1. "The harm is done, and perfectly": Alphonse Didron, aîné, "L'Achèvement des restaurations de Saint-Denis," Annales archéologiques 5 (1846): 111.

2. Lightning struck the spire of the north tower on 9 June 1837. Debret profited by this accident to acquire new funds which enabled him to rebuild the spire and restore the entire facade: Vitry and Brière, L'Eglise abbatiale, 39-40.

3. For an adulatory account of the life and career of the sculptor, see A. Delcourt, J. S. Brun. Sculpteur statuaire. Ancien pensionnaire de Rome. Saint-Denis. L'Arc de Triomphe de l'Etoile. Le Palais de Justice de Rouen (Paris, 1846). In addition to the west portals, his restorations at Saint-Denis included the marble statues of Catherine de' Médici; Henri II; the kings, queens, and princes of the house of Valois; and the tomb of Dagobert: ibid., 56-57. He was also involved in the restoration of the sculpture on other medieval monuments, especially in Rouen. Thanks are owing to Elizabeth Brown for this reference.

4. See below, n. 42.

5. "A disfigured facade, deprived forever of historical interest, and, moreover, extremely ugly": Didron, "Achèvement des restaurations," 109.

6. See, inter alia: [Guilhermy], "Saint-Denis. Restauration," 400-411; idem, "Restauration de l'église royale de Saint-Denis," Annales archéologiques 5 (1846): 200-215; Didron, "Achèvement des restaurations," ibid., 107-113. Didron and Guilhermy also published numerous articles in the Annales archéologiques on aspects of the restorations other than those primarily concerned with the sculpture of the portal: 3 (1844): 245-46; 4 (1845): 175-85, 319-24; 5 (1846): 62-68, 107-113.

7. [Guilhermy], "Saint-Denis. Restauration," 409.

8. Didron, "Achèvement des restaurations," 113.

9. "Rapport sur la restauration de l'église royale de St. Denis [copy]," June, 1841, in

Seine, Saint-Denis, Dossier de l'Administration 1841-1876, Paris, Archives de la Commission des Monuments historiques, fols. 1-12.

10. Debret, "Réponse," fols. 18-22.

11. See "Rapport à l'Académie des Inscriptions et Belles-Lettres et à celle des Beaux-Arts sur la caractère des travaux de restauration exécutés à l'église S ^t Denys, par M ^r Debret, Architecte, Membre de l'Institut. 8 avril 1842," ibid., fols. 43-62.

12. Didron, "Achèvement des restaurations," 109: "By the decision of the minister of public works, Monsieur Debret, member of the Institute, architect of the Royal Abbey of Saint-Denis and of the Royal Academy of Music, had been made a member of the Conseil général des bâtiments." To the outrage of Didron, Debret thus joined the supreme committee on architecture that authorized and oversaw all work on public buildings.

13. Didron, ibid., 113, announced the appointment of M. Duban, architect of the Ecole des Beaux-Arts, of the Sainte-Chapelle in Paris, and of the château at Blois.

14. The spire had already been dismantled, but Viollet-le-Duc recommended that the tower also be dismantled to stabilize the weakened westwork: Viollet-le-Duc, "Rapport sur l'état des constructions au 15 Décembre mil huit cent quarante six, 2 janvier 1847," Seine, Saint-Denis, Dossier de l'Administration 1841-1876, Paris, Archives de la Commission des Monuments historiques, fols. 111-13.

15. The Archives Nationales and the Archives de la Commission des Monuments historiques contain the bulk of accounts and records itemizing works undertaken at Saint-Denis.

16. See, inter alia, "La Part de Suger," 91-102, 253-62, 339-49; and Whitney Stoddard, The West Portals of Saint-Denis and Chartres: Sculpture in the Ile-de-France from 1140-1190 (Cambridge, Mass., 1952), 2-3.

17. For a general outline of work under the direction of Viollet-le-Duc and his successors, see Crosby, The Abbey of St. Denis, 10-12; and more recently Crosby (1987), 170, 227-29, 231-32.

