome's Three Men in a Boat, 1889, "with all first issue points" — €410.
• A first UK Harry Potter and the Philosopher's Stone "with printing error" — €460. The well-known points on that book are the duplicated "1 wand" in the equipment list and the author printed as "Joanne Rowling" on the copyright page; another lot in the data is titled precisely that way.
• A Prisoner of Azkaban described in Dutch as "eerste staat van de eerste druk" — the first state of the first printing — €2,000.

And the counter-example that keeps you honest: one lot advertises a Chamber of Secrets "3rd Print with typo error on page 245". It sold for €45. A misprint only matters when it identifies the earliest state of the earliest printing. On a third printing it is just a misprint.

Reading the label in Dutch, French, German and Italian

This database is heavily continental European, so these are the words you will actually meet.

Language	First edition	Reprint / later printing	Watch for
Dutch	eerste druk, 1e druk, eerste uitgave	herdruk, tweede druk, 3e druk	The druk number is usually printed plainly on the verso — Dutch publishers are unusually helpful here
German	Erstausgabe, Erstdruck	Nachdruck, zweite Auflage	Auflage means impression; "1. Auflage" is a first printing, but Erstausgabe is the stronger claim
Italian	prima edizione	ristampa, seconda edizione	The colophon at the rear often carries the printing date, not the copyright page
French	édition originale, première édition	réimpression, nouvelle édition	Édition originale is the collector's term; premier tirage refers to the first pull of the plates in an illustrated book

In the data, prima edizione lots sell for a typical €100 across 94 sales and Erstausgabe lots for €98 across 38 — both higher than the €65 that plain "first edition" achieves across 8,968 sales. Do not read that as Italian firsts being better books. It is much more likely that a seller who writes prima edizione on a continental antiquarian lot is describing an older and scarcer book to begin with, and these samples are small enough that a few good lots move the figure.

The five checks, in order

1. Does the copy state anything? Copyright page, half-title, colophon. Note the exact wording — and remember "first published in" describes the title, not the copy.
2. Is there a number line? Lowest digit wins. Check whether the publisher omits the 1.
3. Do the dates agree? Title page year against copyright year. Any mismatch means later.
4. Is it a book club edition? Rear board for the blind stamp, jacket flap for the price, trim size and paper against a known trade copy.
5. Are there issue points for this title? Look them up. A named point moves a book from probably to provably.

Do all five and you have converted a claim into a verification — which is exactly the difference between a lot that makes €65 and one that makes €136.

Then price it properly. Search your title and author for completed sales rather than asking prices, read the comparables the way an auctioneer would, and check the answer against what actually drives value in old books. A first printing of a book nobody collects is still worth very little — age alone barely moves a price, and a genuine signature beats a first-edition label. Identification tells you what you have; it does not decide whether anybody wants it.