Archives for category: Book design

Roger Tory Peterson masterminded a revolution in bird-watching by producing a series of field guides with clear, easy-to-use illustrations printed in accurate color with a consistent orientation of his subjects.

Now of course there had been excellent books of color illustrations of birds before Peterson’s Field Guide to the Birds of Britain and Europe, written with Guy Mountfort and P. A. D. Hollom. But like John James Audubon’s they tended to be huge, expensive, and definitely non-portable: the sort of thing men who’d go out and shoot their bird in order to identify it might use. (Audubon himself would work from dead birds he had shot — he once claimed that it was a bad day if he didn’t get at least 100!)

My copy of the British Peterson guide, which I got while I was at still at school, probably in 1956 or 1957, is showing its age, though the damage is pretty much restricted to the jacket, which miraculously survives. The binding, of bright blue linen cloth is still tight, though, as you can see below, the front endpaper has split down the spine fold.

Paradoxically the jacket scarcely fits the book — in the top photo see the type on the back flap almost rolling over onto the back panel. After so many printings this cannot be because the publisher just measured wrong: it has to mean that for this printing a bulkier sheet had to be used. Probably the paper used on the previous printings became unavailable at the last minute, and the nearest substitute was grabbed at.

As you can see the book was printed by Collins, the British publisher, at their own works, doubtless in Glasgow. Unusually, but justifiably in this case, the company who engraved the halftone blocks is credited. Gilchrist Bros., founded in 1893, continues trading in Leeds, now under the name Sun Strategy.

You can see a video about engraving blocks at Engraving a halftone block. I assume that “reproductions” here refers to the line illustrations, including maps as well as the halftones. In a letterpress job you had to have a block (a cut) made for every picture you wanted to print, which would then have to be fitted in with the type. There had to be some raised image for the ink to be carried on.

The color plates are indeed plates, printed separately on different (coated) paper and combined with the black and white text pages in the bindery/folding department. Given its age, unsurprisingly, the book is smyth sewn. The book is 352 pages long, xxxiv of front matter plus 318 text, and has an additional 64 pages of plates, 42 of them in color. It is bound in eleven 32-page sections each of which (after the first and last) is made up of smaller units around which a four-pager of plates is wrapped. These smaller sections are inserted into their neighboring section, resulting in the plates being distributed evenly throughout the book, each separated by 4, 8, or 16 pages of text. Clever book make-up requiring a clear mind in planning, as facing every plate we find its detailed description of it printed on text paper. More modern books in the stable don’t go in for this elaboration — they just print a chunk of the book on coated paper, binding it all together, leaving you to flick back or forward to the related text pages which end up further away from the color plates than in my early edition. Hand work like that clever inserting plan now costs so much more than it used to that we have devised means of avoiding any such elaboration at every turn. The trim size is 4½” x 7¼” — small enough I guess to fit in a pocket, though I have never carried my copy into the wild. The book cost 25 shillings, £1¼.

Roger Tory Peterson ultimately became a sort of mini-franchise, with guides to all sorts of things, trees, butterflies, and so on, plus of course regional bird guides Eastern USA, Western USA, Mexican Birds, just to name ones I own. In the USA these are published by Houghton Mifflin. Peterson was born in Jamestown, New York in 1908 and studied briefly at The Art Students League on 57th Street in Manhattan, before transferring to the more traditional National Academy of Design in the same building. While he was teaching at River’s School in Brookline, Mass. in the early 1920s he decided to create an illustrated guide to birds of the eastern United States. Publishers turned down the obviously too-expensive book till in 1934 the first edition of A Field Guide to the Birds was published by Houghton Mifflin, who printed 2,000 copies. By now of course they have sold millions. Peterson’s “system” was to show all the birds in a similar pose, with emphasis on key identification features indicated by arrows.

