Dan, if you don't have a copy of Edward Johnston's 'Writing, Lettering and Illuminating' you should have.

Otherwise try hunting through the libraries (you can get 'em back on time, honest) and secondhand bookshops near you. Almost all of the lettering books for commercial artists and signwriters from the first half of the 20thC contained at least one version of the authors' take on the Trajan lettering. My own favourite would be p.29 of 'The Art of Lettering' by Geoffrey Mott, but this material is very commonplace and for less than $20, cheap.

In Albert Kapr's "Schriftkunst" is a 3(!)-page fold-out of it.

Ralf

http://www.fonts.info

Here's a fragment of the inscription from a 1951 exhibition catalogue...

There is a nice book "Letters redrawn from The Trajan inscription in Rome" by Edward Catich. You can also search for Trajan letterformas at the Catich collection http://catich.sau.edu/

I have a couple of images not of the Trajan inscription of poorer quality carving at 16mb if they are any interest to you.

Not much contrast in that.
In certain light, it would have been almost a monoline sans serif.
Like Cleland's Della Robbia. He added a lower case. Cool.

[[Trajan Capitals]]

ebay it.

Don't you just lust after the "M" and (for me!) "P"?

Such gloriously elegant forms (like a 1961 E-type Jag or the current Aston-Martin DB9 Cabriolet): go for it, Dan!

I was at looking around in the Powell's rare books room a couple of months ago and noticed this.
It consists of loose-leaf prints of Trajan rubbings. It didn't seem worth the money to me, but it might be available in some university libraries. It probably wouldn't be something you could checkout from a library though, since it isn't bound. The Goudy book is probably much, much more useful.

Catich is indeed the author you want on this.

Daniel, if you make a Trajan, consider:
1) Having a "V" in place of the "U" (or if that's too much for you, do give the "U" the two full stems).
2) Putting that little triangle thing in the blank space.
3) Not including severely out-of-character characters, like the "@". This would make the font less versatile, but that's the point: it would also make it hold its character better over time/usage. The fact that Adobe Trajan is too versatile causes it to be used by the likes of ISUZU, ruining its focus for everybody else.
4) Calling it Imperium. Or IMPERIVM. :-)

We have enough transvestite Trajans - give us something useful in its difference/authenticity.

hhp

Transvestite Trajans? Considering his family tree, that might not be too far off base :)

So if there are a billion Trajan's available, how about something different like the forms timd suggested? Something useful, but not too useful?

Not having a V or W or whatever gives me just that one less piece of lead to conquer the world with! Oi!

Well, one decision you have to make from the get-go is whether you want to make it regular or "folk" - remembering that these days the latter can be more "useful" than the former!

hhp

I tend to like more elegant looking faces. Thinner weights, sharper serifs, etc. Those kinds of forms don't usually associate themselves with "folk" -types, but if anyone has an idea along those lines, I'd love to give it a try.

I have a book titled Calligraphy, The Art of Written Forms by Donald M. Anderson that made many references to Catich and to Johnston. They were the next set of books that I was going to hunt for, but books on type are slim pickins in used bookstores in Seattle. I'll put them on my wishlist.

Good suggestions Hrant.
I have occasionally thought about producing a more authentic roman inscriptional face, distressed, with the features you mention, and also some OpenType coding that converts numbers into roman numerals.

"OpenType coding that converts numbers into roman numerals."

and the @ singn into et tu?

ChrisL

I see lots of distressed typefaces. Most, not so special. It seems that most people just auto trace or something. Something like Founder's Caslon seems more "authentic" than a lot of the others.

What's the proper way to do a distressed font?

Unintentionally.

hhp

There would be two "distress" aspects to doing a font based on lapidary inscriptions.
First would be the original uneven execution of the letter carvings -- they were painted on by brush first, with that process producing irregularity in form, as did the subsequent carving, which was interpretive (even if it was done by the same artist -- which we will never know).
Secondly, the effect of centuries of weather (on outdoor inscriptions).

And with lots of options so repeat characters are rarer.

Tim

The other marvellous book by Father Catich, still available for a very reasonable price: The Origin of the Serif. Full-page facsimiles of each letter in the Trajan inscription. And a fascinating technical investigation.

