Is it all getting a bit samey?

Nick Job's picture

Just been reading the Avenir & Gotham debate initially posted by David Rault and it got me thinking...

In both commercial aircraft design (and car design to an extent), there seems to be a form to which major manufacturers defer to in terms of 'what is most efficient'. As I understand it, a certain deviation from this 'ideal', the stylised tailfin for Boeing's Dreamliner, had the undesired effect of making the plane less efficient. As time has gone on, all commercial planes have begun to look increasingly alike (not helped by the number of manufacturers steadily having being whittled down). In an attempt to look different, the Dreamliners's fuel-efficiency has taken a hit...

If we assume that the market one is trying to hit is sans serif fonts for immersive reading (for which someone incidentally said Gotham was not as efficient as Avenir), is it fair to say that there is an inevitability as time goes on that the best sans serif typefaces designed for large areas of copy will, like the planes at the airport, begin to look more or less the same and have less and less distinguishing characteristics, and that any deviation from the optimum design will result in a move away from efficiency/readablility?

What in your opinion is the current best performing sans serif for immersive reading? Is this the optimum sans serif for immersive reading? If not, what is wrong with it?

Nick Job's picture

I realise that commercial aircraft and typefaces are different verticals. Apart from that, I may not know what I'm talking about so please put me straight.

William Berkson's picture

A lot of people, including me, think that no sans is optimal for extended text--an excellent serif is better for extended reading than any sans. I think sans have inherent problems connected with spacing and leading which means they lose out to serifs for this use, though not for other uses--display, short text.

That being said, I think the demands of functionality for extended text do push toward a fairly narrow range on weight and width of characters--both for serifs and sans. That still leaves room for stylistic variation, but within a limited range. Erik Spiekermann said something like: a good typeface can only vary by 5% from other good typefaces. I think he was talking about types suitable for text there.

I think that narrower sans, based on the oval rather than the circle, are inherently more readable in text. So Meta is more functional than Helvetica for text. PDFs of Gerard Unger's Vesta convinced me that it is one of the best sans for longer texts. He doesn't have up the ones I saw earlier, but this will give you a taste.

I'm sure flared sans like Carl Crossgrove's Beorcana or will perform very well also, but they are 'cheating' because they have a bit of serif in them. Similarly Angie sans and Optima will do well, but also 'cheat'.

brett jordan's picture

i know this isn't technically about reading extended text, but I recently did a test of a number of different typefaces with the congregation of my local church congregation on overhead transparencies with worship song lyrics... i used a number of typefaces, including georgia, minion, myriad, officina (sans and serif, various weights) and helvetica round... in a number of different letter spacings

by far the favourite typeface was officina sans bold, set slightly loose

people especially liked the fact that the italic was easily differentiated and still legible

me, i preferred myriad, but that could well be down to the fact that i use it a lot

the only typeface that was universally disliked was georgia... they thought it was 'old-fashioned'

crossgrove's picture

Your question is complicated; you make the comparison between airplanes and typefaces, which is pretty apt. One good metaphor for type is shoes; are they stylish yet uncomfortable, or are they sensible and trustworthy for long hikes? One gives a visual effect, one will carry you long distances.

You use a term that might not be what aviation or type designers would use. "Effficient" implies that the object does the most work, fastest or with least effort. The issue with real immersive reading, (which is not what you are doing looking at a coffee-table book or magazine) is comfort, ease, smoothness and transparency of form; you want to take in the author's ideas without being distracted or hindered by noticing the type, and you want to be free to connect with the ideas or story for possibly hours at a time. A "page-turner" will keep you fascinated right to the end, and if set in a comfortable typeface, you won't ever notice how the page looks, and you won't ever be bothered by a certain letter or by faults in spacing or other layout factors.

All that can be said for many typefaces, but I think it's agreed that niether Gotham nor Avenir is really ideal. They are both very reduced, abstract, monolinear, and modular. They are both expertly-crafted, high-quality, robust typefaces, but for the narrow purpose of long-distance reading, they are not ideal. William has identified one characteristic of good book typefaces that I think is important: Proportion. Narrower sans designs probably feel more comfortable to read because the heavier horizontals of a sans need more vertical space in order to keep from clogging the letterforms. A larger x-height will also help a sans design in this respect.

Here we start to notice different classes of sans type; the geometrics like Futura, Avenir, Gotham and Avant Garde have these reduced, directionless traits. Humanist sans designs like Syntax, Legacy Sans, Mundo, Fresco, Bliss, and Absara all are based on Renaissance roman types, which are recognized as the ideal book typefaces: Bembo, Minion, Dante, Garamond, Galliard, Kingfisher, Laurentian. So they have enough DNA that makes them more comfortable, with similar proportions, directionality, and humanism (hence that name).

You can go beyond this to another variety of sans designs that William mentions: flared, or semi-, or hybrid, or stressed sans designs (awkward, huh?). These sort of inhabit a gray area between sans and serif because they don't have monoline or simplified strokes, they adopt more of the traits of serifed roman types, and they give a very different impression in text than even humanist sans designs. However, they still (usually) don't have serifs. How to classify them?

