Brilliant Deduction: About the type…

I already wrote, and posted, an “about the type” document for Brilliant Deduction… but there’s a lot more to say about the type than just “Adobe Caslon and Bodoni.”

Some years ago, someone posted somewhere a remark about self-publishing and DIY typesetting along the lines that “there are plenty of perfectly good templates available that you can just use; I, however, am the crazy mayor of a little one-person community called Crazyville and insist on obsessing over serif comparisons and leading and basically doing it all the hard way.” I wish I could remember the source, but I do recall thinking at the time that “well, maybe as comic exaggeration, otherwise I think you’re getting a little carried away in diagnosing mental illness…”

After typesetting my own book, however, I’ve had second thoughts about that initial conclusion.

If nothing else, it certainly feels that there was an element of The Crazy in how much time and effort I spent on the typography for Brilliant Deduction. (The print versions, at least; the ebook follows current “best practices” of a relatively dumbed-down typography, though getting there nonetheless involved its own set of challenges.)

There are just… so, so many options.

The last time I typeset a book, it was mostly a new experience for me; I had to go to the shelf just to get a sense of the basic conventions for page design, organization, etc. So I played it very safe, choosing good old Garamond for most of the copy. This time, I actually ended up with an equally reliable choice for most of the interior text, but a lot of exploring took place before I finally settled down on Adobe Caslon.

I designed the cover, first, and found myself drawn to the Bodoni family for display type. It was never a real serious contender for my main body copy, though; I think modern style typefaces can work as book type, but only in particular circumstances, which usually involve 1) rather short word counts in addition to 2) being just plain rare. I did a test page or two, just to try it out, but never felt like any variant of Bodoni was appropriate for my main body copy.

I tried plenty of other options, from “hybrid” styles like Transitional to other classics like Bembo to Sabon, the workhorse from my days at Old D.U. Eventually I came down to Bembo, Caslon and Sabon, and pushed the leading and point size of each this way and that before finally getting that “feeling” that Caslon was The One. Massively unscientific, but then graphic design is more of an art than a science. I just felt like my final specimen page of Caslon just had that “comfortable old shoe” feeling more than any other sample page I’d looked at; it seemed like this type made reading the page feel the most naturally effortless. So, Caslon it was.

But of course, things are never so simple in Crazyville. After some further trial and error, I had the basic page design and body copy settings established. Huzzah. That only left, hm, something like seventeen more styles to define before all was said and done. Had I written a novel, it might have been simpler. But I didn’t, and thus there was always one more type of information wanting its own typographic style. The basic blockquote: should I indent both sides, or just the left? Use the same point size or something smaller? The sources: what about them? Headings. Subheadings. Captions alone included, potentially, all of: main identifying text, artist (i.e. photographer or illustrator), a publication reference, my source, and the source’s reference number.

I’m going to write another post, later, about the self-publishing experience, but this is one example of why I don’t envision technology just removing all intermediaries between author and audience any time soon. There’s just so much work to do that can’t really be automated because so much of it involves making decisions, and of course the one reliable activity that automation seems unlikely to take over any time soon (we hope) is deciding what people want to accomplish. It reminds me of what I wrote about web site design a while back, as well, as there’s the same kind of irreducible core of decision-making activity that you really just can’t speed up. As with that, you can have the slickest templates in the world, but you still have to 1) select that template and 2) figure out what pieces of your unique, new data get plugged into it and how.

So anyway, I spent a lot of time on the interior, partly because it takes a lot of time when you have even moderately complex information like I had (and want to deal with it with even a moderate degree of sophistication). Also, however, because I got a little showy. I won’t deny it. I felt like being a little more typographically ambitious after my last book design outing, anyway, and Brilliant Deduction seemed to justify it as well. Though I think a lot of actual Victorian typography was frankly just ugly, the overall impression of liveliness and theatricality it often leaves seemed both historically and thematically appropriate for my (theoretically) popular-audience history of larger-than life detectives, mainly set in the 19th century.

And so I’m rather proud of what I came up with for chapter pages:

This, too, obviously involved some difficulty. Probably the two biggest challenges, though, were kind of oddballs, and at any rate had to do with how the information varied. I had elected to open each chapter with a quote, because it seemed fun (and flattered my pretentiousness). These quotes varied a good deal, however, in length, but it seemed like the actual text of the chapter should start at the same point each time. This I resolved by floating the chapter title and quote in the center of the two decorative bars, which were themselves the same on each page. Some were thus packed more or less “full” while others, like the above, were more spacious, but they all seemed to work out.

The real big challenge that actually haunted me all the back from the early manuscript stage was what to do with quote attributions. The problem was that absolute consistency and clarity seemed to require redundancy. My chapters themselves are prosaically named for their subjects, and a majority of the opening quotes are also just quotes from those subjects but not all. And I kept thinking, “what am I going to do about this?” It seemed dumb to have a page begin with, e.g., “Isaiah Lees” in big letters, followed by a brief quote attributed to “Isaiah Lees.” Other quotes really did require a separate attribution/explanation, though.

This is another example, I guess, of how as writer and designer I could have simply rewritten my way out of a dilemma. In pretty much all of those cases that I can think of, though, I actually did not; the writer was implacable on wanting his text the way he had arranged it, even though I was the writer. (Writers, I tell ya—!) In this and another case that a friend actually asked me about, recently, I kind of solved things the same way, too: I decided that people will figure it out. My solution was, as seen above, beginning with the chapter subject (and title) in big type, followed by the quote… with an attribution, where needed, treated (parenthetically). This felt like it made reasonable sense as a system: when it’s just a name and a quote, people will probably work out at that the two go together; when there is a separate attribution, the parentheses seem to justify the fact that attributions are only included for some quotes, not all.

Yeah, I really, really over-think these things.

One final note on the typography, though, I can offer as evidence that even I reach a point where the answer is “live with it.” Given the amount of fussiness already described, it’s almost inevitable that I wanted to use old style numbers. It wasn’t a dealbreaker, but their inclusion in Adobe Caslon Pro’s character set was another point in that typeface’s favor. Then I encountered an unexpected problem, even though I’ve worked with old style numbers before.

Above right, the ordinary bog-standard “1” from Adobe Calson Pro. Second from right, the old style 1. Obviously, it looks like a capital I, which of course it kind of is, though it’s a small-cap I (so far as I can tell it’s the very same glyph as the small-cap I, in fact) so it looks different from the first-person pronoun. Except… as seen in the middle of the above sequence, “11” ends up looking awfully like the Roman numeral “II.” Even I found this unnerving.

By the time I noticed it, I was definitely not going back and switching to another typeface, and didn’t really feel like switching back to lining (“regular”) figures, either. As it happens, Adobe Caslon Pro does include a small “1” that looks more like a “1,” second from left; this is actually included for creating fractions, I guess. (At far right, above, is a dotless lower case “i,” just for comparison’s sake.) I thought about using this, but it’s a little short to fit in with the other numbers. And, ultimately, I didn’t feel like screwing around with it, particularly when the number “11” occurs by itself a grand total of two times in the final manuscript, if I recall correctly.

Even in Crazyville, a time does eventually arrive to say “good enough.”