Archive through March 10, 2003

paul's picture

Kent and Tiffany,

I'm really pleased to see somebody doing scholarship on Dwiggins. I've been collecting his books and writings for years. I've occasionally thought about putting together an article or small book, since I'm not aware of anything recent being written about him

kentlew's picture

The Vale King's Fount can also be seen in Updike, Vol. II,
figure 351 (opposite pg. 211 in the third edition).

The Linotype digitization of Electra will *not*, unfortunately,
yield results anything like the original.

Dorothy Abbe passed away in June 1999. There was
no obituary in the local Hingham newspaper. According
to her own wishes, the book community was not
notified of her death until nine months later. In 1983,
Abbe wrote a personal reminiscence of her time with
Dwiggins, but it remains unpublished in her archives.

Apparently she was not shy (or kind) in her opinion
that no one else was qualified to write a biography of
WAD, but she never wrote one herself. I suspect that,
ironically, she was too close to her subject and that the
task just proved too daunting. She did, however,
produce two works focused on aspects of Dwiggins's
work -- The Dwiggins Marionettes and Stencilled
Ornament & Illustration.


There is a very nice (albeit brief) biography of Dorothy
entitled "Strings Attached" by Anne C. Bromer, designed
by John Kristensen (of Firefly Press) and published by
the Boston Public Library in an edition of 500 in 2001.

-- Kent.

hrant's picture

Congrats John!

----

> many people confuse seriousness and solemnity

Good point.

hhp

paul's picture

HI Kent,

Thanks for the information about Dorothy Abbe. Sounds a lot like Lillian Hellman and Dashiel Hammet. I do have both the Marionettes book and Stencilled Ornamentation, plus her little booklet about WAD published as part of an exhibition by the Boston library.

I was tremendously excited when I bought the digital Electra, and although I do use it, I agree that it sure doesn't have the same effect as the metal version

hrant's picture

> the only alternative is setting the job on a Linotype machine.

If my suspicion that the digital Electra was made from freshly cast type, then there's a much better way to regain its original glory: print it photopolymer/letterpress!

hhp

hrant's picture

BTW, Oak Knoll books has a great Winter Sale going on (with progressive discounts for multiple purchases), and their current catalog (M557) shows 7 works about/by Dwiggins.

http://www.oakknoll.com

hhp

bieler's picture

"I was tremendously excited when I bought the digital Electra, and although I do use it, I agree that it sure doesn't have the same effect as the metal version

kentlew's picture

"The Linotype digitization of Electra will *not*, unfortunately, yield results anything like the original."

I suppose my comment is relative to one's familiarity with Electra and one's pickiness. Anyone who knows me will tell you that I'm picky.

For education purposes, I have prepared a small comparison of the two versions. The first image is from the debut of Electra, Emblems and Electra, published by Mergenthaler in 1935. At this stage, the face was completed in twelve point only and had not yet been graded (i.e., adapted and cut for other sizes). The second is a facsimile that I produced using the digital Electra LH, printed at 1200 dpi on toned paper to match the value of the Lino sample. Both were then scanned and the background normalized. My sample, although produced from a laser printer, is a fair representation of what the digital type looks like when printed in today's books.

Electra Lino sample
Linotype metal

Electra Digital sample
Digital

Certainly some of the difference in the overall color is due to the difference in printing technology. That is unavoidable. Even so, the digital Electra makes a very different texture on the page. It has a higher stroke contrast than the Lino and the fit is a little wider. Many of the letters are visibly wider in the digital version -- look closely at the 'h, n, m.' Observe also that the 'o' and 'e' in the digital are noticeably rounder, the 'a' is more squat. The biggest difference in character shape occurs in the 't.' The original has a substantial tail which adds a great swing to the line, while the digital 't' is precariously balanced on a shy bottom.

The differences are, I admit, subtle. But on the page, they accrue to create a result which is, in my opinion, palpably different from the original.

I strongly suspect that the digital version was patterned after the 10 point Electra font. If so, this would explain the slightly wider character proportions and the looser fit. The original cutting, and by far the most useful size for books, was twelve point. (The samples are set 12 on 16.) I don't know why the 10 point would have been chosen as a model for the digitization of this quintessential book face.

The sentence in the middle of this sample reads, "Make a line of letters so full of energy that it can't wait to get to the end of the measure." The original Electra does this admirably. The letters march in step and fuse wonderfully into rich word shapes. Take a look at a word like 'something' or 'what.' The digital, by comparison, falls a little short of the mark, in my opinion.

