Is Typography a science?

Gunarta's picture

I really confuse with my question, it made dirty my mind for a days, "Is typography a science?"

50/50. I think Typography is mixture of art and science. Because, Typography is art and another side is science. The science side is about how to make the text and letters can be more readable, legible, and proper with people who want to use it.

The characteristics of science is:

1. Knowledge.
I think typography have it.

2. Scientifically Ordered
I think typography doesn't have it yet.

3. Using the power of thought
Have it.

4. Objective
Sure. The object is type.

that's all..


Bendy's picture

Interesting, I was wondering about 'scientifically ordered' yesterday at Ampersand. Tim Brown was explaining his modular scaling system into which you can plug in your 'key' type size, and it returns a set of compatible sizes that click. (Try it out at (still under development).

It made me question whether principles of typography can be so easily translated into Maths. Music can also be translated into Maths, but IMO something is lost in translation, and the results are not useful for creating music.

Jean Paul Beumer's picture

Related book: While you're Reading by Professor of Typography (!) Gerard Unger.

Bert Vanderveen's picture

Typography is a craft.

eliason's picture



This horse has been pretty well beaten around here.

riccard0's picture

Thanks Craig, that was the thread I mostly meant to link to :-)

Gunarta's picture

i think typography may be a science someday. It has so many rules in its knowledge and making type. Now, typography is a part of design art yet.

If typography is science, which one it will be classified? Natural Science (the study of natural phenomena) and Social Science (study human behavior). I think it will included in Social Science. Because, the usage Typography is never separated with human's life. (while Wikipedia explain typography is an art and technique to arrange type). Typography may be included into formal science also, or applied science (the object is mainly engineering).

If typography is NOT a science. Why there is the academy discipline of it? Why there is professor of Typography? Why Typography is studied by people? I expect typography is a science soon. Looks! We only need to find the basic laws that universal.

Science is a systematic enterprise that builds and organizes knowledge in the form of testable explanations and predictions about the world. (Wikipedia)

Some characteristic of Science:
1. Knowledge
Typography is created by the ideas and curiosity of humankind. It can be as craft and engineering.

2. Scientifically ordered
To be a science, something needs a arranged knowledge. That arranged knowledge can be called as science. Typography has knowledge, about its history, legibility and readability, the role to human life. Even more, Typography has scope and experimental things. However, there is none arranging knowledge in typography yet.

3. Using power of thought
Knowledge comes from idea and curiosity of human. The idea and curiosity come from fact. Since Aldine has found, Akzidenz-Grotesk is labelled as wierd, Comic sans hated, I think Typography is really interesting. The people could judge and evaluate some type in majority way. It's a clue that it can be concluded into science.

4. Objective
The object is type and i said before, Typography has scope. And scope is preventing people to not step into outer boundaries. The Comic Sans case, the masters of type said the same thing. Comic Sans may not used as grave text, traffic text, ect. Some other said that Comic Sans have bad kerning. It's clue that Typography is science.

@bendy: What is modular scaling system? I can't understand the function.

ebensorkin's picture

Adien, you are thinking about very interesting things but to do this kind of subject justice you have to bite off bits at a time and really chew them. What you are doing in your reasoning is too broad and too unfocused. As a result it feels like the kind of argument I might hear at a party. It is fun enough in some ways - but it can't be taken seriously.

Also many people are spending a lot of time doing this work now and it is getting less and less hard to find out about it. I suggest that you do look into this work. Take your time, but do your research!

Gunarta's picture

do research is not my job i think. Professor does.

riccard0's picture

It’s students’ job too.