Back in the eighties I was an observer in the production of a two-volume Easton Press edition of Mr Peterson’s American bird paintings in their original size, The Field Guide Art of Roger Tory Peterson — I just happened to be there at the right time. These leather-bound books are huge, 11″ x 17″,* printed by John D. Lucas, Baltimore, on S. D. Warren’s 100# Lustro Dull with color separations by Red Rose Graphics of Lancaster, PA, and bound in Nashville at the Nicholstone Bindery where they’ve added fancy moiré endsheets. Here Mr Peterson’s paintings are presented as art, not as an identification tool.

He describes his “conversion experience” which occurred on 8 April 1920: “It was one of the first warm days of spring, when my friend Carl Hammerstrom, who lived up the street, and I crossed the railway tracks and climbed Swede Hill to explore new terrain south of town. As we entered a wood lot on the crest of the hill near the reservoir, I spotted a bundle of brown feathers clinging to the trunk of a tree. It was a flicker, probably exhausted from migration. The bird was sleeping, with its face buried in the fluffed feathers of its scapulars, but I thought it was dead. Gingerly, I touched it on the back. Instantly, this inert thing jerked its head around, looked at me with wild eyes, then exploded in a flash of golden wings and fled into the woods. What had appeared to be dead was very much alive. Ever since, birds have seemed to me the most vivid expression of life.”


* Audubon’s original paintings were somewhat larger than Peterson’s and were executed on double elephant paper the largest sheet available at 40″ x 27″. They were principally executed in water color but Audubon employed other media such as pencil, pastel and ink. They were definitely not all posed in the same position as an aid to identification as were Peterson’s. English engraver Robert Havell Jr. printed them on the same double elephant size sheet — they measure 29½” x 39½”: probably they got a little trim on all four sides. Printing by a combination of aquatint and etching, was of the outlines only. The colors were added by hand: at one time Havell had fifty men and women working on the job. The prints were offered for sale loose as folios of prints

CorelDraw offers you a free ebook, Preparation for Offset Printing. Go to their website and fill out the form.

I haven’t done this, so I can’t comment on the value of the book, but at that price what’ve you got to lose?

Not too many of these have been produced in the twenty-first century I guess, when pretty much everything is done digitally, but a mechanical was the way we used to create a jacket in the second half of the last century.

You’d start with a board. Measure twice and go to the guillotine to cut once, to yield a piece of white board slightly larger than the size of the jacket you were making. If we are talking about a 6″ x 9″ book, that would mean 4 inches for the back flap, 6⅛” for the back cover, whatever the spine width would be, 6⅛” for the front cover and another 4″ for the front flap plus at least an inch of white space all round. The jacket would be 9¼” deep. If there was to be a solid color you’d get out your drafting pen and ruler and draw a box allowing ⅛” bleed top and bottom, and ¼” overlap onto the front and back flaps.

You’d have gotten the type for the jacket typeset by a local supplier, proofed and corrected. This type would include the title etc. for the front and the spine, all the back copy, usually adverts for a couple of other related books, as well as the flap copy describing book and author. It should also include type for the stamping die. (In some instances you might do the title on front and spine using Letraset, and of course lots of jackets used hand drawn script.) Obviously, before you sent the setting copy off to the typesetter, all spec’d up for typesetting, you’d have to have had a clear idea in your mind what the jacket was going to look like, even though you might have done nothing more than envisage it. At the end of the process the typesetter would send you a repro proof (a reproduction proof) — all the type output on photo paper in galley form. With your Xacto knife and a transparent gridded ruler you’d cut out all the type you needed for the front of the jacket — probably just title, subtitle and author’s name. If your title was going to be in two lines, you’d end up with two little bits of paper, maybe another for the subtitle and then one for the author’s name. Lay these out in approximate position on the ruled-up board and adjust them this way and that until they look like you want them to look — general positioning and the balance between one element and another. Then get your Cow-Gum tin and unscrew the top and appreciate that unforgettable smell. That top comes with a brush attached beneath it which you use to paste glue onto the back of your little bit of paper carrying the first line of the title. Using your tweezers lay it down in position and use your gridded ruler to ensure it’s straight, parallel with the top trim, and aligned correctly before you press down on it to stick in in position. After gluing you can move the type around a little with the tweezers to get it just right. Ditto with all the other elements. Then on to the spine, back ad, and finally flaps. If you can’t fit the entire flap copy onto the front flap you may need to cut a few lines off it and place them at the top of the back flap, moving the author’s bio down to make room. Maybe you’ll have to cobble together a line saying “Continued on back flap” — which you can probably manage because you’d always insist on two sets of repro proof.