There is a perception that sans designs can never perform in text as well as serif designs, solely based on the lack of serifs. There hasn't been a really comprehensive, thorough and informed study to tell us once and for all what typefaces really are more comfortable for immersive reading; we'd all like to know. So far, comparisons have been insufficient, inappropriate, or otherwise far from comprehensive. Meanwhile we all give our opinions and rationalize them afterwards.

My personal opinion, borne out somewhat in the design of Beorcana, is that serifs alone aren't as crucial to reading comfort as they have been thought to be. Generally people consider the lack of serifs a deal-killer; like a gentlemen's club; you have to have the appendage to be admitted. ;)

I think this is overly simplistic, and gives insufficient weight to the other traits, proportions and design features of a typeface. Classic book faces have a number of traits that make them comfortable, and I think all of those traits contribute crucial advantage in reading comfort: spacing, stroke contrast, cap/lowercase proportion, serif thickness, serif bracketing, stress, width, weight or color (how dark is the type). Then of course, apart from the type design, there is line spacing, line length, type size, paper and ink color and other layout factors. There are typefaces with serifs that do not have enough of these other traits, and they are very uncomfortable and difficult to read for long. Certain digital cuts of Bodoni are essentially failures as book types, even though the original designs by Bodoni were intended for text. It isn't the serifs that confer readability, it's a whole group of parameters of the typeface's design. What makes a person beautiful? Is it blue eyes alone? Is everyone with brown eyes ugly? Or, more aptly, is a rubber sole on a shoe enough to make it comfortable for long hikes? Or does there also have to be arch support, ventilation, flexibility, soft fabric or leather, durable stitching, and doesn't the shoe have to be the right size and width? Is the toe pinching? Does the heel rub? After about 4 miles, these factors clearly are of equal importance to what the sole is made out of.

Since it is this complex, interdependent constellation of factors that influence how readable a typeface is in text, I think the influence of serifs on overall reading comfort is widely exaggerated.

But one difficulty in this discussion has always been that some people will lump every sans design together as though they belong in a category together: Avant Garde, Versa Sans and Beorcana, because they lack actual serifs, are somehow considered interchangeable. It might be more useful to discuss which type is good for books without the misleading classifications of serif and sans.

As far as "samey", William has hit it; the demands of function on text faces are so many, and so specific, that text faces have been homogenous for quite some time already.

Trends in advertising and graphic design tend to draw out similar designs; there was a burst of geometric sans designs all at the same time in the earlier 20th century, and it was driven to a large extent by fashion. It's the style of now, the latest thing. The trend, or pervasive qualities I see in sans designs now are what I call the Info sans, which Erik Spiekermann has almost singlehandedly launched; slightly condensed, somewhat humanist, monospaced-feeling, sensible sans designs without a lot of character, possibly somewhat square. See this thread for a rundown (not all relevant faces are even shown).

Nick Shinn's picture

This has nothing to do with readability, and everything to do with corporate concentration.

Scalfin's picture

While corporate liners are somewhat similar, the Airbus and Dreamliner are nothing alike in appearance, aren't they?

David Rault's picture

I unfortunately have no time at the moment to discuss this topic with the attention it deserves, but I just want to say that for me, there is no such thing as immersive reading with a sans. a flare, maybe; I experienced some pretty good reading using cronos or optima, but it doesnt come close to the comfort of baskerville or garamond. someone said that serifs are not that crucial to immersive reading as it was tought to be: try to read a 300 pages novel composed in 10 pt. gotham or ff meta, then. good luck with the headache.

dr

Nick Job's picture

Thanks for your comments, esp Bill and Carl.

>While corporate liners are somewhat similar, the Airbus and Dreamliner are nothing alike in appearance, aren’t they?

In order to move away from the 'standard' in airplane design (particularly in the tailfin area), the resulting Boeing 787 Dreamliner design was less 'eco-friendly', as I understand it. Can the same be said for typefaces (particularly sans) for immersive reading? i.e if one makes a typeface more distinctive will it not be as good at its job? Is there a ceiling which we have just about reached as to just how readable a typeface can be?

When was the last major breakthrough with typeface readability? If it was in the fifteenth or even late nineteenth century, then shouldn't we all give this self-indulgence up as a bad job? After all, what real value is there in expressing oneself if our end product is no better? (I don't really think that btw, or at least I don't think I do! I'm just trying to find out if anyone really believes in what we/they are doing?)

To elaborate a little on my original post up top, is there no point in trying to make a more readable sans serif when serif faces are better at the job? Is that defeatist or am I simply being a realist?

Or, and this for me is the ultimate typographic question, can the current ceiling for (readable) sans serifs be blown away*? i.e. am I wasting my time trying to design a sans serif that reads better than what's gone before? Am I kidding myself that this is possible for anyone, never mind me?

>try to read a 300 pages novel composed in 10 pt. gotham or ff meta, then. good luck with the headache.

Granted, but have you tried to read a novel in Beorcana? Or is Beorcana just Carl's interpretation or opinion of sans serif readability or is there an objective yardstick of perfection that he has moved closer to?

* I think Beorcana hit that ceiling pretty hard if it didn't actually blow it away. I like it a lot.

Nick Shinn's picture

When was the last major breakthrough with typeface readability?

