First of all, it should be pointed out that the frame currently on this panel cannot be described as ‘engaged’ unless it is physically one with the painting it holds. This can be achieved by one of three methods:
- one, the frame forms a raised traylike border (often ornamented) around the pictorial surface, which has been lowered by carving it out from the same piece of wood as the ‘frame’;
- two, the pictorial surface does not take up the whole panel, and the pieces of the frame are glued around the outer margin; the whole object is covered with linen glued on and then gesso’d all over, and the frame is gilded before the image is painted (sometimes also having a gilded ground); the use of linen is more an Italian than a 15th–16th century Netherlandish practice, as the joiners relied on their ability to marry the planks seamlessly;
- three, the image is painted on a panel which has two raised and opposed frame-like edges made as one with it, and then two more free-standing frame mouldings are carved, which have deep grooves which slide onto the other two edges of the panel and are dowelled onto it (as in Van Eyck’s works), making the frame relatively inseparable from the panel unless it is violently split into its component parts.
If a frame on an early, or relatively early, panel painting - as in the one under consideration here – is made to be taken off the picture easily, then it is, like every later modern frame, ‘moveable’. The number of moveable frames on Netherlandish panels in the second half of the 15th century was ten times greater than the number in the first half, and four times greater than the contemporary number of engaged frames, but it is very necessary to use these terms accurately [1].
More than 99% of Netherlandish works were made from oak, and since the panels and frames would have been made in the same workshops - of joiners or carpenters, or occasionally turners - the wood used for both objects would almost always have come from the same tree. In the photo of the reverse of the Pentecost, the colour of the wood looks almost identical, but the grain doesn’t; also, the construction of the panel edge looks as though it was lowered to admit a rabbet on the reverse of the frame (similarly to point 3, above), which does not exist on the present frame. It is possible that this could have been done later, as the revealed lowered edge is lighter in colour, but this would only make sense if there had been another frame with a projecting rabbet between the original frame and the current one.
There were guild rules (e.g. in Antwerp, in 1494) to indicate that the central part of a triptych should be built sturdily enough to support the weight and movement of the wings, which also suggests that a single panel would be less robustly framed, and that, if a later addition of wings was made, the central panel would need to be reframed to cope with the extra burden [2].
In the great majority of late 15th and early 16th century panel paintings from Bruges, the flat faces of the frame (especially those of the bottom rainsill) are painted black, or less frequently another colour, such as grey or brown; they are very rarely gilded. Often, of course, the finish will have been renewed, occasionally with a change of colour where there is polychromy, but there is no reason to think that there would have been a consistent change from black to gold on all the altarpieces with flat faces, since it is easier to paint over gilding than the other way round.
Also, the frame of the Pentecost panel seems to have a small extraneous moulding at the sight edge on the three colonetted rails.
For all these reasons, it seems safe to state that this frame is not original to the panel; that it was made later, replicating the construction and mouldings of a 15th century frame, but that it is not, for instance, a 19th century repro. As wings were attached to it in the 17th century, this seems to be the logical point at which a new frame would have been produced – one which was perhaps more solid than the original and capable of supporting moving shutters, and one which was a very close but not perfect copy of a 15th century version.
[1] Hélène Verougstraete, Frames and supports in 15th and 16th Southern Netherlandish painting, 2015, p.130
[2] Ibid., p.6
