I do not understand how tight leading could speed reading of a fast-paced novel. If leading is either too tight or too loose it slows reading. Tightly leaded prose which causes doubling and reading effort will not make a "page-turner". Poetry whose lines are widely separated will be hard to read as poetry, since the sense often flows across lines (poetry readers rarely "contemplate individual lines", at least this poetry reader doesn't). I suppose there might be something to be said for it with some lyric verse, but not as a general rule.

I would say that approach (1) is the only sensible approach. I suppose one can conceive circumstances where one deliberately wanted to produce unreadability, but they surely must be rare. It is probably much closer to the mark to say that there may be circumstances (e.g. reference works) where readability has to be sacrificed for compactness and leading may be adjusted accordingly.

Yeah, see that's what I thought, but two things bother me about that approach.

Firstly, given that a letter is to a sentence, as a sentence is to a paragraph etc, and that one type face may be more appropriate than another for content (you wouldn't set a book on the Bauhaus in Bembo for example). Doesn't it follow that setting type in one manner may be more appropriate than another for certain content.

Secondly, would that not make it reasonable to develop a set of rules for leading. For example, verdana, set at 12pt needs 2pt leading at 45 characters per line, plus 1pt for every ten characters above 45 per line. And if that's reasonable why wouldn't leading be an inherent part of a type face like kerning?

cheers.

I think there are three forces acting on optimal leading: economy, readability, and atmosphere.

Economy simply wants leading to be minimal. Readability wants leading to be enough to avoid: errant line returns*, and errant fixations to subsequent lines (especially when the type is loose and needs bigger word spaces). However readability also doesn't like very large leading because that increases page jumps/flips. As for atmosphere, it's subjective, and has to do with how airy you want to page to feel (but also the nature of the type - like a Bodoni likes more leading).

* This depends on line-length, as you say. As for "type characteristics such as descenders/ascenders", this is huge, and doesn't get nearly enough airtime. When comparing two fonts for economy for example, people are usually smart enough to vary the point size to match their apparent sizes, but they make the leading identical. You have to worry about apparent leading too.

Pulling atmosphere out of the equation for argument's sake: optimal leading should be as little as possible without affecting readability (unless there's some strange force majeur having to do with economy). BTW, this is also the reason why a real text font has descenders notably shorter than its ascenders.

> develop a set of rules for leading. .... why wouldn't
> leading be an inherent part of a type face like kerning?

That's an interesting proposition, but I think it belongs outside of a font because type designers don't (and shouldn't) have too much control over how their fonts are used.

hhp

> right or wrong. this brings up a good point. there are certain > guidelines and classically, or even better, historically proven > "rules" that designers use as the foundation for building something > that works. this being said, no one is going to break down your door > and haul you in for questioning if you don't follow said rules, but i > think it's also important to know what rules you're breaking. > personally i love to see educated designers challenging the standards > of our trade if it's working for them. there are a million things out > there that i wouldn't necessarily think of, or even want to do, but > that really do work and allow for an interesting, if not valid > solution. with leading, i think legibility is important in some > situations, but i've also seen circumstances where the content > allowed, almost demanded, decreased legibility. when a feeling or > personality created by an exaggeration in leading (one way or another) > adds an important layer to the content. for me, the only real rule is > to respect the content and my solutions have always been found there.

I think one probably could establish guidelines for particular fonts for particular usages ... but I wouldn't necessarily let the font designer do it because (a) of the point Hrant makes, (b) because I'm not sure even a talented font designer would necessarily get it right, and (c) because circumstances vary so much.

I don't actually agree that a letter is to a sentence as a sentence is to a paragraph etc. One reads words, not letters. And one reads groups of words, not sentences (and certaintly not paragraphs). They are quite different units, and the relationship of each unit to the next is actually different.

For me at least in many instances "feel" is low on the list. Of course we can all come up with inappropriate matches of type to subject-matter, and given the multiplicity of usable text faces, they can and should be avoided. But, when you think about it, why not set a book about Bauhaus in Bembo, if you think Bembo is a good readable typeface? It's not, after all, wholly inappropriate for those in the know (a piece of industrial-age type design from the early- mid-20th century which functioned excellently in its time). Which could not be said of a book set in Futura. The idea that a book design should somehow 'enact" its subject-matter seems to me to be crude. One need not set Virgil in rustic capitals printed on parchment with no word spaces. One should not. One need not set Dickens in some ghastly modern with enormous gaps between the words. Books about things are not to be confused with the things they are about.

For (most) readers books are things to be read. In so far as they are things, they are properly the subject of designers, and good book design which makes them the best things they can be is wonderful. But unless they function as things to be read, all the design in the world will not save them. That is what is so ghastly about some of, say Bruce Rogers', book designs: too much emphasis on the thing (the antique type, the fleurons, the bindings etc), not enough on the function. As a reader give me a nicely typeset penguin paperback any day over some handsome quarto volume artily tricked out.

Thanks very much for your replies. I appreciate the clarification. Personally, I've yet to decide where I stand, but you've given me plenty to think about.

cheers.

Certain typefaces need more leading than others to be legible. The measure of a line is also a factor. It's a judgment, rather than a rule book call. We've never had it so good as today when our computers allow us to set paragraphs until we get it right. Years ago those first proofs often came as somewhat of a surprise.

There is a new magazine Design Inside (Something like that). I find it unreadable the line measures are across the lenght of the page and the pull quotes are set with minus leading making acenders and decenders crash into one another. I only mention this because in the mid to late 60's and early 70s it was fashionable to kern type to unreadable tightness and use little or no leading. The opposite trend has been going on now for a while. The point being, as with fashion and color, kerning and leading will change.

The text block is only one element of a page design. The color and readability created by the leading can only be judged within the context of the overall design, its various elements and their relative weight.

My feeling is that leading and word spacing are somehow related, more so than leading and type size.

As the space between words grows, I always feel that I want more space between the lines. As lines get tighter, so, it seems to me, should the leading.

I suppose this makes sense. If I have very tight lines with a large leading, then the lines would start to look like railroad tracks, or bars across the page. If my lines were very loose, with tight leading, well then there would be terrible rivers.