Once your paste-up is complete, you’d cut a piece of tracing paper to cover it all, bend it a couple of inches over to the back of the board along the top and tape it to the back. When you folded it over your pasted-up mechanical it would not only protect the pasted-up type, but allow you to draw on it the instructions for the color breaks, indicating which bits of type or other design elements like pictures were to print black, which red or whatever, which were to drop out to white etc. If the graphic elements were just line drawings, they might be pasted onto the board just like the type. If the graphic was to overprint type it would be pasted onto a separate sheet of tracing paper and pasted in position in just the same way as, and below the protective tissue overlay. A halftone would be handled in the same way, except that the original photo would have to accompany the mechanical so that the printer could originate the halftone themselves, and incorporate it into the job in negative film.

Computers have simplified the job, but let us not disregard the psychic benefits of using your hands as well as your brain.

See also Mechanicals.

During my life time there has already been a huge shift in the book business from, crudely put, aesthetics to commercialism. I came into a business where it was assumed that every book would be made with an optimal individualized appearance whatever that might mean. There was a sort of mystical view that every book had an ideal form which it was our job to release to the waiting world. We now stand on the threshold of a business where every book will be produced in a standard format. In the olden days, if Crawford’s Roman Coins required that the images of the coins should be printed in an oversized format by collotype in the Cotswolds, they’d be printed by collotype in the Cotswolds and the sheets shipped to the plant where the rest of the text was printed, so they could be bound together. If this meant the book would cost £10 or £20 more than it would have done if they’d been printed by offset, well, that’s what it costs to do things the “right” way.*

Now this approach to publishing is not totally dead. I recently had correspondence with Gordon Johnson about a scholarly edition of the two volumes of Tod’s Annals and Antiquities of Rajasthan, published in 1829 and 1832, where it seems a similar level of care and attention is being devoted. Not dead yet, but much less common than it once was.

In the months to come most book publishers are going to have to confront this issue. Problems of paper supply are not going away any time soon. Getting your books printed must be more important than getting your books printed exactly as you have envisaged them. Thus if you have always done your books in a 5¾” x 8½” format, you may just have to face the reality that, if you can only get paper for 5½” x 8¼”, having your books come out a little bit smaller is a whole lot better than having them arrive six months late, or worse never come out at all because neither you nor your printer can source paper in the required roll or sheet size.

Publishers who have embraced print-on-demand have already accepted this necessity. To get into the program you have to adjust your book to the standard specs that are on offer. This makes absolute sense when you remember that you are printing one copy only — no supplier’s going to switch paper every time they are called upon to print a single copy of your book. This is now going to become the necessity for almost all books. If your printer can print this book as a 6″ x 9″ on a white paper bulking moderately, you’d be a fool to insist on having it as a 6⅛” x 9¼” on a thin cream sheet, and waiting indefinitely for this configuration to come about.

The harsh reality is that almost nobody (except for people in publishing and the odd bibliophile) gives a hoot about whether the book’s this size or that size, on white or cream paper, fatter or thinner, with clever foil stamping on the jacket or not. They just want to read the damn thing. So for a publisher to insist on not giving it to them for artificial reasons is just dumb, dumb, dumb. Yet over the years probably every publishing house has acted in exactly this way on many occasions. So get over yourselves — if someone wants your books, let them have them. The word “book” doesn’t come with a paper specification attached to it.

