A sorta taxonomy of pavilions

I. Vertical Center Pole(s)

Many pavilion designs have one or more vertical poles to hold the fabric off the ground, and a bunch of stakes to hold it to the ground; the tension between these opposing forces controls the shape of the tent.

A. Single vertical pole, wall=roof

The simplest imaginable tent design, therefore, would be just a center pole and a bunch of fabric hanging from it, staked to the ground some distance away so there's space inside. Unfortunately, to get any significant room inside the tent you need either an enormously tall center pole (hard to transport and pitch, and vulnerable to wind) or stakes a mile away (wasting a lot of real estate on a portion of the tent that's too close to the ground to use).

B. Multiple vertical poles, wall=roof

One solution to this is to use two or more center poles, often with a ridge pole to hold the shape of the roof in between them. When I was growing up, this design was used for "pup tents" and low-end backpacking tents. The result has triangular ends and rectangular (or slightly trapezoidal) sides at a diagonal (so they serve as both wall and roof).

This shape appears in a number of sources, including King René and a 15th-century Italian treatise on military engineering, De machinis by Paolo Santini (Lat. ms. 7229, Bib. Nat. Paris). Pages 13v-14r of the latter source shows a military camp with twelve circular tents like our first King René tent, "pup tents", and one walled rectangle (looking like a pup tent on top of vertical walls). Bear in mind that this shape can also be attained by an A-frame construction (see below), and in many of the pictures it's hard to tell which design is used.

The Norse Film and Pageant Society has reconstructed such a thing, calling it the Anglo-Saxon Geteld; I don't know what sources they worked from. Cariadoc and Elizabeth (David Friedman and Betty Cook) have likewise reconstructed one based on a 9th-century picture, and it seems to have worked well at Pennsic. Here are two pictures.

(The larger tent housed two children, Rebecca and Bill, while the tiny tent, made from the larger one's scraps, housed their stuffed animals.)

Another interpretation of the geteld is due to Thora Sharptooth and Dofinn-Hallr Morrisson (Carolyn and Greg Priest-Dorman). The sources they examined uniformly showed the ridgepole extending beyond the end the end of the roof, and running not under the roof, but through a sleeve above it formed, apparently, by extending the roof fabric. They're working on a full-fledged Web page about the project, but this picture will have to do for now.

Meghan Roberts (aka Magnunnr Hringsdottir) has built a smaller geteld, of wool fabric with hand-sewn seams. She decided to have the "vertical" posts angled in a little, forming a trapezoid rather than a rectangle, for stability. Click on the picture to see her site with more details on the tent.
Note: the picture at left doesn't show sleeves extending beyond the ends of the roof, but she's adding that.

C. Shoulder between roof and wall

The other solution I know of involves a "shoulder" where roof meets wall at an angle. The challenge, then, is how to keep this angle from collapsing.

1. Side poles
   One common approach is side poles to hold up the roof, as seen in this picture (for which I thank Mike Hannah). You still need guy lines to stakes somewhere beyond the side poles, or the side poles will fall inward, but these guy lines can be at a fairly steep angle so you don't waste a lot of real estate. (Indeed, they must be at a steep angle, or the tent fabric can lift right off the poles in a wind. As a corollary, the stakes must be driven either very deep or at a sharp angle from the vertical, since they're holding down a nearly-vertical guy rope.) The walls can then be hung from the edge of the roof, as in many Berber and Bedouin tents, and either lightly staked, left to hang freely, or even removed to allow more air circulation.

   This approach has several disadvantages. One practical disadvantage is transporting all those side poles. Another is that unless the roof is steeply pitched, rainwater can collect in pools in the vicinity of the side poles; a roof that sheds raindrops nicely may leak like a sieve when rainwater pools on it for minutes or hours. These may explain the third disadvantage: as David Kuijt (Dafydd ap Gwystl) points out, is that there is essentially no evidence that side poles were ever used in medieval European tents.

