Historical background of De Vinne, Howland and other related late 19th century american typefaces

Kent Lew

Along the lines of De Vinne, David Jonathan Ross’s Roslindale has emerged recently as a pretty successful homage.

Claudio Piccinini

Kent Lew said:

Along the lines of De Vinne, David Jonathan Ross’s Roslindale has emerged recently as a pretty successful homage.

Yes, it’s very nice. But, albeit very carefully drawn and faithful to the spirit (i.e. not straying away much compared to late 1960s and 1970s reinterpretations like Bernase or Trooper Roman) is a revisitation.
Myself, for now I am trying to get into the forms and the “spirit of the times” to attempt some pretty faithful digital versions, then I might see what they inspire me.

Claudio Piccinini

I am trying to figure out how many capital /S/ forms De Vinne was originally produced with.
I always believed it just had the basic form (whose counters are pretty open compared to other letters) and the swashed, more decorative one.

But on close inspection, I found that it seems there is a third capital form, which in the later versions called "De Vinne No. 2" from ATF is also used in the lowercase, and this form, which can be seen in the scan below (early Central Type Foundry specimen) in the first line in 48 pt size, is rounder (with rounder counters), and more harmonized with the other capitals. The decorative form, in its lowercase version, can be seen in the second line of the picture in 42 pt size.

What do you think? For historical accuracy I’d be oriented to keep all three, but I wished to know how widely the second form was used in earlier years, as the first seem to predominate.

Claudio Piccinini

Any opinion on this? I know it’s not easy to historically determine whether the three forms of "S" were already there in 1892 (it appears so) and if there is something about this. The two forms are documented (many period uses too) but the third not so much… :-( 

Jacob Casal

It will take a few weeks maybe, but I think I can get you some pictures of DeVinne /S/s and /s/s that aren’t just scans. For now I think there are two pairs of styles per lowercase and capital /S.  I’ll follow up on this message later.

Claudio Piccinini

Jacob Casal said:

It will take a few weeks maybe, but I think I can get you some pictures of DeVinne /S/s and /s/s that aren’t just scans. For now I think there are two pairs of styles per lowercase and capital /S.  I’ll follow up on this message later.

Ooh, thank you. Yes, surely two, for the uppercase definitely three but what’s not clear to determine is whether they were there when it was released to begin with.
Surely the later versions, the Monotype one (I believe) uses the “smoother” and more harmonious form of /S.

Jacob Casal

Claudio! As promised the type is here. I have organized to have some high quality macro shots taken for archival purposes. It is a set of metal 30pt De Vinne, spacing included. It’s going to take some time to take good top down pictures. Evidently there was a 36pt condensed around as well, but I’ve yet to find it. I’ll post more details when the good pictures have been taken.

Here is a rough sneak peek of /S taken via a regular camera.

Claudio Piccinini

Jacob Casal said:

Claudio! As promised the type is here. I have organized to have some high quality macro shots taken for archival purposes. It is a set of metal 30pt De Vinne, spacing included. It’s going to take some time to take good top down pictures. Evidently there was a 36pt condensed around as well, but I’ve yet to find it. I’ll post more details when the good pictures have been taken.

Here is a rough sneak peek of /S taken via a regular camera.

That is very good and much appreciated! I haven’t been able to work on it for the last ten days or so, so it will be very timely.
Although I am working on a master based on 60-72 point samples, it will be great to have an overall sampling of the 30 point size.

My initial idea was to tackle Howland to pass to DeVinne Condensed, and familiarize with the forms while doing a digital version as accurate as possible, but I realized it‘s better to “grasp” the essence of the forms starting with the Roman. And this, of course, will take time.

Claudio Piccinini

I haven’t been updating the thread because I began working on De Vinne and temporarily stopped Howland. In the meantime I have also started a new typeface of mine inspired by Howland and other period condensed styles.

However, knowing my “dispersion potential” I just picked it up again and fixed a pair of letters. Following Craig’s advice, for the glyphs which presents apparently "unsmoothed" inconsistent solutions, I am working to keep both version: a more faithful and a harmonized one.
I still have to restore the "vertical“ counterspaces in /C and /G but here’s an updated Fontlab screenshot with the /Os, /Q and /8 in their more "philological" forms… :-)

The basic character set is not fully complete yet, so I haven’t generated the font yet.

Claudio Piccinini

Ah! the good old inconsistencies…
I’m wondering how to be faithful to the original form(s)… 

Jacob Casal

It’s a tough call, to be sure. I’m not sure as to how much material using the typeface is available to you, but I would likely look within each size for the most common angle of the spine and how often the sharper angle at the end of the spine occurs across all /S’s as a start. In the end it seems to come down to the eye arbitrating which one seems like it belongs with the other letters most. There’s also the off-chance of the idiosyncrasies themselves being intentional.

