The painting sold at Christie's in 2017 for a record sum and attributed to Leonardo da Vinci — the ex-Cook Salvator Mundi — rests its early history on a contested seventeenth-century documentary record. The primary study of that record, the 2019 monograph by Dalivalle, Kemp and Simon, itself poses three unresolved identification questions: whether Wenceslaus Hollar's 1650 etching, declared by Hollar to be made 'from the original', can be matched to a specific painting (extant or lost) among the Cook, Ganay and Naples versions; how the two distinct Christ-as-Salvator-Mundi paintings attributed to Leonardo, recorded in the collections of King Charles I and Queen Henrietta Maria, can be distinguished from one another and identified; and by which route — if any — the royal painting left the collection after the Revolution of 1688, including whether it is the 'Head of Our Saviour' sold from the Duke of Buckingham's collection in 1763. These questions remain open. The problem — to reconstruct, or rigorously to show unreconstructable, the Stuart-era documentary chain and its bearing on the attribution of the Cook version to Leonardo — is open.

Background

A Salvator Mundi (Latin for 'Saviour of the World') is an image of Christ half-length, frontally posed, blessing with the right hand and holding a transparent orb, symbol of the cosmos under Christian dominion. Several paintings and painted copies of this composition in the Leonardesque circle (the orbit of Leonardo's own workshop and his followers) survive, of which three matter most here: the so-called Cook version (named after Sir Francis Cook, in whose collection it was recorded from 1900), bought at a small New Orleans auction in 2005, restored, exhibited in London's National Gallery in 2011 as a newly accepted Leonardo, and sold at Christie's on 15 November 2017 for about 450 million dollars — the highest price ever paid for a painting at auction; the version formerly in the de Ganay collection, Paris, championed in the earlier twentieth century as Leonardo's original but never generally accepted; and a version now in the Museo di Capodimonte, Naples. Whether the Cook version is by Leonardo's own hand, wholly or partly, or is a workshop product, remains one of the most publicized attribution disputes in Old Master studies, and the painting has been inaccessible to scholars since its 2017 sale.

The documentary spine of the dispute is seventeenth-century English. The royal collection of King Charles I (reigned 1625–1649) was inventoried and, after his execution, dispersed in the Commonwealth Sale of 1649–1653 administered by Parliament during the Interregnum, the republican period between Charles I's execution and the restoration of the monarchy under Charles II in 1660. Records of that period document not one but two paintings of Christ attributed to Leonardo in or near the royal collections — one in the king's sphere, one associated with Queen Henrietta Maria — and a key witness is the Bohemian printmaker Wenceslaus Hollar (1607–1677), who had worked in England for the collector Thomas Howard, Earl of Arundel, and who published in 1650 an etching of a Salvator Mundi he stated was made from the original painting by Leonardo. After the Restoration of 1660 a Christ attributed to Leonardo reappears in royal inventories, but the painting then vanishes from view around the Revolution of 1688, when James II was deposed and the dowager queens' goods were dispersed; only in 1763 does a 'Head of Our Saviour' attributed to Leonardo surface at auction from the collection of the Duke of Buckingham. The 2019 monograph by Margaret Dalivalle, Martin Kemp and Robert B. Simon — the attribution's principal scholarly defence — proposes a continuous chain: that Hollar copied Leonardo's original, identifiable as the Cook picture, at the Commonwealth Sale in 1650 (perhaps acting for the agent of Cardinal Mazarin); that the painting was disbursed to Captain John Stone in 1651 and returned to Charles II in 1660, reappearing as the 'Greenwich Salvator Mundi' of the inventory of about 1666; and that it may be the picture sold in 1763. Yet the book's own introduction states that it 'leav[es] the case open for further discussion', its chapter on Hollar asks 'Can we identify Hollar's model?' without clinching the match, and its chapter on the disappearance sets out the competing hypotheses — removal before 1685, passage to a jointure property of Henrietta Maria (the dowager queen's provision secured on crown lands) or to France, or removal to Portugal by Catherine of Braganza, Charles II's widow — as unresolved alternatives.