18. E. de Labédollière, Histoire des environs du nouveau Paris (Paris, ca. 1861), 177-78.

19. The early 1970s saw a campaign to clean the exterior of the church. Aware of the fragility of the sculptures of the western portals, Louis Grodecki was able to convince the authorities of the Monuments historiques not to clean the portals.

20. On the appearance of the three roundels after cleaning, see Blum, "Lateral Portals," 203 and fig. 8.

21. Two unpublished studies have been devoted to a critical examination of the sculptural details of the western portals: a resumé of the thesis of Cécile Goldscheider for the Ecole du Louvre appeared as "Les Origines des portails à statues-colonnes," Bulletin des musées de France 6/7 (1946): 22-25; and the doctoral dissertation of Johann Eckart von Borries, "Die Westportale der Abteikirche von Saint-Denis. Versuch einer Rekonstruktion," Ph.D. diss., Universität Hamburg, 1947. Both Mme Goldscheider and Dr. von Borries kindly gave Professor Crosby permission to use their manuscripts in this study.

A third study, by Paula L. Gerson, The West Facade of St.-Denis. An Iconographic Study, Ph.D. diss., Columbia University, 1970, Ann Arbor, Mich. (UMI, 1970, no. 73-26, 428), was concerned with iconographical interpretations of the portals. But since this study preceded the completion of the archaeological examination of the central portal, Gerson did not have the benefit of the analysis of the restorations.

22. A shorter version of the results of the examination was first published in France: Sumner McKnight Crosby and Pamela Z. Blum, "Le Portail central de la façade occidentale de Saint-Denis," trans. D. Thibaudat, Bmon 131/3 (1973): 209-66. Crosby had previously presented a report summarizing the first, but still preliminary, findings of the 1968 examination of the central portal. Read at the symposium "The Renaissance of the Twelfth Century," held May 14-15, 1969, at the Museum of Art, Rhode Island School of Design, the paper was published as "The West Portals of Saint-Denis and the Saint-Denis Style," Gesta 9/2 (1970): 1-11.

23. Erwin Panofsky, "The History of Art as a Humanistic Discipline," in Meaning in the Visual Arts (Garden City, N.Y., 1955), 19.

24. The following opinions exemplify conflicts in discussions of style. Whitney Stoddard wrote, "The search for possible origins of the Saint-Denis-Chartres style has led us to Burgundy": Stoddard, West Portals, 53. Willibald Sauerländer has maintained, "At Saint-Denis, there is, as far as I can see, no trace of any Burgundian inspiration": Sauerländer, "Sculpture on Early Gothic Churches: The State of Research and Open Questions," Gesta 9/2 (1970): 34. He later concluded, "A variety of styles formed the sculptures on the west doorways of Saint-Denis. Though it is now impossible to reach a definite conclusion, the predominating models seem to have come from Toulouse and its sphere of influence": W. Sauerländer, Gothic Sculpture in France 1140-1270, trans. J. Sondheimer (New York, 1972), 381.

25. Recently two articles have presented and discussed all available evidence with respect to the presence or absence of carved lintels in the three portals and the possible heights of the originals: Paula L. Gerson, "The Lintels of the West Facade of Saint-Denis," Journal of the Society of Architectural Historians 34 (1975): 189-97; and Kathryn A. Morrison, "The Eighteenth-Century 'Restoration' of the West Portals of Saint-Denis and the Problem of the Lintels," Journal of the British Archaeological Association 139 (1986): 134-42 and plates. Using the measurements of the three sets of bronze doors given by Doublet in seventeenth-century pieds and comparing them with the heights of the doors today, Gerson computed their original heights by adding the figure of 0.60 m.—the measurement proposed by Viollet-le-Duc as the twelfth-century level of the sills below today's pavement level. With those givens,

Gerson concluded that there had been no lintel under the mosaic in the tympanum of the left portal, one of 0.13 to 0.17 m. under the central portal, and one of circa 0.42 to 0.45 m. under the tympanum of the right portal. But as Crosby noted, Viollet-le-Duc failed to state from what level he had taken his measurement, and since that time, Crosby's excavations have established the sill level at -0.52 m. That measurement proved compatible with Doublet's figures in pieds —with the pied equaling 0.325 to 0.328 m.: Crosby (1987), 138, 293-94. The correct measurement of the sill level at 0.52 m. below today's pavement modifies Gerson's conclusions and undermines Morrison's assertion that Doublet's measurements in pieds were not sufficiently accurate to allow close computations such as Gerson's. Gerson drew her conclusion on the basis of Doublet's figures compared with the widths of the portals today. Morrison assumed that the widths had never been altered, but the archaeological evidence suggests otherwise (see Chapter 6). She therefore found significant discrepancies between the widths today and Doublet's measurements.

