The Making of "Penric's Demon" Illuminated First Page: Drafting the Page

by Ariela

This is the second in a series of three blog posts on the making of the Penric's Demon Illuminated First Page art print. Read the first part here.

While I have done plenty of text and illumination work before, this was my first time trying to for the style of a page from a medieval codex. When creating ketubot and other similar commissions, I tend to paint the images first and then calligraph the text in the space I left for it. But for this project I decided to follow the order of operations used to make medieval manuscripts: text first, then images.

Of course, unlike medieval copyists and illustrators, I got to work in pencil on a first draft before moving to create the final piece. (Paper culture is kind of awesome. So is the ability to proofread before you work in ink.)

While the two manuscripts I used as my models featured 40 and 36 lines per column respectively, I was making this as a display piece, not an actual book page, so I decided on just 25 lines per column, or five times five, for a nice, theologically significant number in the 5GU.

The next question was which alphabet to write in? As mentioned in the previous post, I decided on a Blackletter hand to capitalize on the association with Olde Stuff, but there are lots of different alphabets within that family. Unfortunately, the very feature that made Blackletter such a desirable hand for medieval scribes - its compact consistency - made it difficult for me to use here. The consistency means that it is very hard to fudge around if you need to stretch or contract letter width or spacing to equalize lines with different numbers of characters. I quickly settled on a Fraktur variant because it was looser than most of the other versions and would be more forgiving if I had to stretch it a bit to justify the text.

Speaking of justifying the text, that wasn't always nearly so much of a thing as it is now. Unsurprisingly, when you write everything out by hand, in ink, no draft, it's hard. Neither of my two primary inspiration documents use it, though BL Royal MS 20 D I at least made an effort.

Histoire ancienne jusqu'à César, c. 1325-1350 CE
BL Royal MS 20 D i fol 2r
The lines are at least similar widths.

So what did a medieval copyist do when they started a word and realized too late that it wouldn't all fit on the line? They moved to the next line. Some just finished on the next line, paying no attention to the line break in the middle of the word. Some would start the word again from the beginning on the next line (this was the most common practice in Hebrew manuscripts).

In the opposite case, where a copyist realized ahead of time that a word wouldn't fit on the line, sometimes they filled the extra space in with designs. Sometimes they just left the whitespace alone.

Unfortunately, letting the lines vary widely in width, continuing words from line to line, or re-starting words on the next line are none of them arrangements that will really fly in a world that has become accustomed to the magic of computer-based text layout. Despite the comparative flexibility of Fraktur, it doesn't stretch enough to allow for perfect justification. Neither of my primary inspiration manuscripts filled in dead space with squiggles. I decided that, if I could fit three letters of a word on the first line, I would use a hyphen and break the word, as I could expect modern audiences to at least recognize and understand that convention.

Early stage draft of "Penric's Demon" Illuminated First Page, photographed in terrible light on a cell phone.

While laying out the text, I realized that my initial plan of illustrating the bottom of the page with an image of Penric kneeling by the stricken Ruchia would not work. The text on this page doesn't get that far, and while illustrations don't always correspond exactly to the text of the page, that was just a bit too far removed to work conceptually. So I changed the plan and decided to portray Penric following Gans as they ride out from Jurald Court to Pen's betrothal ceremony.

Thus followed much research into horses in medieval manuscripts. Oh, the horses.

Apocalypse glosée, c. 1240-1250 CE, BnF Français 403 fol. 8v 

MS Ludwig XII fol. 47v

I know that horse breeds common to Europe at this time had more arched necks than the horses I am accustomed to now, but looking at those pictures made me want to scream at the riders to ease up on their reigns.

Others, though, just made me want to scream.

Sigenot, c. 1470 CE, Cod. Pal. germ. 67 fol. 15r

Lutrell Psalter, c. 1325-1340 CE, BL Add MS 42130, fol. 163r

L'estoire del Saint Graal, c. 1316 CE, BL Add 10292 fol. 213r

I finally chose these two as my main models for Penric and Gans' horses, though I dialed back the decorations on the tack, as Jurald is an impoverished lordly house. I also did a bit of smoothing of the silhouettes to make them prettier to the modern eye.

