by Dmitry Kirsanov
Source: Design Lab

The history of type design that we’ve just skimmed through is probably the only useful guide (besides your own experience and preferences, of course) in choosing fonts for “plain text,” one-face projects. In such cases your single concern should be about a good match between the style, semantics, and intended impact of your text and corresponding properties of the typeface it uses. You can’t set Shakespeare plays in a sans serif font (even of the “humanist” variety), and fragile Modern serifs are not appropriate for a pushy advertisement message.

However, the majority of design projects pose much more difficult problems: Most often, you have to find two or more different typefaces, each matching not only its corresponding text but also all other fonts in the composition. As I’ve once noted, creating solid contrast links between different fonts is a real challenge. Each font being nearly as complex as a human being, it is next to impossible to find two fonts whose features, so to say, point in opposite directions but along the same line—which is a necessary prerequisite for any contrast link.

Therefore, the safest option in any project is using the same font, or its different variants (e.g. bold or italic), for all of your needs. This “monofont” solution may not be perfect, but in many cases it may be better to ensure one more aspect of consistency instead of taking risky aesthetic decisions. Using one font is by far preferable in small compositions such as logos, where two or (God forbid) more fonts would have too little room to show their true nature, and any difference between them would therefore become an annoyance rather than a meaningful opposition.

But, let’s say you really need two fonts for your work (for example, one for headings and the other for body text), and you’re not satisfied by using variations of the same font. Which pair to choose? The basic answer is well known: serif fonts work fine with sans serifs, and vice versa. These two types of fonts are different enough for their contrast to be immediately obvious, resulting in a strong and stable connection.

Here again, the font history may give you valuable hints on which sans serif and serif faces work best together. As these two font types were not developed in parallel, we cannot base our decisions on historical synchronism; most, if not all, today’s serif faces are rooted in past ages, while the majority of sans serifs were created in the current century. More useful is matching the level of “humanization,” liberty and even looseness vs. that of rigidness and artificiality. Remembering what I’ve said about the general trends of dehumanization of serifs and humanization of sans serifs, we may conclude that early serif faces should be a good match for late sans serifs, and vice versa.

In fact, I’m not sure if it is possible to successfully combine late Modern serifs with early geometric sans serifs—both these designs are dehumanized to the point of being dry and even awkward, but their underlying ideologies are too different and even disparate. However, for higher levels of humanization this recipe works perfectly; just consider that the basic pair of typefaces for all operating systems are Times and Helvetica, which are a good match because they’re both transitional—that is, not too “zealous” but not too liberal either.

For another example, look at the logo of a Russian magazine called Internet. [Link no longer in service. -Ed] Although, as I said, different fonts should be avoided in logos, here the two humanist faces, Garamond and Frutiger, work perfectly together because there’s no visual in the composition and the fonts are left to interact only with each other, and because the contrast between them is so deep and multiaspect (in font, face, color, and slant). Moreover, the same two fonts are extensively used throughout the magazine itself (more consistently in the print version than on the site, due to the obvious HTML restrictions).

The AlphaWorks site offers a similar yet contradicting example. As most other IBM sites, this one was developed by Studio Archetype; however, the decision to make a logo from the humanist Meta sans serif font (particularly favored by this studio’s designers) and an old-fashioned Modern serif font (which is traditional for IBM’s corporate style) resulted in a composition of questionable merits—exactly because of the clashing levels of humanization in these two fonts. It is difficult to suppress the feeling that the contrast in this logo is either insufficient or irrelevant.

As you may conclude from these examples, it is usually a good idea to accompany the contrast of fonts by the contrast of other typesetting aspects, particularly font variations (bold, italic, etc.). These variations are possible along the axes of weight, slant, and width—but you should understand that these axes are not equally applicable to different font types.

You may have noticed that many sans serif fonts do not have italics in the proper sense, but only slanted (sometimes called oblique) variations. This is not by accident; the very nature of sans serif design doesn’t allow for an easy transformation into an italic face (or at least, into what we’re accustomed to regard as such). Contrary to that, most serif fonts have really dinky italics that, despite having very different lettershapes, are well dovetailed with the roman variety.

On the other hand, demibold and especially bold variations do not work particularly well with many serif faces—their serifs seem to obstruct the change in weight, and the relative difference in strokes’ thickness (i.e. the level of contrast) is difficult to preserve in different weights. (Interestingly, a high level of contrast is less of an obstacle for changing weight—hence the extra-bold Fat Face and its likes, built upon Modern serif fonts with their increased contrast.) Sans serifs, on the contrary, are very easygoing when it comes to changing weight—many of them have a wide range of weight variations, from extra light to extra bold. This means that each font has its own most natural variations, and you must take these into account when working with fonts; for example, in a professional work you’re much more likely to see a combination of a bold sans serif font and an italic serif font than vice versa.

What are other ways to increase contrast in a multi-font composition? Sure, a serif/sans serif pair is not the most conflicting of all possible font combinations. You could take, for example, an enticing decorative font, such as a stylized blackletter or script face, and put it alongside a plain sans serif font. Unfortunately, such extreme combinations are rarely satisfying. When fonts are too different, they can’t live happily together; as we’ve already found out, contrast is the strongest between two objects when their opposite features are many, but some common traits are still there.

You certainly cannot use a peculiar decorative font for all text on the page, if only because such fonts are often barely legible. If you’ve used a fancy font for a heading or logo, try to be as discreet as possible in your other font choices. As I’ve said, stylistically sans serif is not the best match for decorative faces; however, it is often better to environ a complex-shaped fancy heading with a plain, graphically poor sans serif text blocks than with a serif font which usually has a much stronger personality. It is also advisable to use quite different sizes and colors for the two fonts, in order to move their contrast from the dimension of lettershapes to the more flexible opposition of sizes, colors, and visibility.

It is worth noting that professional designers rarely use something besides the simple, traditional typefaces. I intentionally restricted my historical sketch to the mainstream type design, because I believe that for a beginner designer it’s much more important to carefully study a small set of classical typefaces than to indulge in “cool” designs, often of questionable quality. The well-known fonts are almost “transparent” for perception not because they’re primitive, but on the contrary, because their complexity is well balanced, because they’re carefully polished by their creators and by the centuries of use. It is this transparency that allows the immaterial qualities of your text and design show through the material envelope of the font.