Assessing these questions requires connoisseurship — the traditional practice of judging authorship from a work's handling and quality, supplemented here by archival research in inventories, sale records and correspondence, and by technical study of the paintings (infrared reflectography, an imaging technique that uses infrared light to reveal the artist's underdrawing beneath the paint; analysis of materials; and optical or geometric analysis of the depicted orb). Later literature has deepened the technical side — computer-based inverse rendering of the orb's illumination (computational recovery of how the depicted scene was lit), a convolutional-neural-network authorship analysis (a deep-learning image classifier trained to attribute paintings from their visual style), a crystallographic study of the orb published in 2026 — and scholarly reviews have addressed the monograph's documentation. The unresolved target is therefore the reconstruction, or the rigorous demonstration of irretrievability, of the seventeenth-century documentary chain, and its bearing on whether any extant version — and in particular the Cook version — can be connected to Leonardo himself.

Problem Statement

Reconstruct — or rigorously demonstrate, on a systematic accounting of the surviving evidence, to be unreconstructable — the seventeenth-century documentary chain of the Leonardo-attributed Salvator Mundi paintings of the Stuart courts, and determine what that chain can support for the attribution of an extant painting to Leonardo. Concretely, the three linked questions posed by the primary source are: (i) Hollar's model — whether Wenceslaus Hollar's 1650 etching, which he stated was made from Leonardo's original, can be identified with a specific painting, extant (Cook, Ganay, Naples) or lost, and on what documentary and visual basis Hollar's attribution itself rested; (ii) the two royal paintings — how the two distinct Christ-as-Salvator-Mundi paintings attributed to Leonardo, documented in and around the collections of Charles I and Henrietta Maria, are to be distinguished and identified, and how each traced through the Commonwealth Sale of 1649–1653, the Interregnum and the Restoration; (iii) the disappearance — by which route, if any, the royal Salvator left the collection after the Revolution of 1688, including whether it is the 'Head of Our Saviour' sold from the Duke of Buckingham's collection in 1763. The reconstructed (or failed) chain bears on the central question: whether the documentary record connects any extant version — in particular the Cook version — to Leonardo's own hand. The common objective uniting these components is a single, source-faithful assessment of the documentary evidence bearing on the attribution; partial resolutions of individual links are welcome but must not be presented as resolving the whole.