Apocalypse, c. 1260 CE, BL Add MS 35166 fol. 8r

Codex Manesse 73 r Zurich, c 1300-1340 CE

I modeled Gans after the rider in the first picture, removing the scales to allow his hand to gesture back towards Pen to tell him to "pick up the pace." Penric was more difficult. He's supposed to be wearing a suit with matching jacket and trousers, but none of the manuscripts I was already looking at depicted anyone in a doublet. The ones I did eventually find were much later, which wasn't a problem with historical accuracy, which is a meaningless concept for a fictional world, but the style clashed with my primary models and consistency does matter in worldbuilding. In the end I just kind of winged it.

With Penric and Gans departing toward Pen's betrothal (or so they think), they needed somewhere from whence to depart. Jurald Court is wooden structure, rather modest compared to Castle Martenden. "Large, sprawling, fortified farmhouse" it might be, but in visual shorthand, that meant that I needed to make it rather simple. Unfortunately for me, my reference manuscripts weren't big into simple structures in their illustrations.

Royal MS 20 D I fol. 6r

Royal MS 20 D I fol. 35r

Royal MS 20 D I fol. 16v

Royal MS 20 D I fol. 22r

Eventually I found a picture of an old, simple tower in a wall and used it as a vague inspiration. I also elected not to use the blues and purples in my reference manuscript, assuming it would become a "Tiffany Problem." ("Tiffany Problem" is a term coined by Jo Walton, referring to the tension between perception of history and historical record. Tiffany was a woman's name in medieval times, a variant of Theophania, but if you name a character in a medieval setting Tiffany people will say it's unrealistic.)

Here are two process pictures of the draft version, which was done on drawing paper with a 2B pencil. 

Photo of the mostly complete draft.

Finished draft.

Once the major pieces were all in place on the draft and worked out to a reasonable degree, it was time to move on to the final piece.

Penciling in the text first.

Full pencil in place.

Starting to ink the text.

Embellishing the initial letters of the paragraph breaks.

Underlayers of the historiated initial. I covered the rest of the paper with waxed paper to protect it.

Painting in the vines.

And then it was done!

We're off next week for Shavuot and then Terri and I are off to WisCon, so in two weeks I will explain some of the symbolism behind my artistic choices.

Go back: Part 1: The Artistic Framework
Go forward: Part 3: Symbolism

The Making of "Penric's Demon" Illuminated First Page: Artistic Framework

by Ariela

This is the first in a series of three blog posts on the making of the Penric's Demon Illuminated First Page art print. The other parts in the series can be found here: Part 2: Drafting The Page | Part 3: Symbolism

In making an illuminated manuscript from a fantasy world, there were a lot of decisions to make about the aesthetic underpinnings before I could even get started on the art and the calligraphy. This post will go through some of the major factors I had I consider before setting pencil to paper.

Choosing References

I am <understatement> fond of research </understatement>. So, when tackling any illumination project, I like to ground it by drawing on specific historical examples. But I needed some search criteria, particularly Where and When to look for examples.

Penric's Demon is set in the Weald, which Lois McMaster Bujold has said was inspired by Germany. But it is specifically set in the cantons.  A little bit of googling around found the town of Jura in the Swiss Cantons in our world, and I decided to take that as a rough real-world analog.

Time is easier. The Curse of Chalion is a fantasy re-telling of the marriage of Isabella and Ferdinand, which happened in 1469. "Penric's Demon" takes place about 100 years prior to The Curse of Chalion, which meant that I was now looking for a manuscript from 1350 or so.

So now I had my time and place search criteria for reference materials. But I also had one more criterion: for reasons I cannot adequately explain, I was already envisioning this page with two columns of text.

It turns out I didn't find any manuscripts from Switzerland near the French border in that time period that grabbed me, but that was okay - it's a fantasy world, not historical fiction. I settled on Harley MS 4482 and Royal MS 20 D I as my primary models.

