I’ve railed against the use of automatic leading before, but in this article I want to point out how even when carefully set, consistent leading can look wrong. It’s another case of the numbers in a software program saying one thing but your eyes saying another.
The page you see in Figure 1 is a classic example. It’s a nicely designed page (from the Sunday magazine of France’s Le Monde newspaper), but my eye is immediately drawn to the subhead on the right, with leading that seems to alternate between looser and tighter.
Figure 1: The leading in the lines in the deck at the right seem to tighten and loosen in waves, creating a disruptive overall impression of visual disorder. Click the image below to see a larger version.
The deck type is arguably too tightly leaded to begin with (it’s actually negatively leaded, at 30/26) but the problem lies elsewhere. The tight leading merely exacerbates it, making it more visible.
The real problem is that lines with few or no ascenders appearing below lines with few or no descenders will always appear more loosely leaded than other lines. Likewise, lines with many ascending characters appearing below lines with many descenders will appear to be more tightly leaded than their neighbors. In the Le Monde Magazine page, this gives the whole deck a blotchy and irregular aspect. Figure 2 shows this type block in isolation for a closer look.
Figure 2: When seen isolated and enlarged, it’s easier to tell that source of the apparent leading inconsistency is actually the coincidence of ascenders and descenders between particular lines. Click the image below to see a larger version.
This is a problem generated by happenstance, but just because it was no one’s fault doesn’t mean it shouldn’t be fixed. Figure 3 show the results of my efforts to tweak the leading of individual lines to create a consistent overall leading feel, even at the expense of numeric inconsistency. While I tightened some lines’ leading, I loosened most to various extents.
Figure 3: As with kerning, if you look very closely at leading in display type, it’s apt to look inconsistent no matter how much fiddling you do with it. The goal, as shown here, is to create an overall sense of consistent spacing and — above all — to eliminate eye-catching variations. Click the image below to see a larger version.
Leading anomalies like this are not rare or isolated events, and once you start looking for them, you’ll be surprised how often they pop up. The examples in Figures 4 and 5 took me all of three minutes to spot in paging through a couple of recent issues of The New Yorker.
Figure 4: There are several factors at work to make this headline look irregularly leaded. The angle of the W in the second line is echoed by the v in the line above it, and having all four letters in the line being ascending characters makes the second and third line appear too tightly spaced. The coincidence of the W stacked over the A makes the fourth line start to appear too loose already, but the word space amplifies the effect. The bad kerning and slack tracking in the last line spread their impression of looseness into the line’s leading, making it appear to sag too low beneath the one above it. The quickest fix would be to open the leading of the third line to harmonize with the one above it. In addition, I’d hand-kern the last two lines to narrow the word space and fix the slack spacing in “Monster” and then, if still necessary, slightly tighten the leading of both lines.
Figure 5: In this tightly leaded headline, the last two lines nest neatly together at the expense of making the top line look like it’s drifting upward. You can’t tighten the leading of the second line very much without it crashing into the descenders in the first line, so there’s only one solution: open the leading of the third line to restore a sense of visual balance.
Perfecting leading is complicated by the fact that it’s difficult to see the problem onscreen while composing pages electronically. The low resolution of a computer monitor often masks problems this delicate, as it can with so many other typographic refinements. But in print, the problem becomes terribly obvious. Sometimes an enlarged screen version viewed from a few steps back from the computer will help. But ultimately, the real proof is in the printing.