References

  1. Margaret Dalivalle, Martin Kemp and Robert B. Simon, Leonardo's Salvator Mundi and the Collecting of Leonardo in the Stuart Courts, Oxford University Press, 2019, DOI 10.1093/oso/9780198813835.001.0001, https://doi.org/10.1093/oso/9780198813835.001.0001
  2. Margaret Dalivalle, Martin Kemp and Robert B. Simon, 'Introduction', in Leonardo's Salvator Mundi and the Collecting of Leonardo in the Stuart Courts, Oxford University Press, 2019, pp. 1–2, DOI 10.1093/oso/9780198813835.003.0001, https://doi.org/10.1093/oso/9780198813835.003.0001
  3. Margaret Dalivalle, Martin Kemp and Robert B. Simon, "'A Peece of Christ done by Leonardo' and 'A lords figure. in halfe.'", in Leonardo's Salvator Mundi and the Collecting of Leonardo in the Stuart Courts, Oxford University Press, 2019, pp. 135–152, DOI 10.1093/oso/9780198813835.003.0008, https://doi.org/10.1093/oso/9780198813835.003.0008
  4. Margaret Dalivalle, Martin Kemp and Robert B. Simon, 'After the Original: Hollar and Leonardo's Salvator', in Leonardo's Salvator Mundi and the Collecting of Leonardo in the Stuart Courts, Oxford University Press, 2019, pp. 230–245, DOI 10.1093/oso/9780198813835.003.0013, https://doi.org/10.1093/oso/9780198813835.003.0013
  5. Margaret Dalivalle, Martin Kemp and Robert B. Simon, 'Capitanus Stone', in Leonardo's Salvator Mundi and the Collecting of Leonardo in the Stuart Courts, Oxford University Press, 2019, pp. 246–257, DOI 10.1093/oso/9780198813835.003.0014, https://doi.org/10.1093/oso/9780198813835.003.0014
  6. Margaret Dalivalle, Martin Kemp and Robert B. Simon, 'Nothing is hidden under the sun', in Leonardo's Salvator Mundi and the Collecting of Leonardo in the Stuart Courts, Oxford University Press, 2019, pp. 258–271, DOI 10.1093/oso/9780198813835.003.0015, https://doi.org/10.1093/oso/9780198813835.003.0015
  7. Margaret Dalivalle, Martin Kemp and Robert B. Simon, 'The Picture Vanishes', in Leonardo's Salvator Mundi and the Collecting of Leonardo in the Stuart Courts, Oxford University Press, 2019, pp. 272–279, DOI 10.1093/oso/9780198813835.003.0016, https://doi.org/10.1093/oso/9780198813835.003.0016
  8. Philip McEvansoneya, 'Leonardo's Salvator Mundi and the Collecting of Leonardo in the Stuart Courts' (review of the monograph by Dalivalle, Kemp and Simon), Journal of the History of Collections 33(1): 142–144 (2021), DOI 10.1093/jhc/fhaa031, https://doi.org/10.1093/jhc/fhaa031
  9. Claire Farago, review of Leonardo's 'Salvator Mundi' and the Collecting of Leonardo in the Stuart Courts, Renaissance Quarterly 74(4): 1275–1277 (2021), DOI 10.1017/rqx.2021.211, https://doi.org/10.1017/rqx.2021.211
  10. Laure Fagnart, 'Leonardo's Salvator Mundi and the Collecting of Leonardo in the Stuart Courts, by Margaret Dalivalle, Martin Kemp and Robert B. Simon', The English Historical Review 137(584): 253–254 (2022), DOI 10.1093/ehr/ceab344, https://doi.org/10.1093/ehr/ceab344
  11. Steven J. Frank and Andrea M. Frank, 'A Neural Network Looks at Leonardo's(?) Salvator Mundi', Leonardo 54(6): 619–624 (2021), DOI 10.1162/leon_a_02004, https://doi.org/10.1162/leon_a_02004 (preprint arXiv:2005.10600, https://arxiv.org/abs/2005.10600)
  12. Steven J. Frank and Andrea M. Frank, 'Complementing Connoisseurship With Artificial Intelligence', Curator: The Museum Journal 65(4): 835–868 (2022), DOI 10.1111/cura.12492, https://doi.org/10.1111/cura.12492
  13. Marco (Zhanhang) Liang, Shuang Zhao and Michael T. Goodrich, 'Inverse-Rendering-Based Analysis of the Fine Illumination Effects in Salvator Mundi', Leonardo 53(4): 380–386 (2020), DOI 10.1162/leon_a_01923, https://doi.org/10.1162/leon_a_01923
  14. Jean-Pierre Crettez, 'Internal Geometry of "Salvator Mundi" (The "Cook Version", Attributed to Leonardo da Vinci)', in Ruth Scheps and Marie-Christine Maurel (eds.), Mathematics in the Visual Arts, ISTE/Wiley, 2021, pp. 103–126, DOI 10.1002/9781119801801.ch6, https://doi.org/10.1002/9781119801801.ch6
  15. Margherita Melani, 'Hollar "copia" Leonardo: l'uso della carta lucida per incisioni in scala 1:1', Incontri. Rivista europea di studi italiani 35(1): 21–31 (2020), DOI 10.18352/incontri.10331, https://doi.org/10.18352/incontri.10331
  16. Francesca Fiorani, 'Leonardo da Vinci Studies: The State of the Field at the 500th Anniversary of the Artist's Death', Renaissance Quarterly 75(1): 185–203 (2022), DOI 10.1017/rqx.2021.333, https://doi.org/10.1017/rqx.2021.333
  17. Stephen J. Campbell, 'Salvator Mundi' (pp. 9–18), in Leonardo da Vinci, Princeton University Press, 2025, chapter DOI 10.1515/9780691266220-002, https://doi.org/10.1515/9780691266220-002 (book DOI 10.1515/9780691266220, https://doi.org/10.1515/9780691266220)
  18. Juan Manuel García-Ruiz and Dianne Modestini, 'A crystallographic approach to the orb of the New York Salvator Mundi', npj Heritage Science 14 (2026), DOI 10.1038/s40494-026-02558-9, https://doi.org/10.1038/s40494-026-02558-9