BTW Tim, maybe you should point out that the two inscriptions you showed above are about 15 centuries apart.

The Origin of the Serif has excellent plates of the inscription.

Many other books reproduce photos of the Victorian cast that resides in the British Museum. The BM cast has been modified in several ways from the Roman original. For example, a serif has been added to the inside of the right leg of the "A".

I strongly reccomend working from the Catich plates in "Origin".

Bonnie is right about the Victoria & Albert cast of Trajan, and the Johnston pictures posted above are actually after the cast.

Now let's see, what have I got?

How do you like these serifs? More Galliard than Galliard.

Nice 15th century German inscription with Byzantine elements. Not quite what you're looking for maybe, but has some interesting ideas, especially the N.

And this lovely 11th centuty lettering. I began digitising it myself but it would need more time. Sorry the picture isn't better, but notice the Baskerville C in effecerat :) And the high waist-line plus contrast between narrow and wide letters makes it sort of pre-pre-Art Nouveau (or -New Yorker?).

Look at them ligs! How cool would it be to have a Roman font with extensive ligatures? As in, more ligatures than you could shake a stick at?

So here's the checklist for IMPERIVM, the Ultimate Roman:
1) Having a “V” in place of the “U”.
2) Putting that little triangle thing in the blank space.
3) Not including severely out-of-character characters, like the “@”.
4) OpenType coding that converts numbers into roman numerals.
5) Lots of options so repeat characters are rarer.
6) Ligatures, and keep them coming.
7) Possibly different weights representing various states of erosion from freshly brushed, to finely chiseled, to well-worn, to acid-rain.

Anything else?

I would so buy that. (if it was good) You'd need a full day course to learn how to do everything.

Anything else?

It wd be gd to compile a std set of abbrev.s for Engl and code it in.
In Tim's sample, it lks like whenev. they forgot a letr they just went bk and put a line next door II where it shd have been.

We use Mr, etc. in the same way that IMP was used (Imperator) in Marc's example, but we don't put a "macron" over them.

But seriously...Roman inscriptions had lots of abbreviations, because Trajanus hisselfus had a heck of a title, for one reason, and it helped justification, for another. And made for less work for the carver (paid by the word?). If the same principle were applied and fontized, a model would perhaps be 18th or 19th century correspondence, where writers used a kind of shorthand to speed up writing -- or when the end of a line is getting near and it doesn't look like all your words will fit. Another model would be texting today, but perhaps too anachronistic for a Trajan?

I don't know about #5. Ligation and wear are one thing, but free-spirited glyph selection is another. Fine for a handwriting font, but not a formal monumental style, I don't think.

> “macron”

Yes, it would be great to have that, although the variability of its width (like the Arabic kasheeda) seems challenging to implement (which is probably why it didn't survive the transition from writing/lettering to typography).

> I would so buy that.

I should bring this thread to the attention of David Lemon. :-)
He's been a vocal sceptic concerning the wisdom of having
a "U" that's a "V", not having an "@" sign, etc.

hhp

The abreviations in Tim's first sample are a basic late mediaeval set (and in this case even post-mediaeval): nri for nostri, Lateranen for Lateranensibus, pp for pape, etc.

If you like ligs, you'll love the early 12th century. This is from Moissac in the south of France.

Thanks for the correction, MH. It has been so long since I have been in London that I misplaced the Trajan cast.
I love your inscription photos. I find it wonderful to see that those ancient craftsmen ran out of room at the end of the line, just like I do. Hope springs eternal…

Something else: the #2 is not unlikely to cause technical issues, sometimes and maybe even often. It's possible that a viable work-around would be to have a normal empty blank space, but always replace it with a "that little triangle thing" character using OT.

hhp

Not neccesarily "free-spirited" glyph selection, just a wide range of alternates. For instance, on the Trajan inscriptions the letter I has variable heights.

Since there are going to be plenty of empty slots in this font, might as well fill them with thoughtfully designed alternates.

Oh, Bonnie, I hadn't even noticed I was correcting you. I must have read your post too quickly, and seen V&A where you wrote Victorian.

I like your new hairdo §:)

If you're going to replace 'U' with 'V', you'll have to replace 'J' with 'I' as well, but you probably already knew that.