26. Scamozzi's interest in the western portals is recorded in a minute sketch showing the schema of the bronze doors of the central portal as consisting of scenes within medallions. Unfortunately the scenes are indecipherable. See Franco Barbieri, "Vincenzo Scamozzi, studioso ed artista," Critica d'arte an. 8, ser. 3, fasc. 19 (1949): fig. 171, and Paris, Arch. Phot. no. 53.N.122. The bronze doors lie outside the scope of this article, but see Gerson, West Facade, 100-111; idem, "Suger as Iconographer," in Gerson, ed., Abbot Suger, 186-87; and Crosby (1987), 187-92, for interpretations of the visual and textual evidence concerning the iconographical program of the doors.

27. Album Debret, Paris, Archives de la Commission des Monuments historiques. Inasmuch as Debret's attachements bear the date 1838 and many details included were never completed, the drawing was probably made in preparation for a presentation to the ministry in support of his request for funds. The details apparently represent what he proposed to do and were not working drawings for the restorers. Similar discrepancies exist between his detail drawings of 1840, which supposedly show work completed (Fig. 5). Guilhermy, ms. 6121, pp. 36, 66, noted that the restoration of the central portal began around 15 September 1839. By October of that year he could write, "On commence le bas relief de la porte gauche modelé par Brun. Le même sculpteur venait de restaurer toutes les voussures de la porte centrale, et commençait à restaurer le bas relief du tympan" (Work is beginning on the bas-relief of the left portal modeled by Brun. The same sculptor has finished restoring all the archivolts of the central portal and is beginning to restore the relief [sculptures] of the tympanum).

28. Diagrams superimposed on photographs proved the most accurate and precise way to summarize the results of the examination. Like any schematic record, they present explicit information, not quantitative and qualitative assessments, such as how severely the surfaces were recut or how accurate the design of the insertion. The text will cover those aspects. Coded so that all original twelfth-century carving can be seen at a glance, the unmarked surfaces represent sculpture untouched by the restoration. Generously spaced crosshatching circumscribed by outlines pinpoints every inset. More closely spaced hatching drawn on the opposite diagonal represents repairs made with mastic, mortar, cement, and gesso. The latter hatching thus maps all repaired fractures, as well as repaired masonry joints impinging upon and distorting the sculpture, even those where some of the mortar fill has crumbled and fallen away. Dotted lines represent fractures that postdate the restoration and have never been repaired. Crossed broken lines designate recut surfaces of twelfth-century stone. That last convention marks every recut area, whether the recutting caused real deformation or simply eliminated minor surface abrasions. For those distinctions, again the reader must rely upon the text.

29. Since the publication of the monograph "Le Portail central," by Crosby and Blum, a considerable number of important studies have appeared concerned with the iconography of the portals. See Paula L. Gerson, "Suger as Iconographer," 183-98; Pamela Z. Blum, "Lateral Portals," 199-228; and Zinn, "Suger, Theology," 33-40; all in Gerson, ed., Abbot Suger. See also Charles T. Little, "Monumental Sculpture at Saint-Denis under the Patronage of Abbot Suger," in Crosby et al., Saint-Denis, 25-29; Crosby (1987), 179-213; and Conrad Rudolph, Artistic Change at Saint-Denis: Abbot Suger's Program and the Early Twelfth-Century Controversy over Art (Princeton, N.J., 1990), especially his chap. 5.

30. Because of the installation of the nineteenth-century pavement 0.52 m. above the twelfth-century level, the lowest bed of masonry visible today appears irregular and completely arbitrary. The original twelfth-century mural masonry continues below the pavement.