Now I write this as someone who has spent most of his career trying to make every book a unique and beautiful object. A deeply satisfying and fulfilling task. But I have had to face reality. It’s not altogether different from the tweed industry in my home town. In large mills, yet mills which were small by international standards, we used to weave high-quality, individually designed woolen fabrics. This tweed was much beloved by the bespoke tailoring trade in London, and latterly (late 19th century) in America. You may have noticed that taste in clothing has changed and now no longer does the weavers’ song apply: “the weaving is a trade that never can fail as lang as we need ae cloot tae keep anither hale” (as long as we need one cloth to keep another whole) — who patches a garment nowadays? Apart from me! As a child I remember being told by a mill worker that the tweed trade was a good business “until America imposed the tariff”. I was shocked years later to discover that the tariff dated from the nineteenth century! The McKinley Tariff Act of 1890 imposed an import duty of 49.5% on imported woolen goods, increased to 57% in 1897. In the eighteen eighties annual tweed production in Galashiels was valued at about £1 million (something like $185 million in today’s money).† In 1890 75% of the town’s trade was with the United States; by the beginning of the 20th century this had fallen to 5%. Nobody could be found to buy high quality tweed any more: the configuration of the mills prohibited the efficient manufacture of cheaper fabrics. Consequence; continuous decline. I plan to visit Gala in April and will be fascinated to find if there’s a single tweed mill in operation: during my childhood there must have still been a couple of dozen of them, and I had vacation jobs in half a dozen of them: here’s one of the larger ones.

The remaining half of Netherdale Mill

Modest structures, yet beautiful in their functional design and honest materials.

Beware, beware, oh you publishers — by all means keep your materials honest, but don’t get too fixated on them. Use the best, but if that wonderful sheet isn’t available, use another. Production and manufacturing people are used to compromising — the real problem is with management and especially with editors. Editors are the ones with the most in the way of personal investment in the product, and they tend to have the authority to insist. Without needing to say it, they radiate the idea that when they say we have to have this or that feature in the book, they are speaking for the author; and who’s going to say no to the author? So off we all rush and spend a bundle trying to get ingredient X because (it’s been implied) we’ve promised the author. Editors: stop making demands. Stop insisting on features which can’t be obtained. If your books don’t make it to market on time your company will go under: this is no longer a game; it’s all about survival. We’ve lived through a lengthy period when publishers could snap their fingers and book manufacturers had to fall in line. Times have changed, and now we need them more than they need us. Learn to compromise: if you don’t bend you’ll break.


* We were of course marginally aware that at the end of the year it was a good thing if the business had more money on hand than it had had at the beginning: but this was not seen (by most of us) as the primary function of the company.

† The population of the town peaked in 1890 at 19,500. It’s now just 12,000, 1,000 less than in my youth. In 1801 there had been 844 inhabitants. Behold the industrial revolution at work.

Where does the word dingbat come from? Could it be a mysterious Australian mammal, a cross between a dingo and a wombat? Guess not. It does sound a bit Germanic, with that Ding an sich up front, but the Oxford English Dictionary refuses to commit its corporate self on etymology, stating under that heading “Origin uncertain”. Less rigorous that the Oxford Dictionary editors, I am always ready to imagine a German-speaking USA immigrant influence in the formation of words like this. Dingbat seems, in any of its meanings, to date from the mid-nineteenth century, so such an origin could be possible. One of the meanings listed by the OED includes reference to “thingummy”, a word I was charmed to find in such a formal context. This school playground slang word probably sums up the whole thing. But the earliest reference in their entry on thingummy surprisingly dates to 1737. (Given that meaning, it is far from amazing that another euphemistic usage of dingbat is “penis”.)

