If someone misses the boat, historically speaking, it's very difficult to get them aboard, especially if they're dead.
The Zipf curves swings into action, and the canon gets set with a few examples from an era coming to represent it.

Often, the best bet is a son or daughter to campaign for inclusion.

Otherwise, people are more interested in retrieving not forgotten classics (a genre already represented in the canon), but works which are peculiar to their time.

It's not enough for a connoiseur to express an interest, that's unlikely to sway others to jump into action.
Why not publish a revival of the font(s) yourself Owen, if you feel so strongly about it?

I'm quite sure the reason that not all of van Krimpen's typefaces have been digitized have to do with rights. I once read that Monotype still holds the rights to many of van Krimpen's work, like you said, the Lutetia for instance, and only a few fonts have been up for grabs. Which is a pity of course since the typefaces are of the finest quality around.

The same can be said for the work of de Roos, although Gerard Unger explains that some of his typefaces, like the in holland widely used Dutch Medieaval won't get a revival at least from any dutch designers because of it's place in our history. It was over-used, and through that will always end up looking out of date, being a child of it's time.

It looks like Canada Type has a digital version of Dutch Mediaeval. Don’t know how accurate / well-done it is, though.

DTL has a beautiful revival of Romulus shown on their site, but not in their on line shop. I find it strange how many years it's been up without any statement on whether it's finished and for sale.

I've long wondered about why there's no digital Cancelleresca Bastarda. It seems that both in his lifetime and now his types have suffered the fate of being much admired and little used. How much this is because of failings in the type and how much because of failings in the world, I don't know.

I doubt rights are stopping anyone from reviving a van Krimpen design; there are plenty of capable designers living in the US who need not worry about typeface copyright suits. I think it’s more likely the historical and stylistic issues mentioned above. No matter how great or downright gorgeous the type might be it still needs commercial value to justify the effort of a proper revival, perhaps we just haven’t stumbled on the era when a van Krimpen revival has that commercial value. That doesn’t mean that such a moment will not come, or that the right designer could not just commission a revival for the right client (*nudge*).

Thanks very much for your insight, guys. In response to Nick, I would first like to pardon myself because I don't want to sound like I'm arrogantly sitting here ordering Blokland or De Does to go start working on Spectrum or whatever. I certainly agree that I won't be "swaying anyone into action" ;) I'm merely curious as to why these faces have been in large part left behind, considering the care and mastery that is evident in their craftsmanship.

I would feel so lucky to be able to do some of these faces justice; it would, however, be the peak of arrogance for me to attempt to revive any of them. Monotype's lame cut of Spectrum is evidence of what happens when the job isn't done right. I have never designed a typeface. I work with type, I admire type, but quality type demands more than I could hope to give. That's why I would leave it up to TEFF or DTL, two shops turning out consistently impeccable digitizations. It would be so bitter to work with a digitization of any of these faces that doesn't do them proper tribute. Far worse than not working with one at all. That's precisely why I wouldn't touch cut by a guy whose other work looks like this, and why I find Monotype's digitization of Spectrum upsetting.

Of course you're right, people often have very little interest in looking back. Examining some of the samples of De Roos Romein, which demonstrate such gorgeous notan, I just can't understand why there has been such collective apathy regarding this type, and Romanée, and Cancelleresca Basterda, et cetera, et cetera.

The Romulus situation, William, is simply bizarre. Maybe DTL is waiting for the proper companion italic? But really. This is what I mean. It's just weird.

The commercial point, James, is clearly valid, but unsatisfying, of course. You mention that US designers needn't respect EU copyright concerning typefaces? How's that? I'd be curious to learn more about that. Arthus, Monotype is a British company, I believe, but I would expect its presence to reach the states, as well, as you imply. How does this work? I still think rights are the most likely explanation.

Still, foundries make typefaces, some more beautiful than others, and it seems to me odd that foundries would ignore some of the best type that's sitting right in their laps, choosing rather to "build for the present," which doesn't mean very much to me because many of these designs strike me as quite timeless. I'm certainly not waiting for another slab serif or OTF bonanza. (OpenType is fun, and awesome, and completely necessary for digital, but it's become a bit of a fad. I'd prefer that the time go into the letterforms, not into 130493041 alternates.)