Bendy's picture

The modular scale is a system of ratios. You pick a key type size and the system generates a set of compatible sizes based on whichever ratio you pick. So in that slide, the key type size is 16pt. The modular scale shows two other sizes, 25.88 and 9.88 point, obtained by multiplying and dividing by the Golden Ratio, 1.618. Setting type at these sizes is said to be harmonious on the page/screen. For me there are so many variables (leading, line length, and of course the style of the font itself) that even with the modular scale you can still easily end up with a mess. A design eye is unavoidably crucial.

ebensorkin's picture

I don't mean you should do scientific research. I mean you could just read the information that is easily available on typophile and else where. In any event I don't mean to be insulting. I am just saying that your post covers too many areas at once, and covers ground that has been discussed quite a bit already. That is why many replies were links.

ebensorkin's picture

Bendy, was this reply of yours for another thread?

Bendy's picture

Gunarta had asked what the function of the modular scale was; I'd mentioned it above as one example of a 'scientific' approach to typography.

John Hudson's picture

Modern notions of science tend to focus on method rather than on 'arrangement of knowledge'. Knowledge can be arranged in any number of ways, many of them arbitrary, whimsical or willful. See, for example, the majority of typeface classification schemes, which certainly involve arrangement of knowledge but are anything but scientific. Science isn't something one has: it is something one does. So the question is not whether typography is or is not a science, but what it means to do science in the context of typography.

John Hudson's picture

PS. For a long time my desktop wallpaper consisted of a piece of calligraphy including the quote from Johann Jakob von Weingarten: Inter scientias non minima est typographica. Obviously that is using an older understanding of science.

Chris Dean's picture

@Gunarta: A bit of information to help you get started

Bordens & Abbott’s (2008) diagram of the scientific method

And if you want a nice start into researching the field of a scientific approach to the practice of typography, Kevin Larson has a nice piece of writing which isn’t that hard of a read:

Many people consider Miles Tinker to be one of the earlier contributors to this field. His work is not that hard to read either. You may need a bit of help with the statistics. I also (try to) keep a growing list of literature for my own records:

You will most likely need access to scientific journal databases for the majority of this content, but it has been my experience that if you contact the authour directly they are more than willing to share their work. And Google Scholar seems to be far more successful than it used to be.

Additional threads include:

And Craig (eliason) is correct. There is certainly no shortage of debate on this topic at Typophile. Chances are, the conversation you’re looking for had already happened several times. Good luck with your research.

Té Rowan's picture

"Typography: The Forgotten Link between Science, Art and Craft?" (Wonder if someone somewhere will defend a doctorate on this...)

Nick Shinn's picture

Sssh…let's keep it a secret.

William Berkson's picture

>let's keep it a secret.

Too late. My talk on the "Great Readability Scandal" could be subtitled that. Will you be at TypeCon this year, Nick?

Nick Shinn's picture

Fraid not.
Might go to Iceland, though.

Chris Dean's picture

“Perfect typography is more a science than an art.”

Tschichold, J. (1948). Clay in a Potters Hand, as cited from The Form of the Book. Hartley & Marks Publishers Inc. Washington, USA.

(1st sentence of 1st essay)

Gunarta's picture

OK, this is the fist Basic Law:

1. You can read letters and words.
2. Serifs make letters more readable (in good kerning)
3. Hand written font is not hand writing

(these is the thinking of student's brain like me. Imagine, if professor would think about it)

Gunarta's picture

I agreed with Dean, perfect Typography is more a science.

About Graphology, it's a science that study about human's HAND WRITING and its relationship with soul.

I think Typography is BIGGER than graphology. If Typography studied and ordered scientifically, people will not use Comic Sans for Grave anymore

Chris Dean's picture

2. Serifs make letters more readable (in good kerning)

Refutable. See:

DeLange, R. W., Esterhuizen, H. L. & Beatty, D. (1993). Performance differences between Times 
and Helvetica in a reading task. Electronic Publishing, 6(3), 241–248.

Patterson, D. G. & Tinker, M. A. (1932a). Studies of typographical factors influencing speed of reading: X. Style of type face. Journal of Applied Psychology, 16(6), 605–613.

Nick Shinn's picture

…Performance differences between Times 
and Helvetica…

Which Helvetica and which Times?

Typographers believe the difference in cuts is significant, hence the newly undead Haas Grotesk.