Yesterday.
The world changes every day. Readability is a moving target.

**

Dan, the ClearType faces have a consistent look for several reasons. As I see it:

Firstly, there is the techno groove, something about perpendiculars rendering better than diagonals and curves; verticals better than horizontals.

Secondly, it was a corporate design program, so a degree of homogeneity represents brand, and MS has progressed and become more sophisticated in that respect, when compared with some of its more ad hoc type releases of the past. So this time, stronger directives to the hired hands?

Thirdly, I believe there were some pressures to consistency of glyph shape from the need to conceive of the three major scripts as a whole. Whether this was an overt direction, or just a general mindset, as an outsider I can't say--but I suspect that Jelle Bosma's Cyrillic was found wanting because it didn't toe the party line.

Fourthly, there may have been some orthodoxy caused by readability theory, whether traditional typographic theories, or as a result of tests conducted during the type development.

russellm's picture

The cleverest thing said on this topic was by Eric Gill–something like: Readability amounts to what one is used to reading.

The second cleverest was by Mr. Shinn: Readability is a moving target.

-=®=-

crossgrove's picture

Those two comments are worth emphasizing; we wouldn't think roman type at 10 point was highly readable if printing and typefounding hadn't allowed the design of type to evolve to this point. Earlier, hand-cut typefaces were considerably less precisely refined and cast, and that looked normal at that point. Along the way, we learned to read best what we were reading most. I think this evolution is still happening, and we continue to adapt what we are comfortable with. Now, that is influenced by reading on computer screens, reading newspapers, reading various-length texts set in sans designs, and simply not reading as much (debatable).

Blackletters were considered the most readable typefaces in a culture that had essentially adapted to them. We also read latin types well because that is the script we learned to read in. Kanji, Arabic and Malayalam aren't harder to read, if you're a native reader. Clearly, seen overall, humans are widely adaptable in their ability to learn to read different script systems, and even within those systems, we adapt.

John Hudson's picture

Clearly, seen overall, humans are widely adaptable in their ability to learn to read different script systems, and even within those systems, we adapt.

We do, yes, but that doesn't preclude the possibility that certain shapes, certain sizes, certain spacing, etc. are more easily read than other shapes, sizes, spacing, etc. Adaptation is bounded in various ways by legibility, ease of distinction between potentially similar forms, i.e. features that underly readability. It is also worth remembering that writing systems were developed in order to be read, which means that they are themselves adapted to our ability to adapt; this makes the whole relationship of readability and what I call readerability complicated and not easily disentangled (presuming one want to disentangle them, which I don't particularly). On the other hand, there are examples of specific styles of writing that were deliberately devised to be difficult to read, either because readability was sacrificed to an overall aesthetic appearance (e.g. Russian v'yaz), or because the scribes found the game of making a puzzle of language irresistable (e.g. Ottoman diwani). Such styles demonstrate that there is more to readability than mere familiarity, and that readerability adapts to certain features better than to others. Those features include consistency of form and distinguishability of individual letters, which is why those features are suppressed in v'yaz and diwani.

Gus Winterbottom's picture

Some optimum shapes are imposed by physical laws -- aerodynamics and hydrodynamics, for example. On aircraft, the development of delta and swept wings was mostly driven by what was needed to fly at supersonic speeds (of course there are other reasons, such as build simplicity and fuel capacity). Modern submarines look significantly different from WWII subs because their designers studied the shapes of sharks and tuna and other "efficient" underwater swimmers.

Whether anybody has yet defined, in terms and equations that scientists would accept, the physical laws of readability or legibility is outside my BOK. In other words, what would Richard Feynman have thought about relating boumas and saccades and whatnot to readability and legibility?

Scalfin's picture

In planes, wing shape varies quite greatly, with the WWII-shape still being the most efficient at subsonic speeds, while a full delta seems to be the best at the highest speeds, if memory serves.

The dreamliner's whole body looks different from the airbus. The dreamliner is built like a sailfish, while the airbus looks like a whale w/ an oversized forehead. There is also a fair difference between the wings.

William Berkson's picture

>Whether anybody has yet defined, in terms and equations that scientists would accept, the physical laws of readability or legibility

There are not yet a lot of important results in this area, so far as I know.

roman's picture

Nick, do you have a source on the lowered fuel-effiency on the Boeing due to the wing design?

According to Boeing the plane is 20% more fuel-efficient compared to other jets of similar size flying today. And obviously there was a lot of effort in making a lighter plane out of carbon fiber and titaniun ... so slapping on some stylish wings doesn't seem to fit with one of the main selling points, which is that it's a more efficient plane compared to its only competitor. As an airline executive I'd probably thinking of the millions of dollars lost with that .2% decrease in fuel efficiency or whatever, and not on subtle styling cues as with cars or a typeface.

So far I've found this page which says that "the 787-3 will use winglets to decrease the wingspan and improve efficiency over short distances" while the 787-8 and 9 will use raked wingtips which are optimal for long-range flights.

Nick Job's picture

Roman, take a look at this article entitled Shark fins and bullet trains.

. Having read around a little I now realise that the final 'locked-down' design of the 787 Dreamliner is very dull compared to original concept drawings and that they aren't going to use the stylised sharkfin tail that I make mention of in two of my above posts.