I would think that it would be less work for, say, DTL to do a JvK revival than do an original design. William, your question as to whether the current reality is a product of "failings in the type" or "failings in the world" seems to me quite pertinent. If you look at TEFF, though, a foundry that I have so much respect for, you'll find a small shop churning out consistently top-notch faces, some of them revivals. The recent "Geronimo" (a fun font, tastefully hinting at the Rococo style in its ornamentation, it seems) is a wonderful revival by the great Mário Feliciano. Here's one face, one revival, squeezed out. I think that the JvK/De Roos issue might just be a matter of time as well. If they are ever liberated from metal, they should come out right.

I hope.

DTL Romulus is a rare beast, but its existence is beyond doubt.

Owen wrote: Maybe DTL is waiting for the proper companion italic?

That would be like Waiting for Godot.

That's certainly how its seems.

That Canada Type Dutch Medieval looks pretty good.
What do you think, Owen?
How come you missed it?
To tell the truth, it didn't occur to me, but then de Roos really isn't on my radar, I'm much more familiar with contemporary Anglophone historicism.

I did have a serious point to make, which is that there can be no question of "awaiting" a proper italic for Romulus, for the simple reason that such a thing does not exist.

As we know, Romulus was designed and cut at Enschedé in the early 1930s as a lone face. What prompted Stanley Morison to advise Van Krimpen to draw a sloped roman, I for one cannot fathom, but at that point its fate was sealed.

Fortunately, in 1938 Monotype gave Van Krimpen the opportunity to revise his design. He drew a fresh italic and christened the pair Haarlemmer, now available as a faithful digital revival from DTL. So there you have Romulus, right under your nose. Similarly, Romulus Sans was never completed, but Frank Blokland's Haarlemmer Sans is a most worthy successor.

The unique designs whose absence should pain us are Cancelleresca Bastarda and the Romanée italic.

As for De Roos, not only is there Canada Type's Dutch Mediaeval - his Romein and Cursief are right there as well: http://Roos.

There is some information in this Typeradio episode from 2005: Frank E. Blokland reports how DTL obtained the Jan van Krimpen collection, and talks about revivals in general.

>So there you have Romulus, right under your nose

That's not accurate. While both Haarlemmer and Romulus are closely related designs, they are different. Haarlemmer is more robust, has old style stress and serifs, and has shorter descenders. Romulus has more vertical stress and transitional serifs, is more fine and elegant, and has longer descenders.

Yes, if you ask me DTL would make Romulus usable by adjusting the weight and proportions of the Haarlemer italic so that it is a suitable mate to Romulus. I agree—and I think even JvK did, if I remember rightly—that the sloped roman is a flop.

Romulus is one of the most beautiful romans ever, so it is a pity not to have a usable version available.

William, naturally you are correct. I only meant to indicate the common genealogy of Romulus and Haarlemmer, which had not been broached. I do wonder at Van Krimpen's reasoning behind the revisions made between the two. Perhaps you might enlighten us?

It is very difficult to make a "proper" revival.
So much depends on reinterpretation, going from letterpress to offset (or pixels).
I would say that there are many classics which have still not had a proper digital era reinterpretation, such as Bembo, Baskerville, and Perpetua. Even the mighty Bodoni is still wide open for reinterpretation.

Added to which is the danger that if one does an authentic revival, capable of reproducing a facsimile page, it will add nothing to the pot, and just look like some boring old antique. If one adds a twist, creating a dialectic, then it is not quite so proper, and may well be considered a travesty by the cognoscenti.

Don't overlook this post for high-quality scans of the De Roos Romein, from a Dutch specimen book.

The development of DTL Romulus takes almost 14 years now, but a delicate type like this one has to ripe like a good wine. There are actually more weights/styles of the different versions (Poster, Display and Text) made than are shown on the DTL web site and the current state of DTL Romulus is fine but not perfect yet. So, it is like a Grand Cru at the moment, but it is not a Grand Cru Classé yet. It misses the crispness and sparkling that characterizes DTL Haarlemmer, especially in the sloped roman still. So, some work to do here before it will be released and as always we take our time at DTL.
 