I doubt that science would ever be able to establish an objective measure of the reading-functional difference between the various cuts of Helvetica.

Typography, as practised by typographers, is beyond science.

William Berkson's picture

>beyond science

I think the aesthetic side of typography is not likely to be understood by science any time soon. The craft aspects involving readability I think are.

riccard0's picture

Science is overrated.

Nick Shinn's picture

So Bill, the selection of a particular cut of a typeface is "merely" aesthetic, and has no effect on readability?


In discussing Helvetica Neue, Ilene Strizver says, "Widened crossbars on the lowercase f and t increase character recognition in text."

Are you saying that the widened crossbars are aesthetic, not craft, because their effect on text readability cannot be scientifically measured?

IMO, Ilene has made an aesthetic judgement on what constitutes readability, but aren't such preferences part of the craft of typography, which you believe may be objectively measured?

William Berkson's picture

>So Bill, the selection of a particular cut of a typeface is "merely" aesthetic, and has no effect on readability?

No. Incidentally, Strizver doesn't use the term readability, but "character recognition".

>Are you saying that the widened crossbars are aesthetic, not craft, because their effect on text readability cannot be scientifically measured?

Not at all.

What can't be measured now, and may well be measured soon are two different things. And there are important things that may never be measured. What of readability can be measured and not is a work in progress in reading science.

You have insisted here on typophile that a science of readability, which would include measures of readability, is IMPOSSIBLE.

My TypeCon talk will show the science of readability already exists, and has begun to have interesting results, though a lot of progress is yet to be made.

Whether widened crossbars in themselves help readability I doubt, but they may in some contexts; it's not a settled issue.

Just to be clear, I also don't think that readability always trumps aesthetics. Quite often in display type maximum readability should be sacrificed to aesthetic goals. Also there are, of course, practical considerations other than ease of reading of continuous text involved in designing a page and a document—including hierarchy of ideas, ease of navigation for the reader, and more.

Chris Dean's picture

@Nick: I am trying to contact the first authour as we speak. In the meantime, if it helps:.

Reading material for the Standard Two group was taken from Blanckenberg and Ferreira Blanckenberg [14, pp. 47–53], a discontinued school reading book for Standard Two. A discontinued school reader for Standard Four by the same authors provided the reading material for the Standard Four group [15, pp. 73–77]. Times Roman 13 point was used for the roman typeface and Helvetica 12 point for the sans serif typeface in this experiment. Type size, weight, line and letter spacing of the two faces were matched closely by means of computer software. The heights of the lower case letter x of Helvetica and Times Roman were respectively 2.30 mm and 2.25 mm. The height of the Times Roman capitals was 3.3 mm, and those of Helvetica 3.0 mm. Line spacing for both typefaces was 5.15 mm. The text was set flush left with a maximum line length of 33.75 picas. The pre-test consisted of 536 words and 47 lines and the post-test of 549 words and 45 lines.

Nick Shinn's picture

What of readability can be measured and not is a work in progress in reading science … I also don't think that readability always trumps aesthetics.

You're marginalizing aesthetics.
(i.e. anything that can't be measured isn't readability but aesthetics.)


Christopher, I was only wondering which cut rhetorically, didn't really need to know!
IMO proprietary names of products shouldn't be used in serious research.

John Hudson's picture

Nick, I don't think Bill defined aesthetics as 'anything that can't be measured', and you had to delete a large amount of intervening text to stick those two statements side-by-side. In his final paragraph, Bill is talking about design decisions and how functional readability may be legitimately and knowingly reduced for aesthetic effect in some circumstances. He doesn't suggest anywhere that aesthetics is 'what's left over when you can't measure any more'; rather, he is quite reasonably pointing out that the visual beauty of a text is independent of its readability. That seems to me uncontroversial. Nor is it the case that the aesthetics of typography cannot be measured, but it must be measured in ways that differ from how we might measure readability, e.g. presenting different settings of the same text to people and recording which one they prefer to read, i.e. finding ways to quantify subjective responses to the visual appearance of text before reading.

dezcom's picture

There is far more than enough controversy about how accurate or meaningful readability tests are. If we try to measure aesthetics, we are really in for some crazy stuff ;-)

russellm's picture

> Is Typography a science?

yes and no.