Sorry if I did not make myself clear. First and foremost, I meant the tailfin which was sharklike to begin with, but as the blog above suggests, both the sharkfin tail and the bullet-train nose (as seen in early artists' impressions of the Dreamliner) had to revert for the sake of efficiency. The guy blogging rather disappointingly says, "Now it looks just a like the old plane it replaces." That is exactly what I'm talking about; a radical new design having to be abandoned because it doesn't do the job as well as the current designs.

So my question is; have typefaces (particularly Latin typefaces) to a lesser or greater extent already reached this equilibrium or ceiling, where any change in design, no matter how well-intentioned/executed, actually makes matters worse so you go back to a safer, more robust and proven design?

Nick Shinn's picture

This is not a good analogy, because the air is a natural and relatively unchanging environment, whereas type exists in a cultural environment that is constantly morphing.

In fact, type is an active participant in the cultural ecology.

crossgrove's picture

True; while John Hudson has a point that there probably are limits to this dictated by our physiology and psychology, we do adapt and evolve, and the products of our hands (type, books, computers) also evolve. It isn't a constant like air pressure or friction.

It's not established, or even much discussed, whether we have reached the kind of ceiling you mention. It's also hard to imagine that happening, given the changes we've already seen, just in the 20th century.

Nick Job, you can choose how radical or safe your designs are. Everyone doesn't work the same way, and everyone doesn't focus on the same aspects of type. If you think up a rad idea, just try to make it work. The proof is in the results. There ought to be numerous ways that the idea of the text face can be stretched, evolved, or exploded.

Nick Job's picture

>It’s also hard to imagine that happening, given the changes we’ve already seen, just in the 20th century.

I hope you're right. Nevertheless, in recent times a lot of people have done revivals, as if there's nowhere else to go. There may be other reasons for reviving type but, let's face it, mould-breaking originality isn't one of them.

It also occurs to me that the glyphs that come in for most criticism on the Critique section of Typy are nearly always those that 'jar' with the reader, usually because they have departed from the 'normal'/received shape of the glyph in question. I think that the normal shape of letters may be a far more constant 'pressure' for uniformity than Nick S will admit.

>There ought to be numerous ways that the idea of the text face can be stretched, evolved, or exploded.

Stretched or evolved maybe, but do you really believe the idea of the text face can be exploded? That said, Beorcana was some gauntlet to throw down.

>This is not a good analogy.

I realise that. I'm just not very good at thinking of good analogies. It's the best I could come up with to try to illustrate my original question.

Nick Shinn's picture

the normal shape of letters may be a far more constant ’pressure’ for uniformity than Nick S will admit.

Damn right. The pressure for uniformity comes not from the needs or wants of the reading public (surely a diverse bunch!) but from:

-typography professionals who would like to see things done "properly"
-technology which standardizes for the sake of efficiency
-corporations whose products dominate the marketplace
-accessibility standards
-advertisers and publishers who... &c. &c.

When people get the opportunity to break the rules (handwriting, tattooing, email, texting), all hell breaks loose.

I’m just not very good at thinking of good analogies.

Perhaps I shouldn't have said it was a bad analogy, because it was a good analogy, as analogies go. But analogies break down pretty quickly when it comes to type and reading, because the issue is so complex and there's nothing quite like it.

William Berkson's picture

>comes not from the needs or wants of the reading public

I don't agree. Familiarity with letter shapes is a key to readability. If you make unrecognizable characters it will be very annoying to readers of extended text. So there is an inherent conservatism to type. If you want to be a radical, be a scientist or inventor; that's where radical innovation is highly prized.

One example of the inherent conservatism of type is the non-phonetic spelling of English. It is totally irrational, and a barrier to learning reading, but it sticks around because people are familiar with it. Similarly the keyboard layout, which was designed to *slow* typists down, is still used generally, though it is demonstrably sub-optimal. I think that the roman characters are pretty functional, unlike English spelling and the keyboard, but it just goes to show how conservative these things are.

This doesn't mean that there's not scope for stylistic innovation; this is, but it's just more limited than say in painting in fine arts.

In display type you can of get away with a lot more, of course, depending on the purpose.

crossgrove's picture

"Stretched or evolved maybe, but do you really believe the idea of the text face can be exploded?"

Come on now. Don't we see this every day in coffee-table books? Ha ha. Really that has more to do with people not understanding how text faces work.

You notice that there are a lot of people doing revivals. True. I don't think this is new. Every technology brings this, and some old types really are worth keeping around. Some aren't as well-known or appreciated as they probably should be. Also keep in mind how different type designers have different interests. There are incredibly strong designers who, on examination, stay well within a safe zone of historical models. Other designers experiment more freely and stretch the boundaries, rather than stay within them. Some designers approach type design from the outside and are less inclined to follow conventions.

Here's the thing: Dwiggins made some very successful, but somewhat bizarre experiments with his types intended for harsh printing conditions. Matthew Carter used a technique in Bell Centennial that we now see as extreme, but for a particular purpose. Do you think of those as radical? Compared to say, the "madly daring" extreme contrast of Didot and Bodoni, I think they are pretty out there, but always in the service of making text easy to read. Our reading conditions are now very different. We find ourselves reading onscreen more than we might like to admit. What might be done using the current technology and settings that would improve things, but possibly be kind of un-subtle, disturbing, unprecedented, deconstructive, radical? I think there's plenty of area there to explore.

