At Canterbury Cathedral, Léonie Seliger has built a career around one of conservation’s most demanding responsibilities: caring for fragile historic glass while continuing to understand it as art, architecture, evidence and a living craft. Seen from the floor of a cathedral, a medieval window may appear almost untouched by time. Colour gathers within the glass. Figures emerge from fields of blue, red, green and gold. Sunlight passes through surfaces made centuries ago and transforms the interior of the building. Closer examination reveals something much more fragile. Glass may be pitted, darkened or chemically altered. Painted detail can loosen from its surface. Lead cames may be fatigued, fractured or distorted. Rain, wind and pollution attack the exterior, while condensation repeatedly wets the inner face. Earlier repairs, substitutions and rearrangements may sit beside medieval material so convincingly that their different histories become apparent only through careful examination. To conserve stained glass properly is therefore to work between what can be seen from a distance and what is revealed through patient attention. Few contemporary practitioners embody that combination of craftsmanship, scholarship and close observation more fully than Léonie Seliger ACR, Director of The Cathedral Studios and the person responsible for the conservation of the stained glass windows of Canterbury Cathedral. Her work demonstrates that stained glass conservation is not a matter of making an old window appear complete or new. It is the more difficult practice of recognising what has survived, understanding how it has changed and deciding how it can be protected without erasing the evidence of its long life. Seliger studied glass painting and design at Hadamar College in Germany before completing an MA in Buildings Conservation at the University of York. She joined Canterbury Cathedral’s stained glass studio in 1991 and has led it since 2005. She is an accredited conservator-restorer and a Freeman of the Worshipful Company of Glaziers. She also advises the Dioceses of Canterbury and Rochester on proposals involving the conservation, restoration and design of church windows. That background is important because her formation was not confined to the history of stained glass or to the science of material deterioration. It began with making: drawing, cutting glass, painting, firing, leading and understanding how a window works within architecture. Historic stained glass cannot be read only as an image. Its lead lines are part of its drawing. Variations in thickness, texture and colour affect the transmission of light. Bubbles, streaks and irregular surfaces may reveal how the glass was manufactured. Tool marks, solder joints, painted lines, corrosion, fractures and previous repairs form a physical record of the window’s creation and continuing history. A conservator who understands how glass is made is better able to recognise how it has failed. A glass painter can distinguish an original painted line from a later addition. A glazier can assess whether bowing is related to tired lead, inadequate support, structural movement or an earlier intervention. The intelligence of the hand and the intelligence of historical research become inseparable. Seliger’s career has maintained this connection between maker and conservator. Alongside preserving historic windows, she has designed new stained glass, showing that the craft remains capable of contemporary expression rather than belonging only to the past. The permanent stained glass workshop at Canterbury Cathedral occupies a distinctive place within British conservation. Repair and conservation work at the Cathedral was entrusted for more than 130 years to four generations of the Austin and Caldwell families, beginning after George Austin Senior was appointed Surveyor to the Cathedral in 1819. The present workshop was formally established in 1972 and continues to care for Canterbury’s glass while also undertaking commissions for churches, cathedrals, museums, historic houses and private clients. This continuity matters. A permanent cathedral workshop acquires a depth of knowledge that cannot easily be reproduced through occasional campaigns of repair. Its conservators learn not only individual windows but the behaviour of the building as a whole: its orientations, microclimates, masonry, metal supports, maintenance history and previous interventions. The workshop also preserves records, photographs, drawings, fragments and practical knowledge accumulated across generations. A repair that might otherwise appear unexplained may be traceable through an old studio photograph, a glazier’s estimate or evidence left within the window itself. Yet continuity does not mean repeating inherited practice without question. The history of stained glass restoration includes interventions that would no longer be considered acceptable. Medieval glass was sometimes rearranged, heavily releaded, supplemented with conjectural painting or altered to satisfy changing ideas of visual completeness. Earlier restorers often understood themselves to be improving damaged compositions rather than preserving each stage of their material history. The value of a long-established studio therefore rests not simply in retaining tradition, but in being able to examine that tradition critically. Historic repairs may themselves have acquired significance. Before deciding whether they should remain, however, the conservator must identify, document and understand them. Under Seliger’s direction, the Canterbury workshop operates at the point where institutional memory meets contemporary conservation ethics, environmental research and scientific analysis. The most important tool in stained glass conservation is not always a