Harley MS 4482 f. 76

Royal MS 20 D I f. 11

These weren't the only manuscripts I looked at by a long shot. I consulted many others for reference to individual elements on the page, but these were the ones I used to set up the basic layout of the page and to set the tone for the overall aesthetic.

Adaptational Difficulties

One of the tricky things about this project was that it wasn't just an attempt at figuring out what a manuscript from the Weald would look like and executing it. The finished product needed to be something that modern viewers would immediately identify as a medieval illuminated manuscript. Which is to say, it needed to look less like an actual historical manuscript than like what most viewers think an historical manuscript looks like.

For example, neither of my primary manuscript models are written in a Gothic alphabet. But the stereotype of old manuscripts is that they are in Gothic. So I wrote it in a Gothic alphabet, albeit one that is a bit looser than the standard (I'll go into why I chose that one a bit more in the next post).

It also needed to be attractive to modern viewers. Aesthetic standards have evolved considerably since the medieval period. For example, mermaids are a popular genre in fantasy art and seem to generally sell well with that audience. But compare these two illustrations of mermaids, one by contemporary artist Meredith Dillon, the other two from Bodleian Douce MS 134.  

"Blue Mermaid" art print by Meredith Dillman. You can buy it at MeredithDillman.com. Used with permission.

View fullsize

Mermaids from Bodleian MS Douce 134, c. 1450-70 C.E.

Despite the titillation (all pun intended) of the historical mermaids being topless, which do you think will play better at a convention art show? Aside from not being terribly pretty, it takes a moment to even read the two historical mermaids as mermaids rather than as women who are being swallowed by fish up to their waists.

I knew I would need to strike a happy medium between evoking the historical references strongly and conforming to modern visual vocabulary.

Symbolism

Many of the standards of medieval European illustration and imagery resulted from Christian iconography and biblical allusion. In trying to reproduce recognizable elements from that genre, I needed to make sure that I didn't accidentally include anything that wouldn't make sense in the World of the Five Gods from a worldbuilding standpoint.

A Trinity Knot, for example, is frequently used as a representation of Christian theology, but it would not be an appropriate to use it as a theologically significant icon in a World of the Five Gods manuscript.

View fullsize

Is this the Daughter of Spring appearing at the spring of Limnos? Or the Assumption of the Virgin Mary?
Holford Hours MS M.732 fol. 56v

I also needed to make sure that I didn't accidentally cross any cultural wires. A woman in a blue veil with a halo in a manuscript in the World of the Five Gods would clearly be the Daughter of Spring. But before a real world viewer could identify the Daughter so portrayed, they would first have to unplug that image from their mental slot labeled "Virgin Mary Iconography" and stop to think about the context. (Doesn't everybody have mental slots for medieval imagery?)

Next Monday I will walk through the process of actually drafting the page. Stay tuned!

Read the next posts in the series:
Part 2: Drafting The Page
Part 3: Symbolism

Employing A Sensitivity Checker

by Ariela

Old Economy Steve longs for the days when he could avoid people telling him how offensive his views are.

Sensitivity readers have been quite the hot topic in some parts of teh interwebs lately.

What is a sensitivity checker? When a creative type, in my case an artist, wants to use cultural elements from a culture not their own, you employ someone from that culture to act as an expert guide, telling you things that are important to know, and giving feedback that should help you to portray the culture in question accurately and in a way that is not offensive. This can cover anything from preventing embarrassing errors like those from our Jewish Stock Photography Fail Blog, to the horribly offensive error of making Nazis the good guys in a Holocaust novel supposedly told from the point of view of a Jewish girl (we won't link to that book, but here is a scathing review by Katherine Locke). 

In some ways, it is no different than consulting any other expert so you don't make ignorant mistakes, but here the stakes aren't just your own embarrassment but the possibility of perpetuating oppression of real people. 