 
Although related when it comes to proportions (like most of JvK’s types), the handling of the contrast-flow very much differs in Romulus in comparison with Haarlemmer. For me Romulus is clearly an attempt by JvK to design a typeface with a ‘Baskerville-like’ contrast-flow (‘expansion’), although it also shows characteristics of ‘translation’ with an almost flat angle.
 
Some time ago I started the development of an additional ‘real’ cursive for DTL Romulus, which will be released under the name DTL Remus. After finishing my PhD thesis this summer I will proceed working on it.
 
Most of the trademark rights on JvK’s typefaces are hold by Monotype and some reproduction rights belong to the heirs of Van Krimpen. For Haarlemmer and Romulus I made arrangements in the past with both Monotype and JvK’s son Huib van Krimpen and for Sheldon with HvK (the contract with OUP ended a long time ago). The drawings JvK made for Haarlemmer are in my possession.
 

 
DTL Sheldon is in production for some time now and we have serious plans concerning digitizing Spectrum and Lutetia also.
 
From the DTL web site: The DTL Van Krimpen Project dates back to Autumn 1986, when Dutch Type Library founder Frank E. Blokland met Jan van Krimpen’s son Huib van Krimpen (1917–2002) for the first time at the ATypI conference in Basel. There in Switzerland a plan arose for publishing facsimiles of all the typeface drawings by JvK.
On the 23rd of September 1986, shortly after the conference, Huib van Krimpen sent an extensive letter to Blokland together with the Dutch translation he made of his father’s Memorandum to Monotype from late 1955 / early 1956. The idea was to add the original English version of the Memorandum to the facsimiles.
At that time the Berthold foundry was considering to produce Romulus and Spectrum and Huib van Krimpen suggested in the letter in question to typeset the Memorandum in one of these two new versions.
 
Below three parts from HvK’s letter in question (for those who read Dutch).
 

 

 

 
Although under the American law it is possible to circumvent some rights, like on the design itself, in my opinion the type community should acknowledge and protect all rights, irrespective the meshes of the law. For a decade DTL owned the rights on the Tetterode / Lettergieterij Amsterdam font collection before these rights were transfered to Linotype. So, Linotype is the party now to contact for the designs by Sjoerd de Roos.
 
Roughly fifteen years ago I wrote the following text on the backgrounds of (DTL) Haarlemmer.
 
 
(DTL) Haarlemmer
 
In his Memorandum to Monotype Jan van Krimpen makes no bones about his disapproval of revivals: ‘It hardly needs saying that I am no supporter of copying or even adapting his­torical type faces. Not, it should be understood, because I see them, indeed fear them as competitors of today’s designs; but for no other reason than that they are neither flesh nor fish’.
The most striking aspect of this comment is that Van Krimpen did not consider revivals to represent competition for contemporary designs. Here he meant the design, the form. But revivals may have presented competition for Van Krimpen and other designers in a different way: investment in the production of revivals by type-setting machine-firms, brings little benefit to living designers.
While the quotation from the Memorandum to Monotype leaves hardly any room for doubt, Van Krimpen had been rather less opposed to revivals some twenty years earlier. In the years 1937–1938 he advised on the production of Monotype Van Dijck. This was a revival based on a type design attributed to Christoffel van Dijck, which had been used in a work by Joost van den Vondel in 1671. Only when the project had been completed did Van Krimpen disassociate himself from it. In the appendix to A Tally of Types (Cambridge, 1953/1973) Netty Hoeflake wrote: ‘Whatever he did for this series he did “with love and application”, yet in the end he was “more opposed to it” than ever before, and he called the roman “a doctored adaptation of a seventeenth-century Dutch face”’.
Ten years before the production of Monotype Van Dijck, Van Krimpen had made direct use of historical material for another typeface. The basis for the production of Romanée in 1928 was also a typeface attributed to Van Dijck. The italic had been preserved as a foundry type, the roman, however, was only available in an Enschedé type specimen. Van Krimpen designed a roman based on Van Dijck’s letter. Elements of the seventeenth-century model were employed in the new typeface. In The Work of Jan van Krimpen, A record in honour of his sixtieth birthday (Haarlem/Utrecht, 1952) John Dreyfus wrote: ‘A comparison between Romanée and the reproduction of the original “Kleine Text Romein No. 2” will show that many characteristics of the latter have in fact been retained’.
 