Nick Job's picture

Great links, Carl, thanks. I admire you for trying to bring out the radical in all of us. I'm pretty confident that in my case it's not there! And I remain fairly convinced that it has less place in type design now than the relatively microscopic evolution of letterforms over a period of time. Still, crazy blended wingtips on Boeing 737s came from nowhere and seem to do a job so don't let me be in any way a discouragement to anyone who has it in them to do radically different typography, provided I can read it better for longer from further away in less light. Even better if it's a sans serif; I prefer them.

Nick Shinn's picture

One example of the inherent conservatism of type is the non-phonetic spelling of English.

No doubt that explains the inherent conservatism of handwriting as well?

If you make unrecognizable characters it will be very annoying to readers of extended text.

William, that's a pleasantry, and as so often you're employing false dichotomy as an arguing technique.
As per the topic of this thread, I propose more variety (of recognizable type), not radically unrecognizable letter forms!

William Berkson's picture

>you’re employing false dichotomy

Well I think that you're mistaking my point. I specifically said that there is room for stylistic variation--including innovation--though much greater in display than in text type. My point was simply that you can't be as radical in type as in painting and still serve the reader. When David Carson replaced the letters in an article with Zapf Dingbats he was being more radical than typographers are--but he was ceasing to be a typographer, as he wasn't serving the author or the text.

John Hudson's picture

One example of the inherent conservatism of type is the non-phonetic spelling of English.

That makes no more sense than saying that the phonetic spelling of German is an example of the inherent dynamism -- or, for that matter, conservatism -- of type. The phonetic or non-phonetic nature of a particular orthography has nothing to do with typographic conservatism. If English spelling were reformed tomorrow to make it phonetic, do you think suddenly typography would be less conservative? Is German typography less conservative than English typography?

English spelling isn't even primarily an example of conservatism in spelling: it's evidence of dynamism in pronunciation. :)

William Berkson's picture

John, evidently I didn't make myself clear. It is not that phonetic spelling or not is inherently conservative. It is rather that the reluctance to change a dysfunctional system shows conservatism. English spelling puts up an unnecessary barrier that makes it more difficult for children and non-English speakers to learn to read English. Therefore there is a strong reason to reform. But the unfamiliarity of the new to those already trained means a short or medium term cost to change. Hence the current system is conserved even though it is clear that another one, once established, would be superior. The same with the keyboard. That's the conservative bias I was talking about.

I don't think the latin alphabet letter forms are dysfunctional--though they have some problems--but even if they were it would be difficult to change them a lot, and quickly, because of the same conservative bias going on.

Nick Shinn's picture

I think that you’re mistaking my point.

No, I was objecting to the way you made your point.

And what is the relevance of comparisons to painting and science?
The same social and technological agents of conformity that affect typography also affect art and science.

William Berkson's picture

>The same social and technological agents of conformity that affect typography also affect art and science.

The pressures are different--that's my point. In art you have a speculative market that encourages novelty. In science I believe one of the requirements for a PhD is that your research have novel results. And they give the Nobel prize for important novelty.

This is just not the same in type. A good revival may be both more honored and more successful in the market than a good novel face. And, to the point of this thread, there are much stronger pressures for a text fact not to be radical: both from the demands of our visual apparatus and of the conservative demand I have just written about to instantly recognize the characters.

The key thing in type is that it serve the author and publisher. That may involve some novelty for excitement or persuasion, or it may not. Novelty is one value among many. It's just fundamentally different from science.

John Hudson's picture

William: It is rather that the reluctance to change a dysfunctional system shows conservatism.

So you think 'design' and 'designate' should be spelled phonetically and thereby lose the semantic link provided by the root? English spelling is erratic because of the mongrel nature of the language, the influence of multiple regional dialects on received pronunciation, dramatic historical shifts in pronunciation, and the relatively late standardisation of spelling, but the basic notion that spelling should preserve roots is not in itself erratic --indeed, it can be very systematic as in Arabic, which compounds non-phonetic spelling by not writing the vowels --, and it certainly isn't dysfunctional: it simply has a different function. Yes, it presents a challenge for learning to read; on the other hand it provides dividends in understanding what you are reading, and facilitates the play of meaning across and around words that is one of the hallmarks of much English literature. It is a mistake to assume that the primary functional criterion of a good spelling system should be ease of learning to read. Many cultures with many languages around the world have developed and used non-phonetic writing systems because they found them useful. Such systems favour those who have learned to read, rather than those who are in the process of learning, so you may argue that they encourage the existence of a literate elite, but I think this is an access to education issue not a spelling issue.

John Hudson's picture

By the way, Mark Rosenfelder developed a very good analysis of English spelling rules as they relate to pronunciation. The rules are complex, because they describe how the pronunciation of letters changes depending on context, but the changes are systematic, which Rosenfelder demonstrates by using his rules to teach a computer program to anticipate pronunciation based on English spelling.

http://www.zompist.com/spell.html

Nick Shinn's picture

The pressures are different—that’s my point.