- Lex

Then there's the C and G... But I actually wouldn't link those decisions too strongly, because there are a number of factors at play, and historical authenticity isn't everything (even assuming it can be adequately defined here). One factor is that of chronology, and exactly when the U, J and G were introduced, and by whom; another is that of potential ambiguities in actual names and words, and where you want to draw the line; a third is what people think is authentically Roman (and this is where I think the V/U stands out much more than the other two pairs). You can make the decision to limit the versatility of a font (not out of hooliganism, but out of a hope of helping it maintain its focus) but you can still worry about how it might be used (which is still the point of a font, in the end). For example, BVLGARI works, but IVICY wouldn't. :-)

hhp

An important source: L.C.Evetts. Roman Lettering: A Study of the Letters of the Inscription at the Base of the Trojan Column, With an Outline of the History of Lettering in Britain. London: Pitman, 1938.

Another important source: Walter Kaech. Rhythm and proportion in lettering [Rhythmus und Proportion in der Schrift], Olten und Freiburg im Breisgau: Walter-Verlag, 1956.

This is similar, for example, to what I've noticed about Johnston's Underground face. P22 has an accurate version of it with no lower case bold, etc. Font Bureau's Agenda has 54 styles. Where do you draw the line?

I would think that in this instance, with the desire of making the ULTIMATE ROMAN*, you'd want to keep everything optional. Of course, an intelligent and historically cognizant designer might realize that if you want to have all of your U's look like V's, G's look like C's, or J's look like I's, then all you have to do is type them that way.

So, with this in mind, I hearby proclaim point #1 from my list above null and void.

But still, no @ sign.

*Patent Pending

Wow. Is that an analysis showing the use of the Golden section? Those Romans were sick.

> all you have to do is type them that way.

Of course there's a truth in that, but there's also a danger, the one that I mentioned: the more you facilitate some people's use of the font in a way contrary to its "ideal" character*, the more they'll ruin it for everybody, because the font's associations become diluted in the public eye. On the other hand, having neither a "U", "J" nor a "C" could be said to be too puritanical, too limiting. So maybe a good compromise would be to default in the "archaic spelling" using OT and allow the user to revert to the "Modern" characters, but only proactively.

* I know this seems like a fascist attitude, and I'm all for being nicely surprised by unexpected usage, but I think this really is essentially a matter of long-term functionality. The key thing to me is that we already have highly (too?) versatile Trajan-style fonts, and Dan would be contributing more by being different.

> But still, no @ sign.

Whew.

hhp

What's all this about C and G? OK, J and V achieved complete independence from I and and U only in the 19th century, but the Romans invented (or reinvented) G long before Trajan, as soon as they realised Etruscan letters were not enough for their own needs (since the Etruscans had lost the original sound of Greek gamma, and turned it into C).

Have you had a look at the font Senatus by Prof. Werner Schneider? Its based on the Trajan inscription. Here's a sample of it and some of his calligraphy.

Of course there’s a truth in that, but there’s also a danger, the one that I mentioned: the more you facilitate some people’s use of the font in a way contrary to its “ideal” character*, the more they’ll ruin it for everybody, because the font’s associations become diluted in the public eye. […]
Hi Hrant. Are you still of this opinion?
I agree on what you say (Paul Shaw thought also numerals were out of place in Trajan), but if everything "modern" is removed, we should not have accented characters as well, currency symbols et al.
It seems to me the crucial points among all the ones mentioned are:
1) The [U], which could be kept as an alternative;
2) The addition politypes, where it's appropriate (not excessive, like you said);

* I know this seems like a fascist attitude.
What do you mean by "fascist"? As an Italian, I'm so used to the word that I still can't get what's the dominant feelings associated with it. They are mostly negative, but this doesn't help.

Claudio, yes I still feel that way.
Limiting the versatility of a font, or any element of design, increases its focus. The trick is finding the right balance; pretending that a font can be maximally versatile and not pay for it in some other way is misguided.

> Paul Shaw thought also numerals were out of place in Trajan

Really? Cool. Is this written down anywhere? (I mean besides here. :-)

"Fascism": to me it means telling people how to live their lives.

hhp

Hi Hrant, thanks for the answers.
Limiting the versatility of a font, or any element of design, increases its focus. he trick is finding the right balance; pretending that a font can be maximally versatile and not pay for it in some other way is misguided.
I agree entirely on this, I think it can also be experienced and seen by anyone.