My frame of reference is as a book designer. Electra was arguably the most popular and most widely used text face during the heyday of American book publishing in the 40s and 50s. While I would happily use Linotype Electra for just about every book (if only I could), I will not specify the digital version. It just doesn't do the job. I am always disappointed when I see it used in books today because it is such a pallid ghost of its predecessor. Perhaps if I wasn't so familiar with the original I wouldn't feel so strongly.

-- Kent.

kentlew's picture

I forgot to mention: the digital display version is not actually the exact same as the text. The stems are marginally thinner, the serifs are noticeably thinner and tapered ever so slightly, the characters are a little narrower (closer to the original twelve point), and the x-height is a little smaller.

-- K.

hrant's picture

Kent, thanks for that great comparison!
I never realized that the "t" (the second most frequent letter in English) really is very different.

But one thing that maybe you're not considering:
The fitting is looser in the digital, but you can see that both are fitted about right for their color. So maybe the digitizers consciously chose to make it lighter (catering to [supposed] contemporary tastes? like Crawford* versus Mansfield), and realized -correctly- that it needs to be looser as a consequence? If the vertical proportions are the same, that's a clue that the master was indeed the 12.

* BTW I've seen Cindy Crawford in person (twice!) - she looks like a total camel.

In any case, the original is still way better.

hhp

kentlew's picture

Hrant, it seems to me that the digital is fitted not just
for the color, but also for the change in counters. Yes, it's
not a *bad* fitting, per se. But the result is not as good
as the original. The lighter color may, indeed, be some
sort of contemporary taste. There are undoubtedly
some who prefer the digital version to the original.
Who can account for taste?

My primary point was that digital Electra is a thing unto
itself and not quite like the original -- so those who see
Electra being used in books from the 40s and 50s and
observe that it is a "killer font" may be disappointed
when they license the digital font and attempt to get
the same results.

That doesn't mean that the digital version is
unjustifiable or that it shouldn't be used, necessarily.
But I personally don't think it does Dwiggins's design
justice.

Perhaps someone from Linotype monitors these
forums. If Linotype are interested in continuing their
program of reworking standards -- like they did with
Frutiger, Optima, Sabon, et al. -- then they might give
consideration to an "Electra Next" which would take
another look at the original designs and put some meat
back on the bones and some blood back in the veins of
this classic workhorse.

-- K.

William Berkson's picture

Kent, I am wondering whether there was a general problem with early digital versions of great classics - that they just didn't know how to take into account the spread of letter press.

As one who gets nostalgic looking at the pre-digital books, I am wondering whether the softness of the outlines in letter press from ink spread was part of their charm, and whether this has been lost in the translation to digital. What do those of you with experience in designing typefaces think?

Also, Kent, I think you are too kind to the digital Electra. You comparison makes strikingly clear that the digital kills the poetry of the metal version. The amazing thing about the metal Electra is that it seems so disciplined and free of quirks, yet is full of dynamism and vigor.

kentlew's picture

>Could this boldness be at least partly attributable
to ink gain from the press? And if so, did WAD design it
with this in mind?


Undoubtedly the weight comes in part from the ink
gain. But this amount of gain was standard for the time
and Dwiggins did design with this in mind. Most of the
progressive proofs that Lino made were printed on
several different papers -- both smooth and "antique"
book papers. The sample I presented represents
exactly the sort of thing Dwiggins would have been
looking at and approving.

There is additional discussion of this (and a citation
from Dwiggins's letters, which I won't repeat here) in
this thread over in the archives of the Typo-L
discussion list, from January of this year.

>I am wondering whether the softness of the outlines
in letter press from ink spread was part of their charm,
and whether this has been lost in the translation to
digital.


I think the quality of letterpress does contribute to the
charm and vigor of the great types of that era. And
some of that has definitely been lost in the transition to
digital and offset. I don't necessarily think, however,
that we should strive to *mimic* that softness in our
own technology. But neither do I think we should
ignore these factors when transcribing metal designs to
digital. I think there is something valuable to re-learn
by looking back at metal types; but then we must move
forward.

We can only speculate how Dwiggins would feel about
the digital technology of today. I know that Dwiggins
expressed several times his desire to get sharp, crisp
type and images. He was a progressive thinker. I'm
sure he would have embraced the possibilities. But I
don't think he would have approved of the brittleness
and frailty of digital/offset Electra or Caledonia.

-- K.

bieler's picture

"I am wondering whether there was a general problem with early digital versions of great classics - that they just didn't know how to take into account the spread of letter press....

I think you are too kind to the digital Electra. You comparison makes strikingly clear that the digital kills the poetry of the metal version. The amazing thing about the metal Electra is that it seems so disciplined and free of quirks, yet is full of dynamism and vigor."

There would have been no reason for digital type designers to be concerned with inkspread. What was the probablity anyone would try and print a digital face letterpress?