Claudio Piccinini

Jacob Casal said:

There’s also the off-chance of the idiosyncrasies themselves being intentional.

Thanks much for the considerations and advice to begin with.
Being entirely intentional: I’d say it’s not the case, not given the arbitrary nature of them across point sizes.
To answer your question: in point sizes lower than 24 letters tend to be homogeneous in form, so it’s mostly a matter of the bigger sizes.
I have also discovered from a specimen that point sizes 72, 96 and 120 were produced afterwards, and this at least partly explains the differences in design (/S for them is wider and no spur/sharp angle), so I’d stick with sizes not above 60 for my initial master which have to be consistent with the text/subtitling masters I would like to add afterwards.

Claudio Piccinini

@Mark Simonson, which I thank a lot, recently gave me some precious advice on Howland, that (with his permission) I want to share here.

«It’s quite odd, some of the inconsistencies, almost as if different people worked on different sizes and had different opinions. Or it could just be that the designer felt that exact consistency from size to size was unnecessary or didn't matter.

My inclination would be to pick the form that is most representative of the variations, or even the one you prefer. I don't think it would be that useful to include alternates when the differences are subtle enough that most users would probably not notice. 

Doing a display and text version makes a lot of sense, since they are so different, and would satisfy 99% of the reasons (in my opinion) of doing a Howland revival. Reviving all the little differences between different cuts would be less worthwhile.

I always try to think of the users and not just what would be interesting or satisfying to me to do.»

Claudio Piccinini

It’s quite odd, some of the inconsistencies, almost as if different people worked on different sizes and had different opinions. Or it could just be that the designer felt that exact consistency from size to size was unnecessary or didn't matter.

On this, I’d be inclined to second the first supposition, as John F. Cumming was enough precise when he wanted to be. Some of his designs at least show this.

Claudio Piccinini

An important italian use of Ihlenburg's Columbus (or – after 1899 – "Victor Hugo" as recut in Italy by Nebiolo, and – in the early 1900s – "Columbia" as recut by italian foundry Urania), on the cover of the first edition of Dino Campana’s "Canti Orfici" (1914).

The story of this book, besides its importance, has been adventurous, and this aspect also extends to the design of the book, as featured in some articles by researcher Massimo Gatta. More information here on Wikipedia (unfortunately, in italian only): https://it.wikipedia.org/wiki/Canti_Orfici

Claudio Piccinini

FYI Columbus as "Columbia", in an Urania catalog from 1909 (I recall I sent the scans to @Thomas Phinney when he was still in the early stages of his digital version of Columbus). :-)

Claudio Piccinini

To show the impact the book has had in italian culture, some recent materials making use of Columbus: a concert logo, a leaflet issued for the centennial celebrations of the book and a poster for a "musical marathon.

All of these apparently make use of one of the pre-existing (bad) digitizations of Columbus floating around before Thomas started his project for a better digital version.

Stephen Coles

Claudio, just came across an ad for Howland in the September 1899 issue of Inland Printer in which they announce new large sizes (72–120pt) not shown in previous ATF catalogs. I love how they use it in that narrow sidebar. The big Jenson Heavyface is great too, though that ‘S’ was rushed off the line.

Claudio Piccinini

Stephen Coles said:

Claudio, just came across an ad for Howland in the September 1899 issue of Inland Printer in which they announce new large sizes (72–120pt) not shown in previous ATF catalogs. I love how they use it in that narrow sidebar. The big Jenson Heavyface is great too, though that ‘S’ was rushed off the line.

Ooh, thanks. Very nice of you. I was just talking with Antonio (Cavedoni) about this, and I’ll probably pick up the search once again in short.
The "new sizes" 72-100pt, are – de facto – almost a different design. Some letters (/S, /R but even /E, a very large /C /a etc.) are very different from the larger pre-existing sizes (48-60pt).
As insightfully pointed out by Mark Simonson, there are a lot of inconsistencies across sizes, so – while in some cases these are worthy to be somewhat preserved, in some cases – when a letter is too different across sizes – the best choice seems to capture the "average essence" of the letter.
My design so far is closely based on 48-60 pt sizes, I’m striving for a high degree of respect of the originals, but nonetheless, choices must be made.

Claudio Piccinini

Hello everyone,

after over a year I have decided to pick up my digital version of De Vinne to complete it.

So far I have worked on the largest cut, to be used at display sizes (based on scans of 60-72pt size).