> Paul Shaw thought also numerals were out of place in Trajan
Really? Cool. Is this written down anywhere? (I mean besides here. :-)
No, it was a thing he told me when he visited me in Modena in 2002. I was accompanying him by car at the Train station, and we were still talking about why he did not like the excessively eterogeneous nature of my letter "experiments", and what constituted a typeface's integrity.
But I don't think he would bother re-affirming it, it was not a negative criticism, he considers Carol Twombly's work very well done. A pity it's so widespread and unconsciosly used, I add…

Fascism: I would say "forcing" instead of "telling". Telling it may be, and it generally is, for the good. This is not a correct use of the word to me, anway. Fascism, despite all its negativity in the dictatorial aspect, had good things going on. Anyway, I would use "dictatorial", but this opens another linguistic problem… :P

Certainly the meanings of words vary by geography.

hhp

<note>Because I intend to use some irony in this posting, and contributors to this forum (including myself) sometimes have difficulty in distinguishing between ironic and serious comments, I’ve labeled the various sections to identify my intentions.</note>

<irony>Since font users can’t be trusted to use historically-based fonts appropriately, more needs to be done than just to eliminate historically-inaccurate glyphs (such as U, J, and @). Clearly, users should not be allowed to use historically-based fonts for anachronistic text. For example, when using Imperivm (or similar Trajan based fonts) copy such as TRIVMPHAL ARCH should be allowed, but the open-type font software should screen for and block text such as IVICY FRVIT GVM.</irony>

<deeper irony>On second thought, for true historical accuracy, the Imperivm font should only allow text in Latin. Any other language would clearly be anachronistic.</deeper irony>

<serious comment>The advocates of historically-accurate font use, need to keep in mind that the users of Imperivm (and similar fonts) are NOT in ancient Rome craving the glyphs in stone -- they are using them on paper and electronic devices worldwide in the 21st century, so there is already a built-in anachronism in the use of these fonts. Users should be permitted to use the fonts however they please (whether we like the results or not). All fonts ought to be available with the widest possible set of glyphs (including the dreaded @).</serious comment>

<postscript>Dan, what’s the status of Imperivm? Any samples you could post? </postscript>
 
         ~~~~~~~~~~~~~~~~~
         When going from A to Z,
         I often end up At Oz.

Users should be permitted to use the fonts however they please (whether we like the results or not).
If you allow me, that's my own decision as a designer. If I produce a work, I decide how it functions, and I reflect upon its use with my own choices as a designer and "page planner" in mind.
Whether people will use it according to the functions of its "organism", it depends on how I (as its designer) decide to sell or make it available. It's clear that freedom is largely perceived in an absolute value, but we do not live on an absolute plane, since we have to deal with the sensibility of our fellow men.
So, you will agree that I may propose an intended usage out of assumed expectations, even if this exludes an "@" sign, if this is my cultural intention.

There have been occasions when it was not convenient to find a hammer and I have hammered a tack with a pair of pliers. Their have been cases where I had a pair of scissors handy but preferred to tare the paper by hand. Type is a tool to be used. It can be used badly or quite well, we do not or even should not exclude misuse by those who purchased the tool. Type is not a dangerous weapon like explosives which needs strict regulation. Cast your bread upon the waters but know some of it will sink to the bottom.

ChrisL

Clint, that stance is even more ideological than mine! And this really has nothing to do with any pedantic historical accuracy (I certainly can't be accused of that! :-) it has to do with perception, and its judicious leveraging.

Adults help children do the right thing (because of their experience and knowledge), and everybody is childlike in some contexts. Users need our help (because of our experience and knowledge) whether they realize/admit it or not. This is why Adobe InDesign does not allow fake bold for example.

On top of that, there's actually self-interest involved too. Just because you can sell more copies of a font for the first year because it's more "versatile" doesn't make it smart to sacrifice its long-term viability, which can only come from its focus. Why do you think Trajan has become a parody of itself, and most intelligent graphic designers avoid it?

