An Overlooked Source of Influence for the Fan Vaulting of the Chapel of Henry VII at

Virginia K. Henderson

The vaults of the Divinity School and Christ Church pendant vaults in which the transverse arches pass through Cathedral at are generally regarded as the proto­ the independent shell structure created by the fans (Figure types for the vaulting of the Chapel of Henry VII a1 5). The transverse arches support the eight foot pendants '&suninster Abbey. While undoubtedly influenced by the which are, in fact, elongated voussoirs, the wedged build­ vaults, the design of the fan vaulting in the Chapel ap­ ing blocks of the arch itself. The pendants, in turn, sup­ pears to have another, overlooked source of inspiration. port 360 ° conoids which, together with connecting span­ A marked similarity in conception exists between the drels, constitute the fan vaulting. While the transverse vaulting of the Chapel and the of the arches a1 Westminster Hall are totally exposed, there being Great Hall at Westminster Palace. That becomes apparent no fan superimposed upon them, those at the Chapel in the comparison of the cross-sections of !he two struc­ are hidden above the shell structure for 1he center two­ tures (Figures 1-2), though I have found no reference thirds of the vault. which makes an association between them. Pevsner sug­ A fan vault is composed of inverted conoids, or gests that the use of strong transverse arches, cusping and cones, which are generated by the revolution of an arc tracery in the spandrels of the arches at the Chapel are around its vertical axis (Figure 6). The conoid consists of derived from timberwork but goes no further. ' ribs of the same curvature, spaced at equal angles and The basic designs of the hammerbeam roof and 1he bounded by a horizontal circular rib.' All horizontal sec­ masonry vault are essentially related, and their differences tions of the conoid are segments of circles, or sections can be explained by the inherent disparity of their into which circles can be inscribed.' Were the vertical strut materials. Therefore, it seems possible, and even likely, at Westminster Hall to become the vertical axis of a fan given their proximity, that the timber roof may have sug­ conoid, the arch brace which begins at its base to be gested the solution to the problem of successfully com­ revolved around the vertical strut, according 10 the defini­ bining the fan vault with the pendant vault, accomplished tion of a fan conoid, and 1he hammerbeam itself removed, al the Chapel of Henry VII, 1503-09. That solution was then the design of the Great Hall would be very close to sought but not quite successfully achieved by William that of the Chapel. The hypothetically suspended fans Orchard at the Divinity School, 1480-83, and Christ would then create a curved shell structure independent of Church Cathedral, 1500. Additionally, if in fact the design the central section of the transverse arch, much as it is of the limber roof suggested the design of the masonry found at the Chapel of Henry VII. The change of building vault, then there occurred at London a productive fusion material from timber to stone would enable such a trans­ of the accomplishments of East Anglican timberwork and formation of the design. West Country masonry vaulting. The principal differences between the construction If 1he basic structural elements of the design of the of the hamrnerbeam roof of Westminster Hall and the limber roof at Westminster Hall, that is, the transverse masonry vault of the Chapel relate to the differences in arches, 1he arch braces and 1he vertical strut, or hammer the properties of the material used. The vertical strut of posLS, were rendered in stone, 1he result would be essen­ the hammerbeam roof occupies a comparable position to tially the underpinning structure of the Chapel. It would that of the stone pendant. However, 1he 1wo function anti­ be possible then to superimpose upon that structure the thetically because of the differing constructional proper­ unique system of fan vaulting found in the Chapel. The ties of the materials. While the vertical strut of the timber very nature of executing those forms in stone would roof is supported by the barnmerbeam, its counterpart in eliminate the need for the supporting hammerbeam itself, stone is suspended from its transverse arch because the and therefore, free the vertical strut, now a suspended masonry arch is, to the extent 1hat i1 is properly con­ voussoir-pendant, to serve as the springing for the fan structed, a self-supporting composite of wedged voussoirs. vault.' In seeking prototypes for the design of the vault or stones, eliminating the need for support from below. a1 Henry VII's Chapel, the tendency appears to have been By executing the design of the timber roof in masonry 10 seek the obvious, that is any structure the vault of which and eliminating those features necessitated only by timber relies on pendants for its springing. In fact, the roof of construction, the design of Westminster Hall would yield Wes1mins1er Hall, in its medium, provides the framework the underlying design of Henry Vll's Chapel. for just that. While the vaults of the Divinity School (Figure 7) The most readily apparent similarity between the and Christ Church Cathedral (Figure 8) are more im­ vault of Henry Vll's Chapel (Figure 3) and the hammer­ mediately obvious prototypes for the vaulting of 1he beam roof of Westminster Hall (Figure 4) is their common Chapel, in some respects their designs are not as close to use of prominent transverse arches. A remarkable struc­ that of the Chapel as the harnmerbeam roof of the Great tural as well as aesthetic feat, 1he vault of the Chapel is Hall. The vaults of both these buildings are pendant vaults actually a combination of interpenetrating fan and combined with vaults, a related but altogether