kiln, microscope or analytical instrument. It is sustained attention. A window that appears coherent from the floor of a cathedral may prove, on the light table, to be a complicated assembly of original glass, medieval alteration, early repair, later insertion and modern reconstruction. Seemingly minor details can change the interpretation of an entire composition. The form of a lead line may reveal that a figure was moved. A difference in corrosion may indicate that one piece was protected by an earlier repair. A painted fragment that appears meaningless in its present position may belong to another scene. This is especially important at Canterbury Cathedral, whose glass has survived fire, iconoclasm, neglect, structural change, deliberate rearrangement and successive approaches to restoration. Seliger’s work repeatedly demonstrates that conservation is also a form of discovery. The purpose of examining a window is not merely to determine how it should be repaired. Close study can reveal how the original artists worked, how their windows were understood and how later generations treated them. The conservator is therefore not simply treating an object that historians have already explained. Conservation can itself produce new historical knowledge. This relationship between practical conservation and scholarly discovery is most clearly demonstrated by the Thomas Becket Miracle Windows Research and Conservation Project. Created between approximately 1180 and 1220, the Miracle Windows originally formed a group of twelve surrounding the site of Becket’s shrine in the Trinity Chapel. Eight survive. Their panels translate accounts of illness, pilgrimage and healing into some of the most vivid narrative stained glass of the early Middle Ages. In 2018, Canterbury Cathedral launched a major interdisciplinary project to remove, examine, record and conserve the surviving windows and return them to more stable environmental conditions. The project is jointly led by Seliger and medieval historian Dr Rachel Koopmans. Seliger directs the conservation of the glass, while Koopmans undertakes the historical research underpinning its interpretation. The windows are exceptional not only as works of colour and composition but as records of medieval lives. They show men, women and children travelling to Canterbury, suffering illness, receiving treatment and giving thanks for recovery. Beds, vessels, crutches, clothing, gestures and family groups appear alongside Becket’s tomb and the expectation of miraculous intervention. For the conservation team, however, these narratives are encountered not simply as scenes but as physical constructions. Every piece of glass must be examined. Painted detail, lead lines, corrosion, fractures, stopgaps and altered sections must be recorded. The conservators must distinguish medieval material from later repairs without assuming that everything added later is necessarily without value. The result is historical investigation rooted in the material fabric of the windows. The close examination of the Miracle Windows has transformed understanding of how much medieval glass survives and how the panels have been altered. The first window investigated through the project had undergone corrosion, loss, iconoclasm, repair and rearrangement across eight centuries. Once the glass was removed and examined at close quarters, Seliger and Koopmans were able to identify medieval material more precisely and relate the surviving compositions more closely to the written miracle accounts produced after Becket’s murder. Their findings are presented in The First Miracle Window: Becket’s Earliest Pilgrims in Canterbury’s Stained Glass, written by Koopmans with digital reconstructions and an afterword by Seliger. The work explains how investigation of the glass revealed not only the original artists’ intentions but also the techniques and decisions of the glaziers who repaired the window in later centuries. Further discoveries followed during the examination of window nIII. The window was removed for display in the British Museum exhibition Thomas Becket: Murder and the Making of a Saint, allowing its panels to be studied closely. The investigation enabled Seliger and Koopmans to reinstate the medieval ordering of two stories that had previously been incorrectly arranged. Their findings are described in Journey to Thomas: Pilgrimage and Healing in Window nIII. Such discoveries demonstrate why the physical examination of historic fabric is indispensable. Documentary research may suggest how a window was conceived, but the glass itself records what happened to it: where it was cut, moved, repaired, supplemented or misunderstood. Earlier repairs can complicate a window’s appearance, but they can also preserve evidence. During the investigation of one Miracle Window, conservators found traces of narrow strap leads that had once been soldered across the medieval lead matrix. These structural additions crossed figures and inscriptions with little concern for visual clarity. Yet areas sheltered beneath them could retain surfaces that were better preserved than the surrounding glass. Damage and protection had become intertwined. A repair that impaired the composition may also have preserved part of it. The subsequent removal of that repair may have restored visual clarity while eliminating evidence of an earlier method of intervention. The history of a window therefore cannot be reduced to an original medieval state followed by centuries of decline. Each repair changed both the object and the conditions of its survival. This is why conservation demands restraint. To remove every later addition in pursuit of medieval purity would destroy evidence of how the glass was valued, altered, misunderstood and preserved by successive generations. Reconstruction