Chris: we are all responsible for the results.

hhp

Jayan has exquisite taste! Jayan, did you study with Werner when he was in India?

Michael

Claudio and Hrant,

It may be a type designer's preference that certain glyphs not be available. After all, in the case of the @ glyph, the user could simply spell out the word “at," but you know what happens in the real world -- if a glyph is not available, the user is likely to substitute one from another font instead. (If he or she wants an @, they will get one, one way or another.)

It seems better to me that type designers should control the design of all the glyphs that might be used in contemporary communication.

Glyphs are not only a design issue; they are also a content issue. Users should be allowed to say whatever they want, using a complete set of glyphs to choose from. Omitting glyphs because the type designer doesn't feel they are historically appropriate, results in a form of censorship: "You can’t say that using my font."

It seems better to me that type designers should control the design of all characters that might be used in communication, and also recognize the fact that their fonts are being used in the 21st Century (and hopefully beyond). If you create a font in the 21st Century, it should be capable of being fully used in contemporary documents; otherwise the font just becomes obsolete historical font, or worse a font where users mix in other fonts to add the missing glyphs.

In summary, type designers should not sit in judgment as to which of their glyphs may or may not be used once the font is released to the world, nor should they try control whether or not their fonts are used appropriately. (They, are of course, entitled to their own opinions, and are welcome to speak their minds, but they should not try to enforce those opinions on others by withholding glyphs they feel are inappropriate.)
 
         ~~~~~~~~~~~~~~~~~
         When going from A to Z,
         I often end up At Oz.

It seems better to me that type designers should control the design of all the glyphs that might be used in contemporary communication.

I try to put it in a simpler way: if it is a "historical" lapidary alphabet, you do not have email.
If some user wishes to use a lapidary alphabet to compose email, there is a bigger cultural problem that does not even allow to understand what "censorship" means.

otherwise the font just becomes obsolete historical font, or worse a font where users mix in other fonts to add the missing glyphs.
Of course, but that's ignorance. As I see it, it's this aspect that should be addressed…

“...if it is a ‘historical' lapidary alphabet, you do not have email."

Well, I suppose we will have to agree to disagree.

As I said in my original post on this subject, people "are NOT in ancient Rome craving the glyphs in stone — they are using them on paper and electronic devices worldwide in the 21st century, so there is already a built-in anachronism in the use of these fonts." If people want to use Trajan (or similar fonts) for e-mails, that should be their prerogative, not the type-designer's prerogative.

Should Caslon only be used for text about 18th Century England? Is it "wrong" to set an e-mail address in Calson? Should all historically-based fonts only be used for texts relating to time of their creation? I think not.

Glyphs should not be "censored" in any typeface for the sake of "historical accuracy."
 
         ~~~~~~~~~~~~~~~~~
         When going from A to Z,
         I often end up At Oz.

As the Latin phrase goes, caveat emptor.

As a type designer, I can and should make the tool as I see fit. If I see fit to use Opentype features to replace an exclamation point with "IO", or "QUO" or whatever. That's the designer's prerogative. The same could be said for DANATDANGAYLEDOTCOM or whatever.

I could program it to render dan@dangayle.com as DANATDANGAYLEDOTCOM, but the beauty is that the underlying text remains the same. It would still be encoded as dan@dangayle.com in the document, and still useable.

CAVEAT EMPTOR
If you don't like it, don't buy it.

Now, the philosophical debate aside, a few years ago I created a giant poster for my school's photo club using Trajan on a giant photo of some marble. I thought it looked great, but it came at the expense of readability and understandability. THE PHOTO FORVM didn't fly so well, because the people in the club didn't understand what it meant, and the anachronism was totally lost on them.

As to the IMPERIVM font, I've been working on a Roman typeface here. Not quite up IMPERIVM standards, but still fun to design :)

AtoZ: I think we are talking of different things, and I'm not sure I am able to explain.
On the historical accuracy, the solution offered by Dan is effective.
I care about internal consistency, and conciliation of apparent rivalry, and if an alphabet design is based on a conceptual idea it could also propose censoring or altering, or the fostering of a constructive talk.