Revivals always formed a substantial part of the typeface assortment of the Monotype Corporation. In fact, before Stanley Morison (1899–1967) started his font production program for Monotype, the supply of this typesetting machine firm consisted entirely of revivals such as Monotype Plantin (1913), Caslon (1915) and Bodoni (1921–1928). These typefaces had been produced in response to market demand. At first Morison’s program also supplied revivals. But what made these typefaces different was that they were designed to anticipate demand. Producing new, twentieth-century type designs was the next phase. Introducing A Tally of Types Morison wrote: ‘The original twentieth-century book type might come in due course. But certain lessons had first to be learnt and much essential knowledge discovered and recovered. The way to learn to go forward was to take a step backward’.
Once his type library of revivals became a financial success, Morison was able to try something new. He appointed a contemporary designer to develop a new book face. This was Eric Gill (1882–1940) and the result was Perpetua (1929). In addition to new typefaces, such as Romulus and Spectrum by Jan van Krimpen, Monotype continued to produce revivals such as Erhard (1938) and the aforementioned Van Dijck. Clearly there was still a demand for historic designs, despite the new typefaces.
And today, in the late twentieth century, the demand for historic designs continues. Following a period of nearly five centuries of relative stability, the technology of typesetting and printing has changed significantly. Yet many of the typefaces now available in digital format are as old as the hills. Popular letters with names like Bembo or Garamond originally date from the fifteenth and sixteenth century respectively. Perhaps it is inevitable that the past continues to influence events in a trade that is governed by convention. Even the most famous twentieth-century typeface, Times (New Roman), has its origin in sixteenth-century punches.
 