Nevertheless, what dingbats mean to me is a font of typographical symbols. As the OED puts it, a dingbat is “A typographical device other than a letter or numeral (such as an asterisk or rule), used to signal divisions in text, to replace letters in a euphemistically presented vulgar word, or for ornamentation. Also in plural: a font or typeface consisting of these.” This meaning didn’t come into existence until the very end of the nineteenth century. I wonder what they were called before that — thingummy-jigs? The usage of dingbats “to replace letters in a euphemistically presented vulgar word” takes us straight to grawlix, and must surely now be “obsolete”, as we no longer feel much need to disguise “vulgar words”.

Here are a couple of dingbat collections from Mergenthaler LinoType:

Dingbats live in the same world as Type ornaments. Type ornaments are little drawings which you know you’ll need again and again, so having made them in metal pieces, you store them for the next occasion on which they’ll be called for. Dingbats are little ornaments which you’ll need again, and again, and again, indeed so often that it makes sense to incorporate them into your system just like the letters a, b, c and so on. So dingbats are type ornaments that recur so frequently that they end up being typeset rather than inserted as a block or cut.

Don’t ever do this.

OK, with a title made up of four letters we can see there was maybe a temptation to align them vertically, but damn it, it’s surely not much of a temptation to overcome. The Romans knew that SPQR could only be displayed in one way. The Roman army was a ruthless, efficient killing machine, but they did it with typographical style: they’d never have gone into battle with a standard showing SPQ and R aligned vertically.

If someone’s holding a gun to your head forcing you to align the title vertically, at least insist on a typeface where the characters are of as similar width as possible, and don’t go for the swashed version of any of them. Here the swashed Q is about twice as wide as the S. But the rule is — don’t do it anyway.

The only exception to this rule is when you are doing the job with a paint brush in hand, and are able to “force” all the characters into the same width. For example MASH (Mobile Army Surgical Hospital) works, but only because army painters were able to use a stencil designed so that Cap M would be the same width as all the other letters. If there’d been a Cap I there, they’d have given it massive block serifs so it was almost like an H rotated 90°.

Here’s another example where they just about get away with it: maybe digits are more forgiving than letters, though the typeface used is a sort of military-look stencil-based design, and can be given a rather uniform width by the use of generous slab serifs.

Let’s face it we don’t read stuff from top to bottom. If you’re Chinese you can do this — their script is designed to be read vertically, and each character occupies a similar amount of horizontal space, so the symbols look right one below the other. They also look right aligned horizontally — take up your complaint with the ancient Phoenicians. In an alphabet where different characters occupy different amounts of space you can’t align stuff vertically — it just looks


While we are looking at the SPQR cover, maybe we should comment on the front cover, which is a letterspacing mishmash; actually a general mishmash. The foil-stamped bestseller lines and the subtitle are quite well letterspaced, though ROME needs a bit of attention I think. But if you’re going to letterspace MARY that generously (though I’d ideally like a little more space between R and Y), then you have to inject a little more space into BEARD. The space on either side of the A, establishes the pattern, and the other locations require similar space to save the day. More letterspacing will mean that the size of the author’s name needs to come down. Reduce it about 10% to fit it onto the cover front once it’s been properly letterspaced, and it’ll look better. Yes. Yes, the author is the selling point, I know, but let the title be a little larger, especially as it’s only four letters long.

But the title is a problem with its little golden tripods interfering with the layout. There’s a lot of gold stamping on this cover: no expense spared. I suppose it was inevitable the designer should go for that maximally swashed Q — after all if you’ve got it, flaunt it — but it really screws up the balance of the title. That counter on the Q represents by far and away the greatest amount of space in the whole line. Without getting my ruler out I’d guess we need almost 12 points more space between S and P, 9 points between P and Q (you always have to mind them), and maybe 4 points before the R — maybe more in order to keep its left leg sufficiently clear of the swashed tail of the Q. I expect we can exonerate the cover designer from responsibility for that silly author photo, ruining the alignment and nafly proclaiming “As seen on TV”. Are there really people out there who are impressed that Professor Beard has appeared on television? The skimpy border at top and bottom adds nothing other than slight confusion.

Still, the book was a great success — no doubt still is — which just goes to show how much less important design is than authorship.

