A love letter to Cheltenham

There are typefaces I love for how they look and typefaces I love for what they did. Cheltenham is the second kind.

I cannot honestly tell you Cheltenham is beautiful. It is blunt. The serifs are heavy, the ascenders and descenders stubby, the letters sit on the page like they have somewhere to be. You do not look at Cheltenham and feel the pen-stroke romance of Garamond, or the generous ink-bloom of Mrs Eaves. You look at Cheltenham and see a newspaper.

That is more or less the point.

The architect

Cheltenham was drawn in 1896 by Bertram Grosvenor Goodhue, who was not a type designer. He was an architect, working at Cram, Goodhue and Ferguson in Boston, a firm mostly remembered for Gothic Revival churches. He was years away from the career that would produce the Los Angeles Central Library, St Thomas Church on Fifth Avenue, and the Nebraska State Capitol. The typeface was a sideline, a favour to Ingalls Kimball of the Cheltenham Press in New York, who had commissioned something readable for his private press work and gave the resulting design the press's name.

Goodhue died in 1924, at fifty-four, with a career's worth of buildings behind him and one typeface to his name. The buildings are celebrated. The typeface made him a footnote in every history of twentieth-century American printing.

There is something honest about that. Goodhue was not trying to reinvent the Venetian oldstyle or solve a problem about humanist stress. He was an architect drawing letters the way an architect draws anything: for structure, for durability, for how it will hold up when someone actually uses it. You can see the architect in the letters. They are built, not grown.

What Benton did next

The version of Cheltenham that actually went to work in the world was not Goodhue's. It was Morris Fuller Benton's.

Benton was the in-house type designer at American Type Founders, the consolidated foundry that formed in 1892, and he took Goodhue's original cuts and did something unprecedented with them. Between 1902 and 1913 he expanded Cheltenham into over twenty weights and widths: roman, italic, bold, bold italic, extra-bold, condensed, extended, oldstyle, wide, and on. An entire ecosystem built from one root design.

That was not how type worked in 1910. Most typefaces had one weight, maybe two. A printer who wanted emphasis set italic or moved to a different typeface entirely. The idea that a single voice could stretch across roman and bold and condensed and extended, and still sound like the same voice, did not really exist before Benton's Cheltenham.

This is the part that makes Cheltenham historically important in a way its looks never earn on their own. It is the ancestor of the type family. Univers, Helvetica, Frutiger, every sans-serif superfamily that came after, every modern variable font with a weight axis and a width axis, is descended from the commercial proposition Benton proved: one design, many voices, one brand. Before Cheltenham the idea did not have a shape. Benton gave it one.

The letters worth looking at

Cheltenham has short ascenders and descenders, which is unusual for a classical serif. Compare a lowercase h in Cheltenham to a lowercase h in Caslon and the Cheltenham h looks squashed. The ascender barely clears the cap height. The descender on the p and q stops short. This is not laziness. It is a deliberate trade: short extenders let you set Cheltenham on tighter leading, which means more text per column, which is exactly what a newspaper wants.

The x-height is small relative to the caps, but the letterforms themselves are wide and sturdy. The counters are open. The serifs are heavy and bracketed, which is to say they curve into the stem rather than meeting it at a hard angle, but the bracket is blunt rather than graceful. Compare a serif in Cheltenham to a serif in Baskerville. Baskerville glides. Cheltenham arrives.

Look at the lowercase g. It has a small closed loop for the bowl and a broader, almost chunky lower loop. The ear is short. Nothing about it is elegant. It is doing its job.

Look at the capital A. The apex is flat rather than pointed, with a slight inward cup. That is an architectural detail, the same way you might design a roof parapet to shed water. It gives the letter a sense of being built rather than drawn.

The terminals on letters like the lowercase c, e and r are rounded, not sharp. This is where the hand-drawn quality comes through. Goodhue was not a lettering specialist, and the letters have the slightly quirky feel of someone confidently freehanding a design rather than refining it through dozens of proofs. Benton tidied a lot of this for the ATF cuts, but the personality still shows.

The S

I look at every typeface's S. My name is Sammii and I cannot help it. Cheltenham's S is almost aggressively stable. The spine is nearly vertical, the curves at the top and bottom are close to symmetrical, and the whole letter sits on the baseline like it was dropped there from a crane. There is no diagonal stress, no calligraphic memory. It is an S that has been machined.

I would not call it graceful. I would call it trustworthy, which for a letter that does a lot of work in words like "news" and "story" turns out to be the right quality.

The newspaper

Cheltenham is the typeface of the front page. Not just any front page, though it sat on plenty in the early twentieth century, but specifically the front page of the New York Times from 2003 onwards.

Matthew Carter was commissioned to redraw Cheltenham as a custom cut for the paper, and the result, NYT Cheltenham, became the unified voice of the Times's headlines. Before then the paper used a mix of display types accumulated across decades. Carter's work sharpened the core character of the ATF Cheltenham for modern newsprint, and it has been the sound of the Times ever since. If you have seen a New York Times headline in the last twenty-odd years you have seen Cheltenham.

That alone puts it in a category very few typefaces reach. Most readers of the Times could not name it, but they would recognise a headline set in something else as wrong. Cheltenham has become invisible in the way that only the most thoroughly absorbed design decisions become invisible.

Before that it worked for the first half of the twentieth century as a general-purpose display and advertising face, running on magazine covers, book jackets, paperback spines, theatre programmes, patent medicine packaging. It is the typeface of American public information design in the period where American public information design was being invented.

The revivals

If you want Cheltenham today you have options, none of them perfect.

ITC Cheltenham, redrawn by Tony Stan in 1975, tidied Benton's proportions into something more refined and slightly more elegant than the original ATF cuts. It is the version most designers reach for if they want Cheltenham without the hand-drawn quirks. Some of the character is lost in the tidying; some of the awkwardness that makes Cheltenham Cheltenham gets filed smooth.

NYT Cheltenham, Matthew Carter's 2003 cut for the New York Times, is the most polished digital version in existence and is not commercially available. You can see it, you cannot use it. This is probably correct. A newspaper's voice should belong to the newspaper.

Cheltenham Pro and the various other digital issues from Adobe, Bitstream, and URW cover a usable middle ground. None of them will change your mind about Cheltenham. They will let you set it.

Why it still matters

Cheltenham is hard to love aesthetically and easy to respect historically, and I think that is a useful thing for a typeface to be. Not everything needs to be beautiful. Some things need to be sturdy, available in every weight you could possibly want, and legible at eight in the morning on newsprint that has been rolled, folded, and stuffed through a letter box.

Hate Cheltenham and you are hating a lot of what "newspaper" looks like in the cultural imagination. You are also hating the typeface that taught the industry what a family is, which means the grandparent of every design system that now lets a brand speak with one voice across twelve weights and five widths.

I do not use Cheltenham often. It is not a typeface that suits most of what I make. But every time I see a Times headline I notice it, and every time I notice it I think about an architect with a side project and an American type foundry that decided to make the side project into a system. That is a good story. Not every typeface has one.