Typophile

The perception of text (rhythm, texture and color)

Martin Silvertant

I've recently been doing a lot more research and personal studies in type. I'm thinking of doing an elaborate study on the perception of text, where I set a page in a certain typeface and analyze why the text is being perceived in a certain way. For example, in Maiola has a rather rich texture so if I were to analyze this typeface, it would be predominantly a study in which specific elements cause the texture to increase.

I would also be looking at why specific letters in a typeface are designed a certain way. J. Harris did experimental studies and found that serifs on the lowercase counters appear to reduce visibility, so in the typeface Spencer these elements are removed. Letters like h/m/n are missing the inner serif on the right side and features the suggestion of a serif on the left side and the /b features no spur. What are the benefits and what are the consequences of such decisions?

This would probably be an ongoing study until I feel I've reached a point where the amount of analysis done results in meaningful insights. I could start with a random typeface but I thought it might be better to pick one typeface or perhaps one typeface from sans and serif to use as a standard which I can compare any other typeface I analyze to. I'm aware there's no neutral or standard typeface but which typeface I select is in some ways arbitrary. What I'm looking for is a typeface that doesn't do a lot of strange things, makes use of elements as we probably most expect it and has more or less standardized proportions. I think for the analysis it will be better if I select a typeface with a very consistent rhythm as well. I wouldn't want to select a humanist typeface because it takes too much time to explain the dimensions of a random typeface compared to the humanist typeface. Or quite possibly a humanist typeface is exactly what I need considering it's the first emerging style of typefaces and could therefor more or less be considered a standard. Frank E. Blokland makes an interesting case for the idea that the initial dimensions have been standardized during this period. On the other hand, the humanist model doesn't have the most even color so it probably fails my requirements in that regard.

So my question was going to be whether you can recommend one or a few typefaces that could be used as a standard in my studies, but the issue is a bit more complicated. So any help in considerations in how to go about this are very much appreciated.

Nick Shinn

Look no further.

John Savard

I find the reply Delphic in its brevity.

Had the original post come from a new poster, rather than someone who is a long time poster here and a typographic professional, though, I would have tended to give this kind of advice:

Start with the basics. Overall texture might tend to be more uniform in two cases -

A typeface with a large x-height;

A typeface in a light weight - that is, one with thin strokes.

In both cases, the ascenders and descenders are less prominent, although in the second case, that isn't really true relative to the part within the x-height, present for every letter.

Only after the simple and obvious is dealt with can one look at non-obvious impacts on overall texture by the characteristics of specific typefaces. Of course, there's lots of room for other things to have an effect; how heavy the serifs are; the way letters like "e" and "c" are shaped that are open in part within the x-height region, and so on.

But an approach that looks at one typeface at a time, and that subjectively, seems to risk getting lost.

David Vereschagin

Mr. Shinn may have been instructing Mr. Silvertant to look at what was in front of his face when he wrote his post. (Or, perhaps, to check out his own œuvre.)

Nick Shinn

Right, I was referring to that ubiquitous transitional font, Georgia.
At least, that’s what I get on Typophile.

Martin Silvertant

Nick, that's a rather disappointing answer. As far as I'm aware Georgia was designed for clarity on computer monitors and not to provide the most even rhythm and both most generic and most consistent features. It might equate to the same in some regards, but I'm also skeptical about Georgia still being my best choice after 20 years.

I must also say, perhaps it's just me but I didn't even consider Georgia for a moment because I sporadically try to set my essays in Georgia, thinking it provides clarity and comfort but I'm always disappointed. I keep trying but I've never used Georgia successfully. It just provides a very wrong texture which I find uncomfortable to read.

A typeface with a large x-height;

My concern is that the large x-height is a trend of recent years while I'm looking for a standard sans and serif typeface to compare all typefaces with. I think it's actually impossible to find a typeface with my requirements. I think I should define standard typefaces for each bigger subclass. Perhaps define a typeface for old style and one for transitionals (ignoring Didones for a moment). I could also define a typeface for Venetian, Garalde and transitionals, making it possible to go into more detail about the distinctions between different typefaces within the same subclass. The problem I'm foreseeing with this is that a lot of modern typefaces can't be clearly categorized. I've seen interpretations of Jenson that are so far off I wouldn't even call them humanist anymore.

And this is another reason why Georgia is perhaps not the best choice or not the only choice. Like I said, there are many considerations to be made here. I have my own ideas on how to go about this but I would like some validation and other people's opinions. I will get more insight into the issue by simply doing experimental research but I thought I would first ask you guys as I like to start with theoretical research first.

Had the original post come from a new poster, rather than someone who is a long time poster here and a typographic professional, though, I would have tended to give this kind of advice:

Do you feel I shouldn't have had to ask what I asked? I think my question isn't quite as simple as you make it sound by the kind of responses I'm getting.

A typeface in a light weight - that is, one with thin strokes.

Why do you feel that is important? A concern here is that not all typefaces I would be testing have a light range. Gilman, Maiola and Premiéra for example only feature the regular and bold weights.

William Berkson

It is not clear to me from your description what you are trying to learn from studying the type. What question do you have? What problem are you trying to solve?

Nick Shinn

Martin, are you dismissing Georgia because it is a “screen” font?
If you want to cover both print and screen, combo it with Miller, which I assume to be its print counterpart.

I recommended Georgia because it is transitional, i.e. not favoring old style or modern, and because of its huge distribution as a Microsoft Core TrueType font.

John Savard

I mentioned a typeface in a light weight simply because then the ascenders and descenders would be less visible, and that would contribute to uniformity.

Martin Silvertant

It is not clear to me from your description what you are trying to learn from studying the type. What question do you have? What problem are you trying to solve?

I want to study how the shapes of letters but predominantly the details affect the rhythm, texture and color of text.

I've done some experiments in extreme texture in the past but that's just one aspect of type. Right now I'm doing a study into typography in a more general sense but I also look up the extremes in weight, texture and color. What I want to do now is start a study that is an extension on that one but looks in a more particular area, rather than looking into typography as a whole.

Martin, are you dismissing Georgia because it is a “screen” font?

I haven't dismissed anything just yet. If you can make a point on how Georgia is the typeface I should pick, I'm willing to consider it. What I am dismissing is the notion that Georgia is the best option because it was marketed for its clarity on screen 20 years ago. I'm not at all convinced in those 20 years a better typeface for my current requirements hasn't presented itself.

I recommended Georgia because it is transitional, i.e. not favoring old style or modern, and because of its huge distribution as a Microsoft Core TrueType font.

Actually last week I was considering perhaps it's not about finding the typeface that is most standard in my requirements but about selecting a typeface that meets my requirements in some ways but is also prolific through its use. Why would I compare typefaces to Kristal (just a random example) if no one has ever heard of it? Perhaps I should stick with typefaces like Georgia because they've become a cultural standard.

If that is the way to go, what sans would you recommend? Myriad has become a standard, but it might be too particular. Franklin Gothic is also too particular. Perhaps Akzidenz Grotesk?

I mentioned a typeface in a light weight simply because then the ascenders and descenders would be less visible, and that would contribute to uniformity.

How would that work? Aren't the core letter shapes going to be lighter in proportion to the ascenders and descenders? Shorter ascenders probably contribute to uniformity.

Captured from web.archive.org on 19 Apr 2015. view this capture