I don’t know whether or not Jack Bowles (1922-2017) invented the idea of the three-em indent, but he was certainly the one who introduced it at Cambridge University Press. It provoked quite a storm of conservative objection.

When you design a text page you’ve got two basic options — leaving aside wild irregular, chaotic arrangements, which can never work — it’s either going to be a centered design or a flush-left one. Centering all elements is the traditional way of handling things, with flush-left coming in in the twentieth century as the modern, Bauhaus-lite alternative. (Rare because odd is a flush-right design.) The two alternatives should never be mixed together — that’s a great way to sow confusion — though you will find novels perfectly nicely designed with a centered layout but with the folios at the outside margin, either at the top or the bottom — and this can work because novels tend to have rather few design elements: you have text, you have folios, you have chapter openings, and maybe you have a running head, so it’s not too hard to set up the expectation that this or that element will be found here or there. An academic book will potentially have many different elements — in addition to those listed for the novel there may be several levels of subheadings, quoted extracts, tables, lists, diagrams and photos (artwork), formulae, footnotes, bibliographical references and so on, each requiring some way of straightforward signposting and differentiation. Aligning every element flush left may not be enough for an element-heavy book. Introducing a regular indented alignment doubles your options, and can contribute to a counter-intuitive cleaning up of the overall page layout..

Think of your type area as a flag, and imagine it rotating on its left-hand margin — that’s a flush left design. Now think of it rotating on a pole ½” from the left hand edge — you now have a three-pica indent which you can use in addition to the left hand margin to put design elements on. You can’t however just use this notional flag pole once in a while: you have to establish it in the reader’s subconscious by using it on every page.

I can’t find any good examples from those days to show an illustration. It was used most often in science and social-science monographs where there were often many elements calling for different treatment. Here’s a couple of pictures of a monograph on the elm tree.

This is a bit of mess, I have to confess. The book, R. H. Richens: Elm, is a 7″ x 10″, double-column book. You’d go to 7″ x 10″ because there was lots of art which needed to be reproduced as large as it could be. The photos in this book are, I fear, rather murky, which also would militate in favor of making them as large as possible. No doubt because of the double-columness of it all, the indent is in fact a 2½-em indent, but the principle remains the same. Running head and folio are both on the 2½-em indent, one on the first column, the other on the second. The caption is also on the indent, though the art is flush left to make it as big as possible. Chapter number and Chapter title are on the indent, as are all the display quotes shown. (As befits a university press, even in a book about a tree, Chinese, Russian and Latin are shown in their original, then translated. You have to assume your readers can understand every language.) The paragraph indent ends up having to tag along on the 2½-em indent too — looks a bit odd at first, but one quickly gets used to it. Not shown on these pages: there is only one level of subhead in the book, on the indent of course.

So I think what the examples illustrate is that using a 3-em indent requires taste and judgement. Whether it could ever be successfully applied in a double column format I won’t speculate, but here I think these extra imaginary flag poles get in the way. There really aren’t too many design elements in the book, and the 2½-em indent becomes a bit of a sledgehammer crushing a nut. To show that readers do react to design I can confess that, though I’ve had this book since 1983, I haven’t managed to read it yet — and I am interested in the elm — I swear! I recently chose as my “tree” for a WNYC radio program, one of the ancient elms in Fort Tryon Park, seen here, leafless, between a parting of the ways with the Cloisters in the background and a street lamp in front. That SUV is just going past it. I have to photograph it once a month, so I can learn to appreciate its life-cycle.

Dutch elm disease has reduced the number of elm trees dramatically over the past thirty years. These ones seem to be hanging on in splendid isolation.

Shorter quotes will just be set as part of the text line enclosed in “quotation marks”, also called inverted commas. Britain tends to favor single inverted commas with double used within the quote when needed. America favors the opposite, as do I in this blog. If the quote is of a certain length it will be set separately from the rest of the paragraph. The switch-over will come at different points in different houses: maybe 50 or 60 words. The Chicago Manual of Style says 100 words. More than a couple of lines anyway and you begin thinking about display. Your house style will (should) provide guidance.

As ever in book design the aim is to make the designer’s thought processes invisible to the reader. So you don’t need to clobber the reader with markers of difference between text and quoted matter. You can set your display quotations (called block quotes in America) in smaller type; you can indent them left and right, or just left; you can set them unjustified (ragged right); or you can leave extra space above and below. You should avoid over-marking the display: in other words if it’s set smaller and indented left, don’t feel you also have to leave extra space above and below. The reader just needs to know they are moving from the author’s words to some supporting words from another authority (who will of course be identified by some proper referencing system). After a couple of instances the reader will subconsciously be able to acknowledge the move from one to the other: they should never need to have any thoughts or feelings about the matter. A displayed quotation should not start and end with inverted commas: its layout will be enough to tell you it’s a quote. The WordPress template for this blog clobbers the reader over the head with its display layout; so I don’t often use it. Here’s why:

It is usual to display verse if there is at least one complete line. If the author runs some verse on in the text, consider whether it should be displayed, or if not, whether line breaks should be indicated by capital letters (if the original had them) and/or spaced oblique upright or oblique strokes; strokes are the only completely clear way. If a displayed verse quotation starts with a broken line, the first word should be indented to approximately its true position in the complete line.

Judith Butcher: Copy-editing: The Cambridge Handbook for editors, authors and publishers, Cambridge University Press, 3rd edition 1992

Well this is horrid, isn’t it? Why’s everything italicized? And those ugly inverted commas! Ugh. I try to mark longer quotations by making them blue rather than black, but that’s often a software struggle too. But the basic problem here is that there’s way too much difference between text and displayed quote: massive indent, smaller font, plus line space above and below, plus italic, plus, just in case you’re still in doubt, those tremendous double inverted commas.

Here’s The New York Review of Books dealing with a half-and-half situation: a run-in quote which continues as a display quote:

The first, run-in sentence is 30 words long. NYRB marks the display quote by indenting left and right, and by an additional line space above and below. One might say that the distinction might be made just as clear without the space above and below, but the trouble is the indent left is the same as the paragraph intent, so momentary confusion might easily be introduced. The opening quotes are coincidental — one might argue that they ought to be single quotes — no doubt this is a topic that could occupy those who need a break after hours of debate about the number of angels who can dance on a pinpoint.

Space is always at a premium, so your instinct will be to avoid the line space above and below. If you go for smaller type, you will run into alignment difficulties when text setting is resumed — this can be mitigated by the use of a variable space above and below, allowing you to get back so that the bottom of facing pages will align, as well as the top. Like all design, it’s always a compromise.

Designers, please remember, we readers aren’t stupid: we wouldn’t be reading this book if we were — we can figure a lot of things out from fairly minor hints.

Stephen Sondheim’s two-volume book of lyrics was a clear precursor of Paul McCartney’s.

The layout is less elegant: of course theatre lyrics require more space, so there’s a three-column layout. It’s a Knopf book, so we get more production details than with most books. It’s set in Berkeley Oldstyle composed at North Market Street Graphics in Lancaster, PA, and printed and bound by Quad/Graphics in Taunton, MA. Volume 1, Finishing the Hat was published in 2010, followed in 2011 by Volume 2: Look, I Made a Hat. There’s something a bit odd going on with bold face. Sometimes it’s used for footnotes, sometimes for commentary These books may, I guess, be a case where running feet are not altogether unjustifiable.

Sondheim is interested in teaching, and his books work almost like a college self-supervision. He’s eager to show us where he went wrong, and to explain why he made this or that decision. Musical theater lyrics, and popular song lyrics are a bit different of course, but much of what he says translates I’m sure.

“Lyrics, [for the musical theater] even poetic ones, are not poems. Poems are written to be read, silently or aloud, not sung. Some lyrics, awash with florid imagery, present themselves as poetry, but music only underscores (yes) the self-consciousness of the effort.” “Music straightjackets a poem and prevents it breathing on its own, while it liberates a lyric. Poetry doesn’t need music; lyrics do.”

When you read the lyrics here or in McCartney’s book, you cannot avoid hearing the melody as you read them. And this points up the requirement placed upon the lyricist to allow the music to call the shots — if the music stresses the “wrong” word the performer has to resist the temptation to overemphasize “and” rather than “think”. To illustrate this point here’s a zany 1989 arrangement of a Sondheim song, triumphantly sung (OK, it is a bit mannered), showing the power of the material to overcome the Pet Shop Boys’ arrangement.

Stephen Sondheim (born 22 March 1930) died on Friday 26 November. What a loss! Two of his shows are currently being performed in New York. Sondheim can’t have been right when he wrote (in his introduction to these volumes) “I used to think that the need for live theater would never die. I fear I was wrong.” His body of work is part of the conclusive argument against.

Costs $100, and has been chosen by Barnes & Noble as its book of the year. Well timed for Christmas, obviously.

As Shelf Awareness tells us: “B&N CEO James Daunt commented: ‘The Lyrics is an extraordinary book. It is stunningly beautiful and a masterpiece of book design. Paul McCartney has fashioned, through the explorations of his songs with the poet Paul Muldoon, a fascinating insight into his life and creative genius. No wonder the booksellers of Barnes & Noble have hailed this magnificent and deeply original book.'”

It is indeed a nicely designed book, but unfortunately Volume 1 opens with a design boo-boo. Page vii is blank. You are not allowed to leave a right hand page (recto) with nothing on it once you have started putting ink on the pages. You can begin the book with as many blanks as you can get away with, but as soon as you’ve printed anything on a recto — usually the half-title will be the first such item — you cannot leave any other recto blank until you get to the end of the book, after the index etc., where once again you can leave as many pages blank as you’d like.

Who is it that makes this rule? Not sure. It’s convention, but convention so rigid that everyone in the business seems to have silently internalized it, and looking at the spread above immediately recognizes that something’s wrong. Maybe it began in the same sort of way as the half-title convention did, so that you’d not have a blank on the outside of a section after it had been folded. Such a blank could lead to the possibility of the section being included at the wrong place in the gathered book block, whereas if it showed a bit of text and a page number it couldn’t (as easily) be gathered in the wrong sequence.

The Lyrics error is so “obvious” that it would seem that it had to be a mix-up. The epigraph on page vi must have been intended for p. vii. With a book of this magnitude it’s surprising that this was missed in proof or even at f&gs stage. Maybe they did notice it in f&gs and couldn’t reprint the sig (as most publishers would want to do) because of supply chain problems — scarce press time, or more likely paper backlogs.

I might have preferred to see this pair of lines dropped a bit lower on the page, a comment which would also apply to the dedication on the preceding page, which is aligned flush right. It’s no big deal, but they both look a bit lonely way up there. Having said that, I have to reflect that the alignment of this pair of lines, both flush left, rather militates against my contention that this was just a mix-up. If these lines had been intended for the following page, they would have been aligned at the right hand end of the second line. Maybe there was once something else on this page and it got dropped at the last moment?

When Bob Dylan won the Nobel Prize there was a pretty chaotic and inadequate effort by his publisher to make his newly hallowed lyrics available to an eager public. Paul McCartney has been brilliantly served by his publisher, Liverright (a division of W. W. Norton), and by his editor, Paul Muldoon who contributes a critical essay. The 154 songs, not just Beatles’ songs of course, are printed in alphabetical order each with a substantial and informative commentary by McCartney including lots of four-color photos. Perhaps almost over-cutely the book is set in a specially designed typeface “Rigby” — but even this turns out to be a success.

Get it for Christmas. Shop early. The book’s printed in China, so they won’t be restocking this year.