without falsification Seliger’s work also illustrates how reconstruction can help people understand damaged stained glass without being imposed upon the original. In an article for Vidimus, Koopmans discusses a Miracle Window panel showing a sleeping man threatened by demons. Much of the composition had been altered by an earlier restorer, including the replacement of heads and missing sections. Seliger produced a digital reconstruction suggesting how the damaged scene might originally have appeared. Medieval models from elsewhere in the Canterbury glass were used to propose possible heads, figures and architectural elements. Crucially, the reconstruction was explicitly presented as hypothetical. (vidimus.org) This distinction is fundamental. Earlier restoration practice might have involved painting new pieces of glass and inserting them into the original panel, permanently converting interpretation into physical fabric. A digital reconstruction allows scholars and the public to explore a possible original arrangement while leaving the historic window untouched. It acknowledges the natural desire to see a lost composition more clearly, but it does not disguise uncertainty as fact. The original survives as evidence. The reconstruction remains an argument. Although stained glass conservation remains deeply dependent upon hand skills, Seliger’s work has also involved scientific collaboration. She was a co-author of the open-access study “Dating Nathan: The Oldest Stained Glass Window in England?”, undertaken with Laura Ware Adlington and Professor Ian Freestone. The study examined glass from Canterbury Cathedral’s Ancestors of Christ series using non-destructive, in-situ chemical analysis. A specially developed three-dimensional-printed attachment enabled a portable X-ray fluorescence spectrometer to be positioned consistently against glass surrounded by projecting lead cames. The results supported the argument that figures dating from before the Cathedral fire of 1174 had been reused in later windows. The analysis also suggested that ornamental borders had been reused, indicating that more early glass survives at Canterbury than had previously been recognised. (MDPI) The findings complicate any simple distinction between original and reused material. Medieval glaziers themselves recycled valuable historic glass. Figures and borders were adapted for new architectural settings, allowing older material to continue within later compositions. Scientific analysis did not replace visual examination, craft knowledge or art-historical research. It strengthened and tested interpretations developed through those other forms of expertise. The most productive result emerged through collaboration between conservators, historians and archaeological scientists. One of the greatest challenges in stained glass conservation is that the window is both an artwork and part of the building envelope. Its exterior is exposed to rain, wind and pollution. Its inner surface is vulnerable to condensation, which can contribute to the irreversible loss of glass paint and other applied decoration. One method of reducing these risks is Environmental Protective Glazing. This introduces a secondary external layer while placing the historic stained glass within a more sheltered position. Seliger and environmental consultant Tobit Curteis co-authored Historic England’s major research report, Conserving Stained Glass Using Environmental Protective Glazing. The report makes clear that protective glazing is not a standard product that can be applied in the same manner to every building. Its success depends on temperature differences, ventilation, moisture movement, the condition of surrounding masonry and the detailing of the installation. A poorly designed system may improve conditions for the glass while damaging other parts of the building or altering the external appearance of the window. Environmental Protective Glazing is therefore not a universal solution. Its benefits must be balanced against its architectural and environmental effects. (Historic England) This work captures an essential element of Seliger’s conservation approach: treatment should be based on evidence rather than habit. The window, the wall and the internal environment must be understood as a connected system. Cleaning stained glass is easily misunderstood. To the untrained eye, dirt may appear to be a superficial layer that can simply be removed to uncover the brightness beneath. In reality, a darkened surface may contain corrosion products, unstable paint, deposits, microorganisms or layers that have become physically integrated with the glass. Attempts to improve appearance can therefore cause irreversible loss. Seliger’s article “Cleaning Historic Stained Glass Windows” explains the need to assess the nature and stability of deposits before treatment. Cleaning methods must respond to the particular glass, its painted decoration and the condition of its surfaces. This represents a wider conservation principle. The object does not owe the viewer maximum brightness. The purpose of treatment is not to create a dramatic before-and-after transformation, but to preserve significant material while improving stability and legibility only where this can be achieved safely. A surface marked by age is not necessarily a failed surface. The conservator must distinguish damaging contamination from historically meaningful change. New glass in historic buildings Seliger’s importance as a Guardian of Heritage rests not only on preserving medieval glass. She is also a designer of new stained glass. Her commissions include work for All Saints, Boughton Aluph; the chapel of the Margaret Thatcher Infirmary at the Royal Hospital Chelsea; and St Lawrence the Martyr at Godmersham. The Godmersham