II different concept from fan vaulting. While lierne vaulting Fifteenth-century masons appear to have been cau­ is conceived in linear terms, fan vaulting is conceived in tious about spanning significant distances with masonry terms of solid geome.1ric forms which are necessary 10 fan vaulls and ii appears lhal experimental designs were create an independent shell structure such as that al the often first executed in timber, especially when greater Chapel. The design of the vault at Christ Church Cathe­ spans were 10 be covered. The earliest fan vaults, excluding dral more nearly approaches the design of Henry VIl's minor tomb canopies and small chantries, are found a1 Chapel, than does that of the Divinity School. lls the cloisters of Cathedral, begun between 1351 transverse arches become lost behind the central lierne and 1377 and completed by 1412, which span a mere work and its more greatly extended pendants support rib twelve feet. Fan vaults of significant width and height were pallerns more suggestive of fans. The transverse arches not constructed in stone until the second half of the fif­ of the Divinity School, by contrast, remain fully exposed teenth century. lnteres1ingly enough, the first high vault and support short pendants, and are, therefore, less con­ of fan construction of considerable span was executed in vincingly integrated into the vault design. The most strik­ limber al Winchester College Chapel between 1395 and ing feature common to the three masonry vaults, however, 1400, attributed 10 Hugh Herland, the master carpenter in addition 10 their use of strong transverse arches, is their and designer of the bamrnerbeam roof at Westminster use of a pair of pendants suspended a distance from the Hall, which was built almost simultaneously between 1393 side walls. A pair of pendants is suggested also in the and 1398. design of the roof of Westminster Hall, although allow­ The vault of Henry VII's Chapel was built a century ing for the inherent structural characteristics of timber, later, between 1503 and 1509, probably by Robert Janyns, the pendants are represented as a pair of vertical struts although older scholarship has attributed the design er­ resting on hammerbeams. roneously to Robert or William Vertue. Each structure is The adaptation of timber constructional forms to considered the supreme achievemenl in its medium. John masonry construction was not unprecedented during the Harvey described the 1imberwork at Westminster Hall as Gothic period. Walter Horn suggests 1ha1 the concept of the most splendid of all timber construction.' Francis the medieval bay system used in Gothic masonry construc­ Bond described Henry VII's Chapel as "the most wonder­ tion had its orig.ins in constructional forms logical to ful work of masonry ever pul together by the band of timber but unnatural 10 stone cons1ruc1ion.' Jean Bony man."' 11 seems probable that the earlier timber master­ cites the vaulting of the aisles of Bristol Cathedral, circa piece inspired the later stone masterpiece, not only because 1311-30, as an example of timber constructional forms of its renown and proximity, but primarily because of the rendered in stone.• He states that "it was part of the visual resemblance of the two designs. refinements of Court art 10 translate forms from one material into another," giving further support to the likelihood lha1 lhe limberwork al Westminster Hall may have influenced the stonework at the Chapel of Henry Vil.' The University of Alabama at Birmingham

1 Nikolaus Pevsner and Priscilla MetcaJf. Tht Ca1httlrols ofEngland, 4 Walter C. Leedy, Fan Vaulting: A Study of Form, 'ltchnology ond 2 vols. (New York: Viking Penguin Inc., 198$) 2: 174. Meaning (Santa Monica: Arts & Architecture, 1980) 3.

2 The exact function of the hammerbcam within Hugh Hcrland's design S Walter Horn, ''On the Origins or the Medieval Bay System;' Jour­ remains much debated. For various arguments ooncerning the nal of the Socie1y of Architectural Historians 17 (1958) 3. loadbearing clements of 1he roof, sec L:f. Courtenay and Robert Mark, "The Westminster Hall Roof: A Historiographic and Struc .. 6 Jean Bony, The English Decorated Style: Gothic Architec.ture tural Stud)';' Journal of tht Soc.iety of Archittttural Historians 46 1>rmsformed 1250-/350 (Ithaca: Cornell UP, 1979) 36. (1987) 374-39$. 7 Bony 22. 3 My definition of a fan conoid combines concepts derived from Willi$, HO\loard and Leedy. Rohen Willis, "On the Construction or the Vaults 8 John Harvey, The Perpendicular Style 1330-1485 (London: e:r. of the Middtc Ages:• Tronsociions of the Royal Institute of British Bacsford, 1978) 74. Architects l (1842): 44, SS. Reprinted in Rohen Willis, Archilectural History of Some Engli$h Cathedrals, 2 \'Ols. (Chichcley: Paul Mince, 9 Francis Bond, Westmlnst~r Abbty (London: Oxford UP, 1989) 134. 197:lj 2, Appendix. F. E. Ho.-ard, "Fan Vaul1S:' Arr:hooological Jour­ nol 68 (1911): 2. 12 Figure I. Cross-section, Vault of the Chapel of Henry VII, Westminster Abbey, British Crown Copyright, reproduced with the permission of the Controller of Her Majesty's Stationery Office.

Figure 2. Cross-section, Hammerbeam Roof of the Great Hall, Westminster Palace, Medieval Structure: The Gothic Vault, James H. Aclund (Toronto: 1972).

Figure 4. Hammerbeam Roof, Great Hall, Westminster Palace, Courtesy of the Royal Commission on the His­ torical Monuments of .

13 Figure 3. Vault, Chapel of Henry VII, Westminster Abbey, Courtesy of the Royal Commission on the Historical MonumentS of England. 14 + + Figure 5. Isometric Drawing of the Extrados of the Vault of the Chapel of Henry VII, Westminster Abbey, drawn by Roben Willis (1842).

Figure 6. Drawing of a Fan Conoid. Drawn by James R. Alexander to the author's sketch.

Figure 7. Vault, the Divinity School, Oxford University, An Introduction 10 English Church Architecture, Francis Bond (Oxford: 1913). Figure 8. Vault, Christ Church Cathedral, Oxford, An Introduction to English Church Architecture, Francis Bond (Oxford: 1913).

15