
Apart from two A-frame tents, the 9th c Oseberg grave held some wooden framework (C55000/181) that is often interpreted as a third tent, refferred to as a «wall tent» or «house tent». Let us have a look at what is said about this structure.
In his first volume on the Oseberg find, published in 1917, Anton Wilhelm Brøgger gave a short description of the structure based on the unpublished 1904 excavation reports, drawings and photos by Gabriel Gustafson and his assistant Haakon Shetelig. After cataloging the find together with Brøgger in 1920-26, Grieg folloved Gustafson’s reconstruction in his more elaborate discussion of the structure, published in 1928. The structure was repackaged and photographed in 2007-09.
Brøgger’s initial description

The structure was found in the middle of the foreship. According to Brøgger It consisted of two rectangular frames, each made from four beams mortised together. The beams were unevenly heaved, some partly and some fully rectangular. Interpreted as related to these frames were two crossbeams laying across the frames. In addition were two pairs of simple «vindskier» (or rather rafters), a ridge beam and two beams that no longer could be fitted into the puzzle. In his view the remains pointed to some kind of disassembled house, c. 3 m long by 2,5 m wide, 2 m to the ceiling and 3 m to the top of the roof. He noted that the mortises in the frames had been secured by nails. This excluded, in his view, the possibility that this was a tent frame made for daily dismantling. In stead it would have been a more permanent building. (Osebergfundet 1: 68-69)
Judging from the sketch above, Brøgger placed the two mortised frames vertically, forming opposite walls, with two beams acting as joists binding the two frames together. Above this he placed the rafters and ridge beam. He placed the two unidentified beams as framework for a ceiling.
Grieg’s expanded description

Let us present Grieg’s analysis following Brøgger’s setup:

One of the two rectangular frames consisted of a lower beam (191A) 3,40 m long, c. 6 x 4 cm with obround mortises near both ends secured with nails. No ornament. Formerly tenoned into one of these mortises was a fragmented but originally possibly 1,90 m long post (211F) with c. 4 cm square crossection and remains of tenons in both ends. Another post (188H) c. 4 cm square, 1,9 m long and with remains of tenons in both ends was also considered to have been attached to the lower beam. It had line ornament on two opposing sides, not going all the way to the end on one side. On top of the posts were an upper beam (E), 3,45 m long, roughly the same shape as A but with ornamental lines on all sides.
The other rectangular frame consisted of a lower beam (possibly 210C [nails from 210?) fragmented, no nails mentioned but otherwise as A. To this a post (211B) had been tenoned, c 4 cm square and 1,9 m long. Line ornament on tree sides. Another post (238D), fragmented and with remains of tenons in both ends, were 1,9 m long and c 4 cm square. Also with line ornament on three sides. On top of these posts were a fragmented beam (187G) 3,4 m long, square 4,5 cm at the middle and widening to 5,4 by 4,5 at both ends with obround mortises near both ends. No ornament.

Joining the two frames were two crossbeams. One (224N) was fragmented but might have been c 2,21 m, rectangular 5 cm broad and c 3 cm thick. Rectangular lapping tenons at both ends. No ornament mentioned. The other (252O) were fragmented but all parts remained and added up to 2,21 m. It had retangular crossection c 5 by 4 cm. Line ornament on one broadside. Rectangular lapping tenons near both ends.
Four rectangular beams 212 (1 and 2) and 307 was presumed to have been two pairs, each pair placed with one end crossing each other and the other fastened to the framework’s crossbeams as «vindskier» (rather rafters). All were supposed to be 1,53 m long and had line ornament on both broad sides. Lower end were hewed on the bias. Upper end had remains of a half circular hole. One was 5 by 2,8 cm at the middle and 5,5 by 4,5 cm at the lower end. The other was 5 by 2,5 cm at the middle and 7,2 by 2,4 cm at the upper end, having at the upper end broad side an iron rivet with washer. Where the two half circular holes met the ridge beam was probably attached.

The ridge beam (213X) from ash with round crossection 5 cm at the centre and 4 cm at the ends were fragmented with 3,27 m remaining, brobably matching the 3,4 m long beams at the sides.
Of the two beams deamed uncertain by Brøgger, (233 K or L) was fragmented with 2,23 m remaining, had half round crossection 5,9 by 3,6 cm in the middle and 4,7 by 2,3 cm at the remaining end. A nail remained in one end. Gustafsen placed it as a middle crossbeam in his reconstruction. The other beam (228P) was 3,4 m long, as opposed to the other parts it was from fir or spruce. Round, 4,5 cm at the middle and flattened from one side at both ends were nail holes were bored trough. No ornament. Grieg noted that it must have been nailed to another piece of wood. Gustafson placed it as a crossbeam in the ceiling.
Grieg mentioned a few more pieces not discussed by Gustafson or Brøgger: 233 a was the fragmented 1,22 m remains of of a flat four sided fragmented ash beam. If this had originally been a crossbeam it would have been 2,21 m. It had rectangular tenons at both ends and a nail hole 1 cm behind one tenon. One side had line ornament. 233 b was 75 cm of remaining fragments from an ash beam, rectangular crossection c. 4 by 2,2 cm. It expanded in width and had a hole near one end. Line ornament. a and b might have been part of the same beam, widening at the middle and at both ends. [Possibly this, now digitally labeled as 233 K or L] 211M was the 2,20 m remaining of a fragmented round ash beam, 4,8-5 cm wide at the center, thinned from one side towards each end. Holes were bored trough near both ends where missing nails had been placed.
A bed canopy? Or something else entirely?
Neither Grig nor Brøgger explained properly how the presumed rafter beams were secured to the rectangular frames and crossbeams. We are left with remains of nails and information of rectangular tenons to make sense of it all. Those of the «rafters» 212 and 307 which seem to have the «upper» ends intact have quarter circular cutouts rather than half circular as stated.
The lower ends of the «rafters» are not consistently placed in Gustafson’s sketch. While Grieg mentioned four sided tenons at the lower ends, none of the beams in the structure are described as having corresponding mortises. Thus, it has not been shown with any degree of certainty that the presumed rafters and ridge beam were part of the same structure as the rectangular frames and crossbeams. The same goes for the remaining beams, and it is a good a guess as any if some of them might’ve been used as diagonal bracing.
It should be noted that the measurements of 1,9 m high by 2,21 m wide opens up for the possibility that the frame with or without the «gabled roof» would fit within the large tent as structure for use as a bed canopy or similar.

Remains of tent cloth?
At Oseberg they found a pile of cloth, ropes and fastening pins that Gustafson thought were the remains of a tent, while Brøgger had a hard time accepting that wool could be used for tents at all.(Osebergfundet 1: 45, 215)

Litterature
- Anton Wilhelm Brøgger 1917, Osebergfundet 1.
- Sigurd Grieg «Rammeverket», in Brøgger 1928, Osebergfundet 2. p 263-
- Vegard Vike on the Gokstad A frames from oak and Oseberg tents.
- The two Oseberg A framed tents are from ash.
- Lump of possible tent cloth from Oseberg.
- Remains of the framework at Unimus. See also here. Most beams are found under 187-188 G.