I recently had my first serious sensitivity check. Our product release next week will include a picture of a keris, which is a Malaysian dagger with serious cultural significance. That's the sort of thing you don't want to just assume you can just chuck into a piece of art when you don't know anything about the culture surrounding it. Thanks to the wonders of Twitter, we were connected to Jia-Ling Pan, who was incredibly helpful. We highly recommend her, if you need a sensitivity check for anything from Malaysia. (Let us know if you would like to be put in touch; she said she would prefer a referral than a link.)

Here are a few basic takeaways from the process.

1. Pay them.

This should go without saying, but alas, it needs to be said. Giving someone a crash course in the intricacies of one's culture isn't a privilege, it is work. Listening to and answering questions, many of which may be ignorant and even offensive is hard work. Giving constructive feedback is work.

This is true for art and even more true for reading a manuscript, which takes a heck of a lot longer.

Pay them. If one cannot afford to pay someone from the culture one is writing/singing/art-ing/movie-making about to do a sensitivity check, then one doesn't get to play in that sandbox.

This doesn't mean you don't get to create, just that this particular avenue is not available. Lack of money sucks. It sucks for you as a creator, but it also sucks for people whose culture gets trampled over insensitively and are then asked to help someone else make sure they're doing it right for free. Sensitivity checkers need to be paid.

2. Listen to what they have to say, and give them space to talk.

You employ a sensitivity checker because they know things you don't. Sometimes that means answers to your questions, but other times that means telling you the answers to questions you didn't even know should be asked. Give your sensitivity checker some unstructured space to talk rather than framing everything within a Q&A. Your questions are shaped by your own assumptions after all, and they may not translate into the culture you are trying to learn about. You might even wind up getting more material out of it in ways you didn't anticipate.

3. Accept what they have to say.

Sometimes a sensitivity checker will say things you don't want to hear. You have to be willing to go into the check alert to the possibility that you might have to rework part of your project, a lot of your project, or even scrap your project entirely. The earlier you ask, the less likely this will be, so check early and often! Though even that is not a guarantee.

It's never easy to hear negative feedback, particularly when it means losing a lot of work. If you are only interested in affirmation and will come up with reasons why you don't need to take the negative feedback too, don't bother to have a check done at all. 

Mary Robinette Kowal has an excellent blog post about this. Please check it out.

4. No Group is Homogenous.

As in any group, different people will react to the same thing in different ways. Just because your sensitivity checker reads things one way doesn't mean that other people won't have a different take. Having more than one sensitivity checker is a good idea, particularly if the source material is emotionally significant or you are using a lot of it. And even if all your readers say it's fine, there's no guarantee that someone else will not be offended. Which brings us to:

5. Your creation is your responsibility.

If someone else is offended by the thing you made, even if you had a sensitivity checker, the responsibility is yours. It is not the job of a sensitivity checker to tell you how to do something so that you cannot be criticized for it; they're just there to give your their own read, and possibly make a best guess at how others may feel. The mere fact that you hired a sensitivity checker is not a shield against criticism, and if the work offends someone, then it is not the sensitivity checker's fault. It is on you to own up to any damage caused and decide how to proceed, apologize, make amends, fix it, etc.

Coda: Sensitivity checking is not a substitute for #OwnVoices

#OwnVoices is a campaign to lift up the work of people in various marginalized identities telling their own stories. It is not enough to have white people telling stories centering People of Color, or straight people telling stories about LGB people, or cisgendered people telling stories about trans people. No matter how skilled or well-intentioned the creator, they will not get it all. People have a right to portray themselves. And we should support them when they do by patronizing their work.

That said, I am very much not in the camp that says it is never okay for someone to portray a person outside their own identity. That leads to other kinds of erasure and normalizes the idea of homogenous societies. But when we do paint pictures of other people, with words or brushes or songs or cinema, we need to make sure we do it with care and respect. That's where sensitivity checkers are so important.

Find a Sensitivity Checker for Your Own Work

There are lots of people out there who do this work. Writing in the Margins is a great directory if you are looking for someone to keep you from accidentally putting your foot in it.