Actually, the Linotype printing reveals a lot of ink spread and if you look at it closely you will see a badly disturbed placement of the letterforms on the baseline. This isn't "charm"; it is the result of the casting process and the matrice lineup not the letterforms themselves. Hard to judge a letterpress designed typeface without taking into account alterations brought on by machine and presswork.

If anything, the digital version is a correction of some of these problems.

paul's picture

Kent, thanks for showing the comparison between metal and digital Electra. One of the best things about the original is its solid color; the digital, although I still like it and use it, looks so pallid by comparison.

I'd like to point out that Electra has some of the best figures, ever. Aren

hrant's picture

> Dwiggins did design with this in mind.

In fact he was one of the few to take gain seriously, to the point of measuring it, as well as (apparently) specifying "minimum clearance" values in his glyphs.

hhp

bieler's picture

Wouldn't the digital Electra have been based on original drawings rather than printed specimens? (Where is Bill Troop when you need him?) Even if Dwiggins built into his design some compensation for the letterpress process there would have been no need for the digital version to incorporate these, that is, to alter the drawings to imitate what the face would like as if letterpress printed.

paul's picture

Here is a sample of the original italic for Electra. Obviously, I was wrong in my previous post about only some lower case letters being replaced; obviously, the entire set is different. I also had never realized that the difference in slant between the italic letters was introduced with the cursive version; it's not present in the italic.

I have not seen too many examples of the italic actually used. Electra might not have become the success it was if the italic had not been replaced.

The scan is from The Power of Print

trajan's picture

The Electra numerals are great.

This brings back nostalgia for me, as I remember picking up a vintage paperback when I was an adolescent (in the '80s) and seeing numerals below the baseline for the first time. I immediately liked it and wondered why they weren't like that in any book I'd seen before.

Years later I realized that this was my first initiation into fine typographical design, and though the book is long forgotten in my mind, I feel certain that the font was either Electra or Caledonia.

So I owe a lot to WAD!

hrant's picture

> Wouldn't the digital Electra have been based on original drawings rather than printed specimens?

Original drawings tend to be hard to come by.
My suspicion is that it was based on the metal sorts (not actual printing).

> I greatly prefer the Cursive.

I myself respect the original italic, and see the cursive as a commercial selling-out.
It's easy to like the latter, while the former requires a culturally progressive outlook.

In fact to me the slanted-roman idea was the single good thing Morison came up with - unfortunately he (and others) took it way too literally. It would have been much better to instead promote/implement non-calligraphic, rigid, tame (like no descending "f") italics.

hhp

bieler's picture

> Wouldn't the digital Electra have been based on original drawings rather than printed specimens?

"Original drawings tend to be hard to come by.
My suspicion is that it was based on the metal sorts (not actual printing)."

Hrant

Think I probably screwed up the original return here so here it is again:

But it's a Linotype re-issue. They would have had either had the original drawings, the original matrice master patterns, or, at least, the matrices (which are still around here and there). No need to go to printed specimens at any point here. But, unlike foundry or Monotype, there are no metal sorts with Linotype. This machine is a friggin line caster.

bieler's picture

Another thing that has occured to me. Did Linotype ever issue Electra as photofilm? If so, they probably made the digital from this. A lot of photofilm was issued based on 10 or 12 point patterns only, a legacy kindly followed in digital practice.

kentlew's picture

>Actually, the Linotype printing reveals a lot of ink spread and if you look at it closely you will see a badly disturbed placement of the letterforms on the baseline. This isn't "charm"; it is the result of the casting process and the matrice lineup not the letterforms themselves.

Gerald, I'm not sure this is the whole reason. Granted, there are idiosyncracies and imperfections introduced by the machine and the process. But, first of all, this specimen was produced by Mergenthaler and, presumably, they would have produced their specimens to the highest possible standard -- in the setting, at least, if not the printing impression.

Secondly, the alignment in the letterpress sample is not substantially different from progressive proofs that I have with Dwiggins's annotations on them. The dropped 't', for instance, and the extreme bottom overshoot of the 's' and 'w' -- these are present in the proofs. There is even evidence of the 't' in some of the original 10x12 ink drawings. Surely if Dwiggins disapproved of these features, he could have had them recut and recast. The machining process was quite precise and the tolerances quite small. Instead, I maintain that, for better or for worse, these things were intended by the designer.

>Another thing that has occured to me. Did Linotype ever issue Electra as photofilm? If so, they probably made the digital from this.

This may be closer to the truth. According to the Linotype Library GmbH catalog, Electra was digitized in 1994. American Linotype was dissolved by then, I'm pretty sure; and while the assets went to Germany, the original drawings and brass patterns went to the Smithsonian (some materials have since been distributed elsewhere). I'm fairly certain that Electra was converted to photofilm, so it is quite possible that the digitizations were made from film masters. I can try to check up on this.