I don't believe they are, we all operate under the same socio-economic system.
But I don't want to debate to what extent the market for art is speculative, or how much novelty is rewarded in science--suffice it to say that on those issues, as on type, we differ!

So let's talk about type.

there are much stronger pressures for a text fact not to be radical

Aren't we talking at cross purposes?
I'm talking about the cultural pressures that produce uniformity in typography, whereas you're talking about how "radical" faces won't catch on because they're too hard to read--a circular argument if ever there was one, because if a face catches on, obviously it's not radical!

How do you explain, according to your theory, how Futura ever became established, and in such a hurry? At the time, typography had just experienced 30 years of near total historicism--old-style serif faces everywhere. Surely people would have found it impossible to read?! Yes, I know, book typography is the only true immersive reading...

William Berkson's picture

>it provides dividends in understanding what you are reading, and facilitates the play of meaning across and around words

John, I think this is really a stretch. I am very literate and I don't know whether 'should' is spelled that way because of its Saxon origin or whatever, and how that silent 'l' is related to antique conjugation or whatever. Also as anyone reads, once they have learned to read, spelling is largely irrelevant unless the word is unfamiliar. In Chinese, where the addition time to learn reading is multiplied many fold, there is indeed much more of a semantic pay-off in knowing what character represents what sound. But not in English.

>How do you explain, according to your theory, how Futura ever became established, and in such a hurry?

Hurry? How about at least 100 years? In the 1830 commercial sans began to be manufactured and used. And these had ancient roots, and were used in signage long before. That's reflected in the French name for sans, "antique". And Johnston's sans was all over London well over a decade before. And the letter forms never presented a problem to legibility, as they had long been present in hand lettering, I believe. In Futura, the one story a, for example, was in italic and hand writing all along. And the radical form of the g was rejected.

The main change that the Germans made, and it was the biggest change in type in the past hundred years, was to use sans in extended text. Still, even that took specific, larger social pressures to happen. Many Germans, particularly on the left, wanted to be more a part of Enlightenment Europe. They were used to reading roman, blackletter, and sans. I think that going for extensive use of sans was a way to get away from black letter without simply following the rest of Europe and being accused of betraying specific German roots. --They were departing from both German custom and the rest of Europe.

So the change, though the biggest, was simply in usage, and not the introduction of new letter forms.

And it took extraordinary changes in society, as well as talented designers doing visually exciting things, to make even this change.

In our time, I see the main pressures as being technological, rather than social. The introduction of digital type, the need for low-resolution screen fonts, etc. The only real potential for big typographic change is globalization. But here, the pressures seem to have mainly been to 'romanize' non-latin scripts, rather than to affect latin letter forms.

>a circular argument if ever there was one, because if a face catches on, obviously it’s not radical!

No, that circularity is only in your head. You can check whether the letter forms were present earlier, and whether people reported difficulty reading them. That would constitute radical change, and historically changes in letter forms just seem to happen rarely and in small steps. Unlike in science where we've had real revolutions.

Nick Shinn's picture

Like I said, I'm not going to debate whether "science" is more revolutionary than "type". That's comparing apples with albatrosses, and a complete waste of time.

And it took extraordinary changes in society, as well as talented designers doing visually exciting things, to make even this change.

Well, there's hope for us yet!

John Hudson's picture

William: I am very literate and I don’t know whether ’should’ is spelled that way because of its Saxon origin or whatever, and how that silent ’l’ is related to antique conjugation or whatever.

(Currently) monosyllablic words of Old English (Germanic, Anglo-Saxon) origin are not always the best examples of the value of root-signifier retention in spelling, but in this case you've picked a good one, because it is the presence of the l that maintains the relationship of should/shall, sceolde the past tense of sceal, as it does for would/will. The fact that should retains the sense of obligation that modern usage has largely dropped from shall, yet preserving the orthographic relationship despite changes in pronunciation, is a helpful reminder of earlier usage of shall, which is important if one actually reads literature that is more than fifty years old.

As Mark Rosenfelder and others have demonstrated, English spelling is, in fact, very largely phonetic: it's just that a) the rules are complex and b) they favour determination of pronunciation from spelling rather than the other way round. So let's make a distinction between complex phonetic spelling and simple phonetic spelling. In the latter, there is an effort to provide a single mechanism to record each sound of spoken language (with the caveats that the mechanisms may involve digraphs or accents, and the sounds may be compound, i.e. not pure phonemes; some native American orthographies have close to perfect phonetic representation, since they were invented by linguists, and they tend to be typographical nightmares).

What would be the implications of a simple phonetic spelling for English?

Well, we'd almost certainly end up needing more vowel letters, so probably some diacritic accents or more and better regularised diphthongs. Picking up on your example above, how would we spell 'would'? 'Wood'? Is that useful? And of course 'wood' doesn't rhyme with 'snood', so maybe both 'would' and 'wood' should be 'woud', or perhaps 'wūd'? Either way, having two words of different meaning with the same spelling is one of the complaints already leveled against English, so what have you gained while losing the etymological relationship of would/will?