The design of Haarlemmer
 
Given Van Krimpen’s distaste for revivals, the re-drawing of Haarlemmer for digital use was a risky enterprise. Especially since the designer had been dissatisfied with the original, unfinished, pre-war production of the typeface. Not that the type lacked quality, which would have been inconceivable for a Van Krimpen design. But there were considerable technical handicaps at the time. Furthermore the outbreak of war prevented him from perfecting the design.
The Vereeniging voor Druk- en Boekkunst (Society for the Art of Printing & Books) invited Van Krimpen to design a new typeface for an edition of the Dutch Authorized Version of the Bible (Staten Bijbel) in 1938. In the design, however, Van Krimpen’s freedom was severely limited. In The Work of Jan van Krimpen John Dreyfus noted: ‘The type was designed to work on an existing Monotype keybar layout in order to keep down costs of production’.
Normally, when producing a type design for Monotype, a layout was chosen so that the letter could undergo a minimum number of adaptations. For other type designs by Jan van Krimpen, such as Lutetia and Romulus, the Monotype drawing room went to considerable lengths to adjust the typefaces, originally designed for foundry type, to fit in their unit arrangement. But this always involved compromise. In his memorandum to Monotype, Van Krimpen described the problems that accompanied this kind of production. He declared himself in favour of a typeface within a unit arrangement. But with the following proviso: ‘As I have said, no designer should try to make it on an existing unit arrangement that does not correspond with his own particular rhythm’.
In all probability, the design for Haarlemmer was not only based on an existing keyboard layout but also on a fixed unit arrangement. Van Krimpen was forced to work with an arrangement that clashed with his own personal rhythm. From his notes to the translation of Jan van Krimpen’s On Designing and Devising Type (New York, 1957) Huib van Krimpen already suspected this in 1990: ‘Has the Monotype Corporation tried to find the best fitting unit arrangement on the basis of a sketch or has the design been made on a previously given grid? Looking at the h, n and u that are too narrow, the latter is most likely’. His supposition was confirmed when he brought the original Haarlemmer drawings, once thought lost, from England in 1992. The widths of the drawn letters have all been neatly divided into units and inked in.
The production of Monotype Haarlemmer was never finished. ‘The first trial sheets of Haarlemmer were pulled in June 1938. Further work on the design was done at the Monotype works in England during the early months of the war, but all plans for the Staten Bijbel were abandoned after the invasion of Holland’, John Dreyfus wrote in The Work of Jan van Krimpen. Dreyfus was not convinced of the qualities of Haarlemmer. ‘As soon as the first trial sheets were pulled, the faults of the type became disconcertingly clear. The roman was far too dazzling, and the italic a good deal too wide. There was an uncomfortable difference in colour between the roman and the italic, and even within the two alphabets the round letters tended to show up lighter than the perpendicular letters. The stiffness of the design is partly due to restrictions imposed by drawing a type to fit an existing keybar layout. Other type designers have also found that a preoccupation with mechanical limitations frequently leads to an æsthetic failure. It is better for the designer to draw the letters as he wishes them to be, and to make modifications only when the engineers show that they are essential for the machinery involved’. Unfortunately, Dreyfus fails to take account of the fact that Van Krimpen had no option than to design Haarlemmer within an existing unit arrangement.
Distancing himself from Dreyfus’s ideas, Walter Tracey (1914–1995) deals with these limitations when discussing his solution for this problem in Letters of Credit. A View of Type Design (London, 1986): ‘I think the design had more promise than John Dreyfus’s account of it suggests. Apart from an awkward W, due to an existing machine layout, there were no disturbing letter shapes. Certainly the narrow h, n, u and rather wide o and c were out of proportion and needed revision; and if the italic lowercase could have been a little narrower it would have seemed to gain weight and would have harmonised better with the roman. The design would then have become an effective book type’.
Huib van Krimpen, in his remarks on On Designing and Devising Type, came to the following conclusion: ‘Probably, the truth lies somewhere in between. Dreyfus’s (JvK?) objections and remarks are well-founded in every respect, but Tracey’s views are no less just. Only, his suggestions are impracticable within the limits of the existing unit arrangement’. Bearing in mind the shortcomings of Monotype Haarlemmer and the criticism of the design, there appears to be little justification for a digital revival of this typeface. Nevertheless, there were good reasons to breathe new life into it. Firstly, the design, the concept, is good. As Walter Tracey has already remarked, the fact that the Monotype version had its faults does not mean that the typeface had no good qualities. Secondly, the restrictions of the Monotype unit-arrangement system no longer apply. The widths and therefore the shapes of the characters can be adapted at will. Thirdly, the italic of Haarlemmer is an expanded type, all other Van Krimpen italics being condensed ones. In Letters of Credit Walter Tracey suggests the following concerning Spectrum italic: ‘The italic is a distinguished design in itself, but it does not quite harmonize with the roman. A foreign phrase or a longish book title in a page of Spectrum becomes over-emphatic and the texture of the page becomes patchy’.
In essence, Haarlemmer is the prototype of Spectrum. The differences between the two designs are quite significant, showing most obviously in the capitals and figures. As far as the widths of the characters are concerned, the italics are antipoles. Haarlemmer’s unique wide italic was probably the result of the unit arrangement Van Krimpen had to employ. So, while this restriction ensured that the project failed it also produced a unique design.
 
The production of a digital Haarlemmer
 
The basis for the digital Haarlemmer was the set of Jan van Krimpen’s original drawings of 1938. However, the quality of the original Haarlemmer was too poor for the purpose. Moreover, there was no reason to take the restrictions of the unit arrangement system on board. The original drawings were not directly digitized. First, they were interpreted, so that a typeface emerged that corresponded as much as possible with Jan van Krimpen’s original concept. In fact, this procedure is comparable to the way in which Van Krimpen’s foundry types were produced.
Jan van Krimpen worked as a typographer and advisor with the Joh. Enschedé en Zonen at Haarlem from 1925 until his death in 1958. Most of his types were cut in steel by the Enschedé punch cutter P.H. Rädisch (1891–1976). Under the auspices of Van Krimpen, Rädisch interpreted his drawings and made the required modifications for the different point sizes. Van Krimpen’s ink design obtained its definitive shape in Rädisch’s punches. Later, these punches served as the basis for Monotype. The original Haarlemmer drawings were similarly interpreted to ‘cut’ digital ‘punches’. Alas, Van Krimpen’s hand was no longer on the tiller, although a thorough study of his work made up for his absence.
 
