Cover image for Opening Doors: The Early Netherlandish Triptych Reinterpreted By Lynn F. Jacobs

Opening Doors

The Early Netherlandish Triptych Reinterpreted

Lynn F. Jacobs


$103.95 | Hardcover Edition
ISBN: 978-0-271-04840-6

328 pages
10.5" × 9.5"
40 color/140 b&w illustrations

Opening Doors

The Early Netherlandish Triptych Reinterpreted

Lynn F. Jacobs

“This remarkable, lucid book takes on a big and complex subject, still somewhat invisible to scholarship. It fully reconsiders a major late medieval art form: the triptych format of hinged altarpieces. The triptych—along with its components, oil paint technique and refined oak carving—rose to Golden Age prominence in fifteenth-century Flemish art. Offering an impressive survey of this great artistic achievement, Opening Doors truly lives up to its name and contributes fresh new interpretations. Scholars and their students will use this book as a standard work for many years to come.”


  • Description
  • Reviews
  • Bio
  • Table of Contents
  • Sample Chapters
  • Subjects
Opening Doors is the first book of its kind: a comprehensive study of the emergence and evolution of the Netherlandish triptych from the early fifteenth through the early seventeenth centuries. The modern term “triptych” did not exist during the period Lynn Jacobs discusses. Rather, contemporary French, Dutch, and Latin documents employ a very telling description—they call the triptych a “painting with doors.” Using this term as her springboard, Jacobs considers its implications for the structure and meaning of the triptych. The fundamental nature of the format created doors that established thresholds, boundaries, and interconnections between physical parts of the triptych—the center and wings, the interior and the exterior—and between types of meaning, the sacred and the earthly, different narrative moments, different spaces, different levels of status, and, ultimately, different worlds. Moving chronologically from early triptychs such as Campin’s Mérode Triptych and Van Eyck’s Dresden Triptych to sixteenth-century works by Bosch, and closing with a discussion of Rubens, Jacobs considers how artists negotiated the idea of the threshold. From her analysis of Campin’s ambiguous divisions between the space represented across the panels, to Van der Weyden’s invention of the “arch motif” that organized relations between the viewer and the painting, to Van der Goes’s complex hierarchical structures, to Bosch’s unprecedentedly unified spaces, Jacobs shows us how Netherlandish artists’ approach to the format changed and evolved, culminating in the early seventeenth century with Rubens’s great Antwerp altarpieces.
“This remarkable, lucid book takes on a big and complex subject, still somewhat invisible to scholarship. It fully reconsiders a major late medieval art form: the triptych format of hinged altarpieces. The triptych—along with its components, oil paint technique and refined oak carving—rose to Golden Age prominence in fifteenth-century Flemish art. Offering an impressive survey of this great artistic achievement, Opening Doors truly lives up to its name and contributes fresh new interpretations. Scholars and their students will use this book as a standard work for many years to come.”
“Admirably broad in its sweep—from Jan van Eyck to Rubens—this book tackles a fundamentally important question: how the form of the triptych affected its meaning. Noting that archival evidence reveals that this art form was envisioned as a panel covered by doors, Lynn Jacobs develops the idea of the ‘miraculous threshold.’ She explores the rich and complex relationships between the triptych’s exterior and interior, and between the central panel, the most important from a theological point of view, and the wings. This book will undoubtedly have a major impact on the field.”
“With her characteristic meticulous scholarship and intellectual verve, Lynn Jacobs opens doors in our understanding of the triptych, one of the defining formats of early Netherlandish painting. Using a wealth of contemporary sources and her sensitive readings of individual works, she convincingly demonstrates how ‘paintings with doors,’ as triptychs were termed, structured and generated meaning for artists and audiences alike. