dot-font: Tables of Contents
The table of contents is one of those essential pieces of any publication that gets short shrift. Some contemporary magazines go in for highly visual tables of contents, but in most publications, the ToC is an afterthought — an obligation, rather than an important part of the navigation of the information. In a book, especially, the table of contents may be treated almost perfunctorily.
Most of the time, a table of contents just lists the elements in the publication and tells you what page each one is on. That’s the most basic function. Other contents pages may add further information: authors’ names, section titles, even fairly extensive descriptive text. But the first purpose is simply to point the reader to the contents. If the ToC doesn’t do that, it’s failing in its task.
It’s surprising how many different ways you can organize the few elements in a table of contents. Let’s look at a few real-world examples, both good and bad. Most of these are in books published by independent presses in the last few years.
Connect the Dots
Figure 1 shows a ToC with a very traditional arrangement: titles ranged to the left, page numbers ranged to the right, with a dotted-line leader connecting one to the other. The ornaments and the choice of text typeface in Best Short Novels 2005 (ed. Jonathan Strahan; Science Fiction Book Club, 2005) echo book designs from the 1930s by W.A. Dwiggins and others, though here the text is unusually large, and the use of the enormously wide typeface Americana for the word “Contents” is anachronistic and clashes with the rest of the page.
This design is functional, but a little oversized. If I were executing a design like this, I would make all the text type smaller, letterspace the all-caps titles slightly, and use old-style figures for the page numbers, not the cap-height modern figures used here. (Apart from my preferences, old-style figures are the natural complement to Baskerville, the typeface in this example. John Baskerville didn’t cut figures like these; they’re just the quick ‘n’ easy default found in the digital font.)
Figure 2, the contents page of Black Juice (Margo Lanagan; Eos, 2004), is based on the same structure — titles left, page numbers right, the word “contents” centered above — but stripped down. The type is all much smaller than in the previous example; in fact, the type here is smaller than the text type in the body of the book. Perhaps the designer was trying to emulate small caps, but these are not small caps; they’re full caps set at a smaller point size, so they look light and spindly. The page numbers are still set in cap-height modern numerals, instead of the old-style figures that would look appropriate with small caps.
And why are the page numbers aligned left, rather than right? That looks very odd, as though the designer had set a tab but made it a flush-left tab. If the word “CONTENTS” was meant to be centered over the whole table of contents, then the page numbers must have been set flush-left in a fairly wide right-hand column. (Oddly, the word “CONTENTS” is set in a different typeface from the rest of the page — different, yet not different enough to stand out visually or have any purpose.) Something is peculiar here.
The most obvious difference between this page and the previous one is the lack of leaders. Many designers consider a leader to look old-fashioned, so they eliminate it, to make a cleaner, more modern-looking page. But they’re still using a very old-fashioned arrangement on that page, and the lack of leaders means that every title sits clear across the page from its page number. The reader’s eye has to track the wide white space correctly, with no visual aid, to see which page number goes with which story. Visually, this page is two columns, one of titles, the other of numbers; this invites the eye to read not across but down.
Figure 3, from Starwater Strains (Gene Wolfe; Orb, 2005), is a more extreme version of the same arrangement, with acres of space between title and page number. But the type is more easily readable, so in some ways it works better. (Here, the modern figures are appropriate to the typeface.)
When you add further elements to the table of contents, it gets more complicated. Figure 4 shows a detail from the contents page of Daughters of the Earth (ed. Justine Larbalestier; Wesleyan University Press, 2006); since the book reprints short stories and follows each one with an essay about it, the table of contents needed to group these paired elements together.
The arrangement of title and page number is still old-fashioned, with the page number off to the right. But the long titles, and the added information below them, reduce the gap to a manageable size; it’s not hard to associate the page number with the correct story.
The contrasting typefaces, however, don’t work very well. To stand out unambiguously, the differing elements need at least two kinds of contrast. Here, the sans serif type in the secondary elements (a version of Futura Light) is still too busy. Maybe if it had been set a size or two smaller, and letterspaced a little so it wasn’t so tightly packed, it might have worked. A greater difference of weight between the two typefaces would certainly have helped; here, they both look light.
To put my money where my mouth is, typographically, I’ll show you two examples from books that I designed. Figure 5, from Contemporary Newspaper Design (ed. John D. Berry; Mark Batty Publisher, 2006), had to group four elements for each item: title, author, page number, and a short description. By grouping the primary elements and using two kinds of separators (a vertical bar and a centered dot), I could keep them together yet clearly separate, and I could still use italics when necessary for titles within the essay titles. The descriptions are clearly subordinate but easy to find and easy to read.
Figure 6 is from Look at the Evidence (John Clute; Serconia Press, 1995), a collection of essays and reviews. Its contents pages include more complex information; they need to be organized to give an overview of the book’s contents and provide a clear way to find any particular essay or review. There are several levels of organization in the book’s contents: parts, sections, and groups of related pieces. In a sense, the page numbers are the least important elements, but readers must be able to quickly find them for each individual essay. I considered making the page numbers stand out more, maybe by using bold, but that would have made the page too busy; a plain space was the least obtrusive way to set off the page numbers — small enough so they wouldn’t get lost, floating out in the white spaces, but large enough to separate them from any numerals that were part of the entry itself.
Just to show another approach (not mine, this time), Figure 7 is a more graphically bold solution to a similar problem, in that popular book on using type, Stop Stealing Sheep (Erik Spiekermann & E.M. Ginger; second edition, Adobe Press, 2003). The information isn’t as complex as it is in Look at the Evidence, but this design could handle it. Having a second color to work with does help.
Even when the rest of a book is laid out in symmetrical text blocks, there’s no need to treat the table of contents the same way. By its nature, the information on the contents page is different from the rest of the book. It’s more like movie credits than a book page. So sometimes it’s more effective to ignore the margins — the frame of the page — and let all that information clump together within the larger space. (Of course, you still have to pay attention to how and where it clumps together, so in that sense you aren’t really ignoring the margins.)
The simplest way to do that is to center the contents. This means putting each element on its own line (or lines, if it runs too long). It’s not a method to use in a long or tightly packed book, since this design usually takes up a lot of space.
Figure 8 is an unsuccessful attempt to center the contents, yet save space by using two columns, in a book of history called Islamic Spain: 1250 to 1500 (L.P. Harvey; University of Chicago Press, 1990).
The extra-large chapter numbers are meant to visually organize these small, spotty elements, but the numerals look huge. The chapter titles are typeset in faked small caps (just capital letters shrunk to a smaller size, which makes them look too light), and they haven’t been properly letterspaced; they look crammed together and are thus hard to read. The full-size initial caps on all the words in the titles just add visual clutter; if these titles were going to be set in small caps, they should have been all small caps, with no full-size caps at all. And of course the numerals should all have been old-style figures, not full-height modern figures. (If the typeface had specially designed small-cap figures, that would have been an acceptable alternative. But this typeface, ITC Galliard, does not; and most of the time, I think that old-style figures are more effective with small caps.)
A much more successful way of grouping elements along a central axis is seen in Figure 9 from So Long Been Dreaming (ed. Nalo Hopkinson & Uppinder Mehan; Arsenal Pulp Press, 2004). Visually, it works like a mobile, with the center of gravity running down the middle and elements of varying length hanging off that center line.
The use of old-style figures and true small caps makes this look elegant and also makes it easier to read. The italic type for story titles contrasts with the roman type for the authors’ names, but in fact they could all have been set in roman if necessary; their position alone makes it clear what’s a title and what’s an author.
The spaces between columns are a little big; the whole design might have held together a little better if these had been smaller. But that might have drawn attention to the way the page numbers are aligned flush right, which is logical, but a contradiction in this otherwise centered design. I suspect that the designer spent quite a while experimenting with these small details to get them right. When they’ve been done correctly, the reader doesn’t even notice; the design is just doing its job.
One final example shows an asymmetric approach to the “centered” arrangement. Rather than being actually centered, the contents page in Figure 10 uses dynamic symmetry to group the information. It’s another book that I designed, Dot-font: Talking About Design (John D. Berry; Mark Batty Publisher, 2007), so I’m responsible for its success or failure. Here, I had the luxury of a second color.
Whether the contents of a book are complex and complicated or simple and straightforward, whether the visual style of the contents page is symmetrical or asymmetrical, the purpose of the table of contents is to provide access to what’s in the book. It is a miniature exercise in information design. The contents page needs to blend into the design of the book as a whole, yet it is fundamentally different from any other element in the book. That’s a challenge.
Designing a table of contents requires subtlety and attention to detail, and all this effort will go largely unrewarded. In a sense, the design has to get out of the reader’s way, but that’s not an easy thing to achieve. Only when the design fails does it draw attention to itself; when it succeeds, it’s invisible. But a few of us, those who are both designers and readers, will notice.