Probably we did not understand each other because you were thinking of normal typefaces which should address the expected uses, while I was talking more of an artistic approach with different degrees of functionality, since type design is always on a fascinating balance between art and technique, vision and analisys. I enjoy a constant confrontation between the two approaches, and of course my lapidary logic if modern would include an [@], if ancient, not.

piccic: I agree that we have probably been approaching the subject from different viewpoints. As you point out, a type designer must find a "...balance between art and technique..." In that sense, type design is similar to architecture and other crafts that mix art with utility, and there are no firm rules as to where and how that balance should be achieved.

My own feeling is that if type designer is creating a typeface for his own use (or has been commissioned to create a typeface for a private use), then, of course, the typeface need only contain the glyphs that are needed and desired for it's particular use. However, if the typeface is released for public use, then I believe that the typeface should contain all the glyphs that users might reasonably be expected to use.

I am aware that some feel it would be unreasonable to include, for example, a @ glyph in a historically-based typeface even when released for public use, but to return to the comparison with architecture -- if one were to design a new public building in the style of a Greek temple, should the architect refuse to include rest rooms in the building, or air conditioning, since those items where not part of the original design vocabulary in ancient Greece?
 
         ~~~~~~~~~~~~~~~~~
         When going from A to Z,
         I often end up At Oz.

"Anyone have a copy of them?"

Isn't there an entire rubbing at Rochester Institute of Technology or somewhere?

Cheers!

@dberlow

Ooo. Thanks for that hint! I'll start poking around.

Regarding some of the previous conversations, now past the 2 year mark, what would you as the head of major foundry and well respected in the industry say about another version of Trajan? Particularly, if some magazine were to come to you and say that they wanted their own Trajan, how would you approach it?

Obviously in that case the silliness of the opentype substitutions would be gone, and all of the requisite glyphs would need to be created, right?

"...if some magazine were to come to you and say that they wanted their own Trajan, how would you approach it?"

Well, first I'd make sure they were prepared for absolute authority, because that's what Trajan's about. Then 'd go about the usual specificational due diligence to end up with a few letters that fit the Imperial Roman ambitions of the spec. (why no silliness of the opentype substitutions?). Then, I'd take a minute to check and see if I was still the head of major foundry and well respected in the industry and I'd send them a bill. If they liked everything so far, I'd finish the face. I could not say from this point of view, what the contrast would be nor the extension of the R's leg might be into the future.

Cheers!

Well, first I’d make sure they were prepared for absolute authority, because that’s what Trajan’s about.
That's what I was thinking about, and to reply AtoZ: yes, I am generally motivated on a personal level. I guess I'll never be a type designer full-time doing large families: I like to use type as much as to design it, and I think all that had motivated me most is when I don't find what I am looking for among existing typefaces (I recall Jon Barnbrook wrote something similar once).

About the Trajan, what I wonder is if it's better to have more than one model, especially for Capitalis. After all, there are other notable inscriptions from the period, and examples, which may not be so refined as the Trajan column inscriptions, but nonethelss give material for a more comprehensive inclusion.

RE: Opentype
I meant only that the substitutional rhetoric would be done away with. The @ sign would be just that. Not "AT".

RE: Absolute Authority
That's an interesting perspective that I hadn't thought about before. Why else would they have erected such columns, if not to boast about their military prowess.

Perhaps that is why Trajan the typeface feels out of place on many movie posters, because it is used for the wrong reason, and without any thought to the attitude of the typeface.

I'll have to look into that more.

Why else would they have erected such columns, if not to boast about their military prowess.
Hmm… I did not think about it in terms of "military prowess".
What's that suggests you such an association?
The adjective "authority" comes from the "author", which here is the designer (or stonecarver), the authoritative people which seriously and honestly study some matter, it's the "creator", anyone to which you could turn to if you need to grade your work and improve.

Negative associations with "authority" are more a modern phenomena, related to the apparent disillusion people nurture towards the conduction of a community, state et al. and towards the possibility of peace, refusing the idea of war, which is sometimes unavoidable.
It's a negative mood which sets a "pessimistic" view upon one's personal improvement and upon a charitative attitude which should be universal, and not just "tolerance".
Of course, the Roman model may not be the better to convey the inherently positive concept of "authority", but the forms are beautiful, associated with strenghth and grace, and – at least to me – friendly and not intimidatory.