Going back to my earlier example regarding retention of roots, what should (shūd?) we do about 'design', 'designate', 'sign' and 'signature'? Presuming even a Websterian approach to phonetic spelling, itself far from a truly simple phonetic system, we end up with 'dezine', 'dezignate', 'sine' and 'signatyur'. The common Latin root, which has survived intact through divers paths of Mediaeval Latin, Old French and Mediaeval French -- survived intact despite changes in the pronunciation of all those intermediate languages -- is thereby lost.

William, my intent here is not to argue that English spelling is wonderful or that it is not, as you say, a challenge to those learning to read -- perhaps especially to second-language learners --, but that you mischaracterise the reasons for its retention when you cite it as an example of 'conservatism'. Perhaps this too, though, reflects a specifically modern usage, in which the term conservatism is presumed to reflect an attitude, an inertia of desire that conserves things because it doesn't like change. So I would like to suggest an alternative word: conservationism, which implies, I hope, a recognition that there is something of value in what is conserved.

William Berkson's picture

>a complete waste of time.

Well we have a difference of opinion. I think that different fields have significantly different cultures, and they are each worth studying and comparing. I will leave it at that.

Nick Shinn's picture

they are each worth studying and comparing.

Yes they are, but that's not what you've been doing.
You have just compared conservatism in the shape of the alphabet, with "real revolutions in science".
The shape of the alphabet is not a field.
Typography is not even a field, in the broad sense that science is, although design might be.

What, in science, is a meaningful parallel to the shape of the alphabet?
Nothing.

Science has the human genome project. We have Unicode.
Science has manufactured diseases. We have Comic Sans.
It's easy to pick and choose similarities and differences, to try to prove a point, but all it ends up doing is creating dispute over whether the comparison is justified, which doesn't help the topic.

William Berkson's picture

>you mischaracterise the reasons for its retention when you cite it as an example of ’conservatism’.

Well, I think we would have to look at efforts at spelling reform, and the reception they met to answer that. My feeling is that the word origins in roots are so distant from most people's consciousness that that wouldn't have been an obstacle, but I may be wrong. I enjoy looking up word origins--I also looked up 'should' after I posted--and learn from the process, but that could be done also with reformed spelling. Spelling reform is not an obstacle to that, and the roots are sufficiently far from the origins to require looking up in most cases.

russellm's picture

real revolutions in science

99% of science, (including airplane design) is applied. Which is to say, suspect, or at least not pure. Just like the arts.

-=®=-

russellm's picture

arhg - dble post.

=®=-

William Berkson's picture

>>they are each worth studying and comparing.

>Yes they are, but that’s not what you’ve been doing.

My comments on the development of science are the results of my years of study in the history of science, including a PhD in the field. You can read my two books on the subject if you are interested. I could explain my views further on the differences between the evolution of type design and of science, but you have made it clear enough that nothing I say will make the least difference to your views.

William Berkson's picture

Russell, according to the data you can download at this site the United States' National Science Foundation last year funded 75% basic research and 25% applied research. The whole picture is very complex, as it involves the National Institute of Health and industry-funded research, which is overwhelmingly applied. Still, I suspect that your 99% figure is too high.

In any case, I don't see why applied research is 'suspect'. What do you mean by 'suspect'? And why are the arts 'suspect'?

Nick Shinn's picture

>>they are each worth studying and comparing.

>Yes they are, but that’s not what you’ve been doing.

Bill, I wasn't questioning your knowledge of science, just pointing out that you weren't comparing different fields (i.e. disciplines), because the shape of the alphabet is not a field--so how can it be compared to science?

You said, "historically changes in letter forms just seem to happen rarely and in small steps. Unlike in science where we’ve had real revolutions."

I don't dispute either half of your sentence, just their connection, which is a bad analogy.

It seems to me that in order to support you argument for conservatism in typographic form, you are saying that alphabetic form is slow to change, therefore typographic form must be slow to change. That is a non sequitur.

No, we have not had revolutions in alphabetic form; but yes, we have had revolutions in typographic form--and Futura/Gill/Kabel c.1930 was surely just such a thing, following thirty years of sustained and near complete serif-based historicism in the graphic arts--historicism which was itself a revolution. I don't deny that every revolution has its precedents, and indeed, as you mention, the new genres of commercial type that emerged in the 1830s were another revolution. Type history is full of revolutions!

You have also argued that because English spelling is complex, therefore typography should not change too fast. That too, as John has explained, is a non sequitur caused by a bad analogy.

William Berkson's picture

>how can it be compared to science?

Just as you can compare anything else. Above you argued that all fields are influenced the same way by cultural pressures. Now you argue that they can't be compared at all. Your arguments are completely contradictory. In any case, there is absolutely no reason we can't compare the evolution of type design to the evolution of architecture, poetry, airplane design or any other field. How useful that will be will depend on the question we are discussing.

I was addressing the issue of innovation, and if you look at type of say, 1540--Claude Garamond--it is not only perfectly readable today, but greatly admired, continually revived, and widely used as a text font.

If you look at science of 1540, Copernicus was about to be published. People believed the earth was at the center of the universe, and the sun and stars revolved around it. Yet to come, nearly a century later, was Galileo, with his telescope and new physics. And William Harvey and the discovery of circulation of the blood. Yet to come was Newton and astoundingly accurate theory of motion in the heavens and earth--and a concomitant weakening of religious belief. Yet to come too was the theory and use of electricity, the atomic theory of chemistry, relativity, quantuum theory, and the discovery of the structure of DNA.