>My feeling is that WAD might well have approved of the digital version that everyone is now convinced is a trevesty compared to the metal orig.

Obviously, I disagree. Don't get me wrong, I do not think that Dwiggins would have been adamant about mimicking the letterpress look -- quite the opposite, he would have been interested in progress. And if he'd been involved in any conversion process, I don't doubt that he would have taken the opportunity to make improvements based on experience gained since his first text design. But from everything I've learned so far in my research about his process and his tastes -- which types he approved of and which he didn't -- I don't think he would have been satisfied by the letters produced by the current digital version.

-- K.

William Berkson's picture

I think Kent's revelation about Dwiggin's proofs shows that, contrary to Gerald, most of the variation was deliberately designed. I have a 1962 book set in Electra, and the 't' dipped below the baseline is there, although less pronounced than in Kent's sample.

Also Gerald's view (if I understand it rightly) that you shouldn't take into account the ink spread of classics in doing revivals, I don't think holds up. If Dwiggins took it into account, why shouldn't someone digitizing Electra try to reproduce the effect of the printed version rather than the original matrices? It is also inconceivable to me that Caslon or Baskerville weren't also taking it into account.

The classic standard for beauty is "unity, variety, and balance". It seems to me that the option of being incredibly precise with digital type & photo offset printing may have tempted designers of text type to make the typeface too uniform. I suspect this affects not only beauty of words, but also readability.

hrant's picture

> This machine is a friggin line caster.

Good point, although they could still have digitized from the lead. But I agree that's unlikely (and your phototype idea makes the most sense*). BTW, how could you digitize from matrices?

* Although of course the question simply becomes where that came from...

> The classic standard for beauty

No such thing, my friend!

--

BTW, I once bought an issue of The Colophon (New Series, V3#2, Spring '38) simply because it used the original Electra (with the Italic, not the Cursive), and although "lone" settings of the Italic do look ungainly, I think it works very well for emphasis. Which makes sense considering the impetus behind the slanted-roman idea.

hhp

John Hudson's picture

Regarding ink spread:
I think it is poor design to try to slavishly mimic the precise effects of ink spread in digital revivals of metal type, because one has to reckon that some aspects of that spread are beyond the control of the original type designer or punchcutter, reflecting the particular ink composition, presssure, paper surface, etc. employed in a particular printing. On the other hand, reference to letterpress printed sources is absolutely necessary to get a sense of the desirable weight of the type (for this purpose, multiple examples are better than single sources, but best of all are examples printed by or under the supervision of the original type designer). So the particular pattern of ink spread, e.g. at junctions, is not necessarily something to be revived, but the overall impact of ink gain on stems, bowls and hairlines is.

kentlew's picture

John --

Precisely! In typical fashion, you have struck right to
the core of the matter. I couldn't have said it better
myself. This is my position exactly; thank you for
putting it so clearly.

-- Kent.

anonymous's picture

"The Linotype digitization of Electra will *not*, unfortunately, yield results anything like the original."

I printed out the digital version, and it looks pretty good, I think. I mean, the only alternative is setting the job on a Linotype machine.

anonymous's picture

Kent, thank you for the education. The metal version, at least this sample you provide, seems too bold for typical body type. Could this boldness be at least partly attributable to ink gain from the press? And if so, did WAD design it with this in mind?

Though it deviates mildly from the original, I still think the digital version is pretty good. At the very least, better than Times New Roman!

The main thing I don't like about the digital is the hyphen. It slanted at a distinct angle in WADs' orig, now it's just a straight line. Now THAT really is senseless.

anonymous's picture

"Actually, the Linotype printing reveals a lot of ink spread and if you look at it closely you will see a badly disturbed placement of the letterforms on the baseline. This isn't "charm"; it is the result of the casting process and the matrice lineup not the letterforms themselves. Hard to judge a letterpress designed typeface without taking into account alterations brought on by machine and presswork. If anything, the digital version is a correction of some of these problems."

Exactly. Everything's a trade off. My feeling is that WAD might well have approved of the digital version that everyone is now convinced is a trevesty compared to the metal orig. There are inherent flaws in the casting process, and it undoubtedly drove men like WAD nuts to see jagged baselines, for example. But, granted, every typophile knows that the "perfection" of digital fonts + offset printing certainly doesn't have the same impact as metal + letterpress.

So, again, a trade off.

The Electra Cursive was indeed an entirely different font offered as an alternative to the "slanted" quasi-italic of the orig. I greatly prefer the Cursive.

Syndicate content Syndicate content