Most of the visible blocks of mural masonry are, in fact, original. Stoddard, basing his conclusion on the masonry sloping inward above the level of the decorated plinths and on the depth of the recession, originally posited that every stone and all the carving above the plinths had been cut back 4 cm. in 1839: Stoddard, West Portals, 2-3. He later retracted this conclusion after learning the results of the archaeological examination of the central portal: idem, Sculptors, 113. Yet no explanation for the recession of the stones of the jambs directly above the plinths seems adequate. Perhaps it represents a peculiarity of the portals, a reflection of difficulties encountered when the masonry of the portals was assembled after the sculptures had been carved in the workshop, or even a tectonic thickening of the structure to strengthen it.

31. The dimension of 29.5 cm. closely approximates the accepted length of the Roman foot: George Forsyth, Jr., The Church of St. Martin at Angers (Princeton, N.J., 1953), 23 n. 4. See also below, Chapter 6, n. 2.

32. The fractures may have occurred during the alterations to the portal in 1770-1771, when the trumeau was removed, but more probably in the nineteenth century, when the rebuilding of the north spire caused movement within the masonry of the whole facade.

33. On the exterior surface of the lunette of the upper tympanum, behind an angel's nineteenth-century wing, the end of one of the bolts is visible. On the interior, the heads of the bolts are now obscured by a heavy coat of plaster. A number of holes filled with plaster suggest additional bolts were inserted from the exterior as well. See Chapter 3, "Lunette with Tympanum Angels."

34. See n. 28 above, explaining how the various types of restorations are differentiated in the diagrams.

35. In the Chapter Acts book, in which the canons of Salisbury cathedral transcribed all the plans and specifications of George Gilbert Scott for the restoration of the cathedral in 1876, we find an entry specifying the addition of linseed oil to the mortar used in replacing masonry in the most exposed locations, such as the tower and spire. Scott believed that linseed oil made the mortar impervious to the damaging effects of weather. At Saint-Denis the oil may well have been used for hardening gypsum used to make plaster of Paris, and certainly the oil would have tinted it a buff color.

36. [Guilhermy], "Saint-Denis. Restauration," 407, mentioned that "mastique et la terre cuits" had been used in the restoration of the portals.

37. Guilhermy, ms. 6121, pp. 49, 53, 61, identified the material as "pierre factice." The composite stone appears impervious to weather and acid rain. See also Chapter 6, "The Jamb Colonnettes."

38. Archival records for Saint-Denis contain a payment to M. Dihl (Diehl), maker of mastic imperméable: "L'Eglise abbatiale de Saint-Denis, 1811-1822," Paris, Archives Nationales, F ¹³ 1295, fol. 76, as well as numerous references to the use of mastic in the early nineteenth-century restorations and alterations. In one instance, the formula, which included linseed oil, seems to have produced a mastic used for coating the walls in the crypt to prevent dampness from penetrating the stone: ibid., F ¹³ 1296, fol. 100. In his discussion of restorations to French buildings, Paul Léon, La Vie des monuments français. Destruction. Restauration (Paris, 1951), 366, wrote: "Later Debret in the work at the basilica of Saint-Denis, [and] Godde in the churches of Paris, authorized the facing of masonry by extraordinary materials, the mastic of Diehl, the cement of Molesmes or of Wasy, which, thanks to the quick and uniform scraping [of the stones], gave a new appearance to the most degraded monuments."

39. On the extent of the eighteenth-century work and aspects distinguishing it from the nineteenth-century restoration, see Blum, "Lateral Portals," 200-202.

40. Von Borries, "Die Westportale," 116-17. See also Arch. Phot., Paris, no. 83675. For a possible explanation of the function of the fragment, see the ornamental animal-head finials, typical embellishments of bench ends that rise above the level of the seat: Eugène Viollet-le-Duc, Dictionnaire raisonné du mobilier français de l'époque carlovingienne à la renaissance, 1, Meubles (Paris, 1871), "Bancs," 33-34. See Bildarchiv Foto Marburg no. 36991, and Arch. Phot., Paris, nos. 11471, 83627.

41. Nevertheless, as noted above, the accounts indicate that Brun alone restored the figures. He had considerable assistance, mainly for restorations to and replacement of