   

2. Long guy ropes
   Another approach, which we've taken in our first and second King René tents, skips the side poles. The roof is held out by guy lines extending a long way from the center pole(s), but the roof doesn't go all the way out to the ground. As before, the walls can be hung from the edge of the roof, staked or not, or removed completely. This avoids the need to transport lots of side poles, and the (in my experience, common) problem of side poles falling down. The main disadvantage is that the stakes need to be far out. If you have several such tents together, you can save some real estate by crossing their guy lines.

   This approach is taken in a surviving 16th-century (?) round tent. Note that in this picture, it's been retrofitted with an internal structure to set it up inside a museum, but originally it was just a single center-pole and guy-ropes. Note also that the guy ropes attach to the shoulder, rather than going through the shoulder and up to the peak (as in our King René tents). My thanks to Terafan Greydragon for finding this surviving tent in a museum in Basel, Switzerland, photographing it, and Webbing it; thanks also to Tanya Guptill for pointing it out to me.
   

3. Shoulder hoops
   A third approach is to put a rigid hoop at the shoulder. This allows you to use the walls themselves for tension, staking them down firmly but leaving no guy lines at all beyond the walls. This approach is favored by John LaTorre of Dragonwing Pavilions, who discusses evidence for and against it in his article, What held tents up?. In any case, the question then is how and from what to build the hoop.

   

   a. Welded steel hoop

   The first incarnation of Marvin used a segmented steel hoop, but it tended to twist into the shape of a saddle or a Pringle's potato chip (or so I'm told; I wasn't there): some parts of the hoop went above the desired horizontal, other parts went below, and the whole tent lost a lot of volume. To fix this, the builders added a bunch of cross-braces. This made the hoop very heavy, and a pain to assemble and disassemble; it was hoisted to shoulder height by ropes that ran from the hoop, up to the peak, through a dozen steel eyelets and back down along the center pole.
   

   b. Jointed wooden hoop

   Cariadoc's "Conjecturally Period Pavilion", the black and white one on the right of this picture, has a hoop built of wooden dowels, jointed loosely together at the corners. Unfortunately, this design tends to buckle into a zigzag pattern, a more complex version of the Pringle's phenomenon mentioned above, with half of the joints going up and the other half going down.
   

   c. Masonite or rigid wooden hoop

   A small pavilion by Roberto di Milano and Niccola di Sebastiani of Myrkfaelinn (I don't have a picture to post, but you may have seen it at Pennsic -- it's yellow and round, with red flame-patterns on the roof) uses, if I remember right, a Masonite hoop several inches wide, which is enough to keep it from twisting very much, but it's less sturdy, which may be why their tent still uses guy lines extending out from the shoulder.
   Bill Hubbard, a 13th-century re-enactor from Britain, built a very attractive tent with a rigid wooden hoop. So did his friend Andy Goddard, who has a web site about it, including detailed discussion and photos of construction techniques.
   

   d. Flexible poles

   Another small round pavilion we noticed at Pennsic 1997 has a hoop made of several of the semirigid poles used in modern dome tents. (I imagine something similar could have been done in the Middle Ages with saplings.) Its maker, Syr William the Finn, says it has never had any problem with twisting or buckling. I suspect this is because the poles resist bending superlinearly in curvature, so the potential energy with the hoop bent into a saddle shape is greater than that with it in a horizontal circle. Anyway, we found it quite attractive.

   
   

   

4. Wagon-wheel spokes
   Yet another approach to holding out the shoulder is often referred to as "the wagon-wheel approach". Just above head height, a hub is attached to the center pole, with spokes extending horizontally to pockets at the shoulder. This design is used by the Past Tents company, as well as by David Kuijt (Earl Dafydd ap Gwystl), who says Past Tents's design is allegedly based on surviving tents from the 1476 battle of Grandison, but he hasn't been able to confirm the information. His household refers to the design as the "pavilino", after an article they read that consistently misspelled "pavilion" thus. To learn more, read David Kuijt's article about this design.

III. Standing Hoops