The first step in the production process was to photograph the originals. These photos were digitized on a flatbed scanner. The scanned letters were separated, and lightly retouch­ed. The Haarlemmer typefaces which had been provided with outlines through Linus M were placed on the exact widths as in the original drawings. Then PostScript Type1 fonts were generated from the Ikarus formats. This made it possible to typeset on the computer with the authentic Haarlemmer typeface designs.
The Haarlemmer contours were given a face-lift. The roman was revised especially thoroughly. Apart from weight and contrast, form and adjustment were considerably improved. The unique, expanded italic required fewer corrections. At the end Medium and Bold weights were added.
 
FEB

Well, Frank, I think you've certainly answered the Van Krimpen question. What a rich and fascinating illumination of Van Krimpen type history and the wonderful part DTL has played in its reformulation. The point you make about revivals, and about JvK's opinion of revivals, is important, I think. The transition from metal to digital insists, it seems, that the design be reformulated for what is almost a different medium. The information you've passed on concerning Haarlemmer elucidates that. Thank you for clearing the Van Krimpen issue up.

@Christopher, thanks for the link to the De Roos specimens. I had them already safely downloaded. I want to revisit the more complex De Roos issue when I have more time.

Happy to hear of "Remus," which as I said I agree is needed to make Romulus usable. Personally, I don't think that a revival of the sloped roman is needed, as it is a poor companion to the roman, being too wide among other things. Perhaps progress has slowed because the sloped roman is inherently a bad idea, and there is no way to rescue it? Just a thought. But I don't see why if you have created a good companion italic it should hold back a release of the roman and italic.

On the issue of revivals generally, developing great design ideas from the past into a contemporary face, with appropriate changes for modern technology of printing or screen is I think a very worthy undertaking. But I am of the opinion that pursuit of authenticity is a trap that can cripple the usability of a design. And the point of the desirability of revivals, and trap of authenticity I think does apply to Van Krimpen's faces.

As Tracy argues, in spite of their beauty, there were problems with practically every one of Van Krimpen's typefaces that got in the way of their usage. Even with Spectrum, the beautiful italic is too narrow as a companion for the Roman, though it is usable. I really wonder whether if Van Krimpen had been able to design as we do now, being able to instantly see how all the glyphs work together, whether he could have avoided these problems. At the time, did his reputed cranky personality get in the way of effective collaboration?

Anyway best of luck on this difficult and worthy project.

Good insight Bill.
Think of W.A.D. struggling with his "shrinking glass".
I spent years trying to imagine how the letters I had drawn would look as type, Fontographer was a godsend in font development, multiple feedback looping of DTP a quantum jump, transcendental.

Owen, late to the discussion, but just stumbled on this thread. I share your sentiments regarding JvK's types, but at the same time I'm somewhat glad the good ones haven't been widely available. Romanée (which hasn't been discussed much in this thread) and Romulus/CB are two of the finest types of the 20th century, and I'd hate to see either of them used for menus or paperback novels.

I've redrawn and developed a revival of Romanée, which is now complete, including the roman, italic, small caps and swash caps, but I have no desire whatsoever to release these fonts. I attempted to touch base with TIFF on this front, but never heard back from them. I did make contact with Fred Smeijers, who worked on the TIFF version of Romanée, but he had no information on if/when the TIFF version would be released. My plans for my version are to print a specimen letterpress from polymer for the time being. I've also recently purchased a good amount of foundry Romanée from a retiring printer in Holland, but it is only the roman. Thus, I may, eventually, cut new matrices of my italic/small caps/swash drawings from which to cast new metal type to accompany the roman. That's a long term plan, and the type would only be used for projects from Greenboathouse Press, but it is my goal.

With apologies, I meant, of course, TEFF.