The Collector Earl’s Garden, Arundel Castle, Part 1: Green Oak and Arches

The Collector Earl’s Garden at Arundel Castle is the masterpiece of Julian and Isabel Bannerman. It is an ‘attempt to bring to life the garden belonging to Thomas Howard, 14th Earl of Arundel, and his wife Alathea at Arundel House’.[1] It is full of fascinating inventions, and works as a series of spaces, although I seem to recall Roy Strong writing that it was a bit to sophisticated for the average garden goer and needed to be used for, say, the performance of Shakespeare plays.

Oberon’s Palace, based on a stage design by Inigo Jones, is perhaps the finest garden fabrique of recent times. Here, however, I want to concentrate on the two arches that face each other on two sides of a square lawn, on the third side of which is Oberon’s Castle, and their construction. One is based on an Inigo Jones design that in turn owes much to Michelangelo’s Porta Pia. The Porta Pia is now the model for the inside of the Montacute Gateway Arch (somehow Serlio got pushed to one side), so the Bannerman’s take on this design is of particular interest. The other is rather simpler and is apparently based on an archway that appears in the Daniel Mytens portrait of Thomas Howard’s sculpture gallery (National Portrait Gallery, on loan to Arundel Castle) but I can’t see it in reproductions: it must be something tiny in the distance. It is also based on a drawing by Jones identified as the ‘gateway at Newhall, Essex’.[2] This the Bannerman’s call the ‘Inigo Gate’.

These buildings, like all the structures in this garden, are made from green oak. What takes a while to register is that they are made from whole tree trunks, some of which would have been huge. This is visible in the Inigo Gate (Fig. 1). The piers seen from the front are clearly made from a single slab, most clearly visible on the right (Fig. 1a). On the left the grain is more complex (Fig. 1b). These slabs do not seem to be radial – the core is probably on the inside. These slabs go half way through the thickness of the arch. On the inside face (Fig. 2) on the second level you can see the join between the front and back slabs. The central blocks on the first and third courses have a different grain: these are separate thin pieces that cross the join. They are attached by four nails, the heads of which are visible. These blocks rest on a plinth block (Fig. 3) that as far as I can tell is a single slab of wood.

The impost block looks like a single slab, but if you look at the end grain you can see there are two cores (Fig. 4). At first this looks like it is made from two trunks meeting on their curved edges. But the bark between them is too irregularly natural. The only explanation is that this is the cross section of a tree that has branched, made just at the point where the bark is forming between the two trunks.

Above this is a wedge-shaped block that accommodates the first two voussoirs, again two blocks thick with shrinkage separating them at the middle (Fig. 5). The join comes at the groove of the second voussoir (Fig. 6). The block above forming the third and fourth voussoirs is a single block with a 120 degree wedge cut out of it (Fig. 7). Here you see a knot and a big flaw. The grain implies that the block is not radially sawn.

The Bannermans describe how they got the idea for this use of green oak from looking at a cork grand tour model of the Temple of Concord at Agrigrento ‘which we had bought in the visitor’s car park’ at Highgrove, Prince Charles’ garden, when they were designing the two temples in the stumpery there.[3] That is, the idea of using slabs of wood as if they were blocks of stone. This is certainly the way those temples work, but to carry the idea through logically, the Inigo Gate ought to have been assembled of blocks of wood like building blocks, rather that a few huge blocks with the divisions between ‘stones’ made with a router. The identity of each slab is obscured, rather that being made explicit as at the Highgrove temples. But this in fact makes a point about illustrates the nature of ashlar rustication, which develops easily from being individual blocks with their edges shaped to make the distinction between blocks explicit, to a surface that is decorated with channels that represent the joins between blocks.

Because these structures are made with green wood there is a certain amount of splitting and twisting. This is, of course, part of the aesthetic, but the virtue of oak is evidently that this is not too extreme. The movement is also constrained because the blocks are so big: on the Park Temple (another of the buildings, on the lines of the Highgrove temples) I noticed that the smaller pieces that make up the pedestal zone have moved a lot more than they appear in the photo in the Bannerman’s book. Attempts to do something similar with readily available Australian timbers have not always fared so well. Clinton Krause’s Chinese barbecue on Lake Daylesford has the four main posts constructed of heavy sections of Victorian ash, presumably from a local sawmill when these existed, but one split very badly early in the piece. It seems that in the UK very large oak tree trunks are available for such projects. No Australian tree is ever allowed to get that big. To cut these blocks major sawmilling equipment is needed, not to mention cranes to assemble it. So the apparently simple rusticity is quite deceptive: these structures require a great quantity of resources.

For this reason the Montacute Entrance Gate will be constructed of treated pine and plywood over the core that I had Ben make out of 120 mm square treated pine, clad in cement sheet. One is limited to timber 290 x 45 mm, which is big as you can get and which can be cut with a drop saw. Accoya would be better but this is available only in smaller sections, as well as being hard to get and expensive. The Accoya battlements of Montacute are glued together from lots of smaller pieces, but this kind of gluing required an industrial setup. And because my ability to machine timber is limited—I can only do basic ripping and crosscuts—the design has to be carefully calibrated to use available timber dimensions.


[1] See the description in Isabel and Julian Bannerman, Landscape of Dreams: the gardens of Isabel & Julian Bannerman, The Pimpernel Pres, 2016, pp. 202–225, see p. 203.

[2] Bannerman and Bannerman, Landscape of Dreams, p. 212.

[3] Bannerman and Bannerman, Landscape of Dreams, p. 73.

This entry was posted in Uncategorized. Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s