windows use hand-blown glass in warm, silvery tones, interrupted by smaller areas of stronger colour. Painted imagery draws upon fossilised plants, insects and early human footprints, introducing a meditation on human presence, memory and the passage of time. (Art + Christianity) In 2023, Seliger also designed and created a new window for Brewers’ Hall, commemorating the Worshipful Company of Brewers’ historic association with Thomas Becket. The window brings together the Company’s earlier and present coats of arms and was produced through the Canterbury studio. (Brewers' Company) These commissions are not separate from Seliger’s conservation work. They demonstrate the continuity of the craft. A designer of new stained glass confronts many of the same fundamental questions as the medieval glazier: how colour changes through the day, how lead lines carry a composition, how the window relates to masonry and how imagery can speak within a particular place. New work in a historic building must neither imitate the past so closely that it becomes deceptive nor disregard its surroundings in the pursuit of novelty. It must possess an identity of its own while contributing to a setting already shaped by other generations. Seliger’s work as both conservator and designer reinforces the idea that heritage survives most fully when its crafts remain active. A tradition maintained only through repair will eventually narrow. A living craft must also retain the confidence to create. The survival of specialist conservation depends upon more than the achievement of an individual practitioner. The Cathedral Studios bring together conservators, glass painters, designers and glaziers with varied training and experience. The workshop contains dedicated conservation, painting and design spaces, a glazing workshop, lead-milling facilities, photographic resources and climate-controlled storage. (thecathedralstudios.com) Within such an environment, knowledge is transferred through practice. A developing conservator learns how to handle unstable glass, recognise vulnerable paint, document a panel and judge whether an intervention is necessary. These skills cannot be acquired solely through written guidance. They depend upon observation, repetition, discussion and working alongside experienced practitioners. Seliger’s advisory work for the Dioceses of Canterbury and Rochester extends that influence beyond the studio. Proposals for conservation, restoration or new church windows must be assessed not only for technical competence, but for their effect upon architectural significance, historic fabric and the life of the building. This advisory responsibility is an often overlooked form of conservation. The most successful intervention may be the one that is reconsidered before work begins. Informed scrutiny can prevent unnecessary replacement, excessive cleaning, poorly designed protective glazing or new work that fails to respect its setting. The phrase Guardian of Heritage can suggest someone standing protectively between an object and change. Seliger’s work presents a more active and demanding model. To guard historic glass is to investigate it. It is to recognise that established interpretations may be mistaken. It is to preserve fragments without pretending that they are complete. It is to understand old repairs rather than automatically removing them. It is to use scientific analysis without allowing technology to displace craft judgement. It is to design protective systems that respond to the whole building. It is to advise, teach and sustain the workshop culture through which specialist knowledge survives. It is also to make new work. Historic stained glass was never static. Medieval glaziers reused earlier material. Later generations repaired, rearranged and reinterpreted it. Cathedrals and churches continued to commission windows in response to new patrons, events, beliefs and communities. The responsibility of the present is not to freeze that history, but to participate in it honestly. Seliger’s career shows that rigorous conservation and creativity are not opposing forces. Both depend upon close observation, material knowledge and respect for the consequences of every decision. Stained glass occupies a unique place within historic buildings. It is image and structure, colour and weathering surface, fragile material and architectural boundary. It can survive for eight centuries and lose irreplaceable detail through one insensitive treatment. It can appear complete while containing fragments from several periods. It can illuminate the lives of medieval pilgrims while preserving evidence of early glaziers, later repairers and modern conservators. Léonie Seliger’s work at Canterbury Cathedral encourages us to recognise all these layers. Through the Cathedral Studios, the Miracle Windows project, environmental research, scientific collaboration and the design of new glass, she has helped demonstrate that conservation is not an attempt to remove time from an object. It is the practice of understanding what time has left behind. The achievement is not a window made new. It is a historic window allowed to remain truthful, legible and alive in the light. Publications by or involving Léonie Seliger Adlington, Laura Ware; Freestone, Ian C.; and Seliger, Léonie. Curteis, Tobit, and Seliger, Léonie. Koopmans, Rachel, and Seliger, Léonie. Koopmans, Rachel, with digital reconstructions and an afterword by Léonie Seliger. Seliger, Léonie. Publications on Canterbury Cathedral’s stained glass Caviness, Madeline Harrison. Caviness, Madeline Harrison. Koopmans, Rachel. Weaver, Jeffrey, and Caviness, Madeline Harrison. Wider guidance and conservation research Brown, Sarah. Historic England. Canterbury Cathedral.Stained glass can conceal its own vulnerability.