Same answer applies the the question "Is Type design an art?".

dezcom's picture

A better question is, "Does it matter?" What would you do differently if it was than if it wasn't?

John Hudson's picture

Chris: What would you do differently if it was than if it wasn't?

I think the question of whether typography is or could be a science is less interesting than the question of the potential impact of science on practice. What about this question: if compelling scientific evidence suggested that readability would benefit from doing something that was contrary to the established canons of text typography, would you do it?

dezcom's picture

That is a far better question! and put better than mine. For me, the problem is "what constitutes scientific evidence"? Just because scientists have done a study and post their results does not mean that their conclusion is a direct correlation of what the test measured. My biggest concern is the quantum leap between measuring saccade count and inferring something more about it that what it shows.

If I were convinced that something that came out of a number of scientific studies countered traditional typographic practice, I surely would value that over tradition. That being said, I don't particularly follow standard typographic tradition anyway. Not that I have anything against it, it just does not interest me since there are thousands who already take that road and one more won't be of much value.

Gunarta's picture

Everything can be measured by Survey Methodology

William Berkson's picture

>anything that can't be measured isn't readability but aesthetics.

Nick, I am getting the feeling that you identify science with positivism—which includes such views as: if you can't measure it, it's meaningless or it's unimportant or it's purely subjective. And you think when I advocate the value of science I'm advocating positivism. But science is not positivism, and I'm not a positivist. And, as John said, I certainly don't believe the above idea that you seem to think I believe.

For starters, I think there is a lot about readability that we can't yet measure, but will be able to in the not too distant future.

I think the problem with most past legibility research is, to be blunt, that it has been poor science. If it had been good science, the results would have been much more compelling, and those working with type would have been more convinced of its value, and would be guided by its results. Most people dealing with type professionally haven't been much impressed by has come from scientific researchers on legibility, and I think their skepticism has been largely justified.

I wasn't intending to beat my drum again, but at TypeCon I will tell a lost story about readability, and some outstanding work on it. I think the whole story, which I only learned of in the past few months, will make a lot clearer what we have been going round and round about over the years here on Typophile, not very fruitfully. I don't mean that it resolves all the problems, but it does clarify a lot.

enne_son's picture

a lost story about readability, and some outstanding work on it

Bill has generously given me a bit of an inside track on this.

What he's found is highly relevant, historically regrettable, and explains a lot about where we are on these issues today.

Nick Shinn's picture

Bill, this has nothing to do with positivism or philosophy.

My posts have never been against science, I have used my knowledge of typography to debunk bad science, under which I include the idea that "readability" can be measured with regard to anything other than a specific document — certainly not to the individual constituent elements such as x-height, weight, leading, line length, and typeface choice, in any manner that will be prescriptive to typographers. A template is the kiss of death*.

Just because some scientists think they are conducting useful experiments doesn't mean that what they're doing is any better than phrenology or graphology.

I made a criticism of what you wrote in this post, in particular your definition of aesthetics and readability as two non-overlapping concepts, when you said, "I also don't think that readability always trumps aesthetics."

In my opinion, aesthetics, courtesy of the skill and good taste that a practising typographer has acquired and embedded in a document, is an important part of readability. When I decide on the width of a sidebearing by dragging it until it looks just right, this is an aesthetic decision which I believe contributes to the readability of text which may be set with the font.

When I choose a particular combination of typeface, size, line length, and page layout for a particular text, by playing around in InDesign or Quark until it looks just right on the monitor, and printing it out on a laser printer, and trying to imagine the final job offset on a particular paper stock, I am using my aesthetic judgement to set values which may subsequently be measured.