My intention is to continue to design the family as faithfully as possible to the original, in its various series, and also to do digital versions of its forerunners and related typefaces (like Howland). The purpose is twofold: having them in digital version with a good degree of faithfulness and to study the forms for a new design of my own that I have been researching and studying, which references the period, but mostly not from typographic sources (rather from lettering and inscriptions), and more from an italian/european angle.

While I wish to thank again all of you that helped me and contributed (Jacob Casal in 2019 sent me a number of pictures taken from the lead typeface), I will try to ask more questions as I am trying to figure out how to find glyphs which are not usually available in specimens, as I wish to be as faithful as possible with my rendition.

Here’s a small screenshot to show its current state. :-)

Claudio Piccinini

My first request for advice is about historical accuracy. I tried to find as many samples as possible of De Vinne, but there are some glyphs which eluded me so far, even at smaller sizes and that I am pretty convinced they should have been there.

For some diacritics, printed samples in other languages had been precious, like this one from a 1905 German ad (yes, De Vinne’s /hyphen is fashioned that way). :-)



Aside from brackets and braces, which couldn’t have been there, I am wondering if any of you have ever stumbled in printed examples of the following glyphs:
— section;
— paragraph;
— registered/copyright/trademark;
— numbersign.

I doubt De Vinne included currency symbols other than Dollar and Sterling since it appears from official specimens they were using /c as the /cent, but any other unusual or less used glyph one may be able to spot, I’d be very grateful to know!

Stephen Coles

De Vinne was frequently used in ads during the early 20th century. I recommend trawling the American/Inland Printer and other magazine and newspaper hi-res archives from 1900–1920.

Claudio Piccinini

Stephen Coles said:

De Vinne was frequently used in ads during the early 20th century. I recommend trawling the American/Inland Printer and other magazine and newspaper hi-res archives from 1900–1920.

Thank you very much for the searches! I recall the samples often shown from period ads, but this is very handy. I will check the ones closer to 1892 first. :-)

Claudio Piccinini

Huh? What is this supposed to be? A page from a 1892 (or 1893) Inland Printer issue. It’s an outline version of De Vinne (which was issued by a few foundries some years later, BTW) which seems to have been patented as an “original” design?

Stephen Coles

Interesting! The lowercase is significantly different from De Vinne, though. It’s not mentioned in the article, but Gnichwitz patented several designs for Keystone. Here’s their 1906 showing of De Vinne Outline:


The earlier design doesn't appear in this catalog, which is the earliest Keystone I know of online.

Claudio Piccinini

Stephen Coles said:

Interesting! The lowercase is significantly different from De Vinne, though. It’s not mentioned in the article, but Gnichwitz patented several designs for Keystone. Here’s their 1906 showing of De Vinne Outline:

The earlier design doesn't appear in this catalog, which is the earliest Keystone I know of online.

Thank you, Stephen! It might actually be be the Outline version, then. In patents, they are usually reproduced small, possibly in non-final versions.

Now I have entered the rabbit-hole of studying Times, I must stay focused and finish De Vinne to proceed with order! :-)

Claudio Piccinini

@Thomas Phinney: Did you see this Hermann Ihlenburg’s sketch for Columbus? :-)
“Black Hen”!

Thomas Phinney

I do not recall now if I saw this, but it is typical of many Ihlenburg sketches I looked at in the Cary collection:

• looks like India ink (over light pencil, right? did you see evidence of pencil marks? I think I remember this)
• very polished and finished, even though it is just a couple of words
• has all caps and lowercase both
• very close to the finished typeface, not a lot changed between this and the final version. (Although that “C” looks a bit wider.)

Another one I remember was “WAVERY harmony” (for an unproduced typeface).

Claudio Piccinini

Thomas Phinney said:

I do not recall now if I saw this, but it is typical of many Ihlenburg sketches I looked at in the Cary collection:

• looks like India ink (over light pencil, right? did you see evidence of pencil marks? I think I remember this)
• very polished and finished, even though it is just a couple of words
• has all caps and lowercase both
• very close to the finished typeface, not a lot changed between this and the final version. (Although that “C” looks a bit wider.)

Another one I remember was “WAVERY harmony” (for an unproduced typeface).

Yes! – the sketchbook is from that collection.
I wonder when he designed these very polished and finished samples, as they’re undated. The sketchbook in itself is dated as "1868-1888", but maybe it’s just an extimation?

Claudio Piccinini

De Vinne’s, despite the fact it has been released for months, in 1892, was patented in 1893. Here’s the news from a 1893 issue of "The Inland Printer" (note that the patent uses the alternative form of /S).