A craft learned through making
The continuity of the Cathedral Studio
Conservation begins with looking
The Thomas Becket Miracle Windows
Recovering the windows’ histories
Reading repair as evidence
Science in the service of the object
Protecting stained glass in place
The ethics of cleaning
Knowledge passed through the workshop
Why Léonie Seliger is a Guardian of Heritage
The living craft of light
Selected Publications and Further Reading
“Dating Nathan: The Oldest Stained Glass Window in England?”
Heritage, volume 4, issue 2, 2021, pp. 937–960.
An open-access scientific and art-historical study of early glass reused in Canterbury Cathedral’s Ancestors of Christ windows. (MDPI)
Conserving Stained Glass Using Environmental Protective Glazing
Historic England Research Report Series 43/2017.
A major technical examination of the operation, design and potential limitations of Environmental Protective Glazing.
Journey to Thomas: Pilgrimage and Healing in Window nIII
Canterbury Cathedral.
An illustrated account of the investigation of Miracle Window nIII and the reinstatement of the medieval ordering of two narrative sequences.
The First Miracle Window: Becket’s Earliest Pilgrims in Canterbury’s Stained Glass
Canterbury Cathedral, 2022.
A concise study of discoveries made through the close examination of the first Thomas Becket Miracle Window.
“Cleaning Historic Stained Glass Windows”
The Building Conservation Directory, 2013.
Practical guidance on surface deterioration, fragile glass paint, deposits and the risks associated with inappropriate cleaning.
The Early Stained Glass of Canterbury Cathedral, circa 1175–1220
Princeton University Press, 1977.
A foundational study of Canterbury Cathedral’s early glazing programme. (Internet Archive)
The Windows of Christ Church Cathedral, Canterbury
Corpus Vitrearum Medii Aevi, Great Britain, volume II, British Academy, 1981.
The principal scholarly catalogue of Canterbury Cathedral’s medieval and early Renaissance stained glass. (Google Books)
“Demons and Discoveries in a Miracle Window of Canterbury Cathedral”
Vidimus, issue 123.
An account of the examination of a Miracle Window and Seliger’s hypothetical digital reconstruction of its damaged demons panel.
The Ancestors of Christ Windows at Canterbury Cathedral
J. Paul Getty Museum, 2013.
An illustrated study of the surviving figures from Canterbury’s monumental Ancestors of Christ glazing programme.
“Medieval Stained Glass and the Victorian Restorer”
19: Interdisciplinary Studies in the Long Nineteenth Century, issue 30, 2020.
A study of the complex relationship between nineteenth-century restoration, scholarship, commerce and the development of modern conservation ethics.
Stained Glass Windows: Managing Environmental Deterioration
Published 2020 and amended in July 2024.
Guidance for owners, building managers and professional advisers on recognising environmental deterioration and commissioning specialist investigation.
Thomas Becket Miracle Windows Research and Conservation Project
The principal online introduction to the continuing project, including films, interpretation and information about the surviving windows. (Canterbury Cathedral)
| Author: | Mark Shaw |
|---|---|
| Date of Publication: | 17/07/2026 |
New listings, project updates, and specialist insights—every month.