I don't think this issue is much different than the way people have been trying to create scientific advertising for over 100 years, with only limited success. If creating predictably readable (and actionable) documents has met with such limited success there, why should these efforts by any different? These new experiments don't exist in a vacuum, but have that baggage to deal with.

To clarify: reading research, good; readability research, bad.

*The most conventional, template-friendly area of typography is the book. So it's it's no surprise that this is the area (continuous reading of immersive text) where scientific studies are considered to have the most relevance. However, how is that engaged by readability studies that use, for instance, the RSVP method?!

William Berkson's picture

Nick, agree that readability and aesthetics aesthetics overlap, and never indicated otherwise. But they aren't the same, and sometimes, such as in a wedding invitation, or a monogram, an elaborate script that is not very readable may be just the thing aesthetically. And even in running text, throwing in italics, which are slightly less readable, serves the larger purpose of slowing down the reader, to indicate that this point is more important.

>reading research, good; readability research, bad.

Well, you're wrong, and I suspect your error is because you think of such research in the mold of some bad research that's been done. But as I keep saying and you keep not believing, there is more to science than that.

But I take this up later after I've discussed this all at TypeCon.

You remind me of the guy in The Princess Bride who keeps saying "Inconceivable!"

Nick Shinn's picture

But they aren't the same, and sometimes, such as in a wedding invitation, or a monogram, an elaborate script that is not very readable may be just the thing aesthetically.

I disagree. The fancy script for a wedding invitation is used because it is more readable in that context.

Well, you're wrong, and I suspect your error is because you think of such research in the mold of some bad research that's been done.

There you go again, with the false interpretations.
I'm right, because, as I keep saying, only the readability of documents can be measured—it's impossible to isolate the readability of individual typographic parameters and use such indices to predict the readability of future documents.

Even if one predicts the readability of a template, that will fail with diminishing returns, because people get bored, culture changes.

Don't you recall the thread where I showed some pages from a Paul Rand book, with the body text set in Univers Bold?

You remind me of the guy in The Princess Bride who keeps saying "Inconceivable!"

I remind me of the guy in Shakespeare in Love who keeps saying "It's a Mystery!"

William Berkson's picture

Nick, if the readability of documents can be measured, then there can be a science of readability, which you deny. You contradict yourself.

>it's impossible to isolate the readability of individual typographic parameters and use such indices to predict the readability of future documents.

Three factors involved in myopia are the shape of the eyeball, of the cornea, and the flexibility of the lens. If the eyeball is too long front to back, then one cannot focus at a distance as well. If the lens becomes too rigid, as in presbyopia, then also the lens cannot be adjusted as much for distance vs close work. And the shape of the cornea can cause myopia.

So by your reasoning, even though we can measure whether individual eyes suffer from myopia, it is IMPOSSIBLE to measure the separate factors of eyeball shape, cornea shape and the lens's ability to adjust, right? And to predict the total amount of myopia on the basis of those three factors, and a knowledge of their complex but determinate interaction, is IMPOSSIBLE. That's beyond the realm of science, IMPOSSIBLE, right?

Nick Shinn's picture

You contradict yourself.

Very well.
However, isn't it true that the "science" of readability, as practised, claims to be somewhat more than it is, and almost invariably draws deductions that pertain to more than the specific documents tested?

Your myopia analogy is so BAD I won't even begin to address it.

How do you explain the readability of the Paul Rand book set in Univers Bold?
Readability theory says that a bold sans serif is not suitable for the text of a book, such things should never be.
But the body text in question is not just "aesthetic", it is functionally readable.

Isn't that a scientifically valid debunking?—the example which disproves carrying more weight than any which may confirm.

William Berkson's picture

Nick, name-calling and haughty dismissal is a poor excuse for argument. If you can't answer my myopia analogy, admit it; otherwise show me what's wrong with it.

Té Rowan's picture

Oh, joy, f5g joy... they're talking east and west again...

John Hudson's picture

Nick: The fancy script for a wedding invitation is used because it is more readable in that context.

Is it also more legible in that context?

Syndicate content Syndicate content