If you wanted to use the science of 1540 today you would be regarded as insane--or criminal if you are an engineer. If you want to use the type of Claude Garamond--well, conventional but perfectly acceptable.

I know how very able a type designer and knowledgeable a type historian you are, but your implication that Gill is on the same level as an innovator as Newton or Crick & Watson is simply ludicrous.

>You have also argued that because English spelling is complex, therefore typography should not change too fast.

That isn't even remotely what I was arguing.

John Hudson's picture

William: Well, I think we would have to look at efforts at spelling reform, and the reception they met to answer that. My feeling is that the word origins in roots are so distant from most people’s consciousness that that wouldn’t have been an obstacle, but I may be wrong.

I think opposition to spelling reform comes from a number of different directions. Some opposition is what I would call conservationist, i.e. from people who understand and appreciate the positive aspects of an etymological spelling system; such opposition tends to be focused among academics and writers. A broader opposition derives from what I could classify as practical concerns about both the process and results of spelling reform. Depending on the nature and extent of a spelling reform, it makes the majority of the population temporarily illiterate or, at least, of reduced literacy. This is why exercises in spelling reform, as in script change, have been most successful in situations where the majority of the population is already illiterate (80% in Ataturk's Turkey when the Latin script orthography was introduced in the 1920s). Once people can read, they are not inclined to have to learn to read again. This is a kind of conservatism, I suppose, but it is one founded on practical concerns about lost time, impact on the economy, etc. and not in a conservative attitude or disposition. There are also practical concerns about which approach to take to spelling reform, and hence what the results will look like. Once you open the door to spelling reform, people are unlikely to agree on which reform should be adopted. [If Hrant were here, I'm sure he would point out that spelling reform would be much easier if one had a king/dictator to lay down the law, and he is certainly right.] Finally, we have the example of attempted spelling reforms from other languages -- the monotonous Greek example, the traditional/reformed Malayalam debate, and the less far-reaching German reform of 1996 -- and what we see is that the old spellings never actually go away, and one ends up with two (or more) different orthographies for the same language, with some people using one and some the other.

Nick Shinn's picture

you argued that all fields are influenced the same way by cultural pressures. Now you argue that they can’t be compared at all.

No, I said that the shape of the alphabet is not, as you had clearly stated, a field. Therefore it cannot be compared to science. You talk about the shape of the alphabet not changing as being a fundament, well so is the fact that chemists still mix liquids in glass vessels and heat them!

If you wanted to use the science of 1540 today you would be regarded as insane—or criminal if you are an engineer. If you want to use the type of Claude Garamond—well, conventional but perfectly acceptable.

Firstly, the heart of science--mathemetics--was in place in 1540. They used the same number system as we do. In observation and classification, would you not say that a medieval herbal book would be regarded as more sane than, say, string theory, with its DC Comic multiverses? Also, their cathedrals seem to have been generally constructed on sound principles. Of course, the theoretical aspect of 16th century science is not very convincing today, but the same could be said of 16th century typographic theory. As with science, a core has remained (e.g. the format of the "codex"); we stand on the shoulders of giants, but much else has changed.

Secondly, Garamond's type, although it would be legible, would not be acceptable--even for that tiny amount of typesetting done in letterpress shops. Consider that the verbatim transcriptions of Renaissance type (e.g. Poliphilus) are of little use.

Thirdly, I have to point out again the irrelevance of your comparison, because you are not comparing discipline with discipline. A revolution in a discipline is a revolution in a discipline. Crick, Watson et al revolutionized genetic theory as it was understood at the time. Gill, Renner, and Koch revolutionized typography as it was practised at the time.

my years of study in the history of science, including a PhD in the field.
I know how very able a type designer

Bill, I'm not pulling rank on you with the "appeal to authority" argument. This kind of thing is irrelevant and unworthy in a forum on a multi-disciplinary subject where participants come from all kinds of backgrounds. What one says should stand on its own feet.

your implication that Gill is on the same level as an innovator as Newton or Crick & Watson is simply ludicrous.

Rubbish. Gill had comparable, if not more effect on typography than Crick, Watson et al had on science.
As an innovator, political activist, style icon and all-round eccentric celebrity, Gill would best be compared with Einstein.

Scalfin's picture

One reason not to adopt a phonetic alphabet is that it would make the letter r obsolete (and if that doesn't tell you where I'm from, nothing will).
The qwerty layout was designed to maximize hands alternation, not slow you down. That myth is actually struck down on wikipedia, for god's sake!

Letter forms have changed quite a bit over time. I can say this with some certainty because I researched the history of each letter for my font, and found out how long blackletter was the primary type in many areas (you'd also be surprised to learn that some areas of New York spoke Dutch exclusively well into the 80's)

russellm's picture

according to the data you can download at this site the United States’ National Science Foundation last year funded 75% basic research and 25% applied research.

I am sorry, but a 25/75 split simply doesn't lend itself to hyperbole.

-=®=-

Syndicate content Syndicate content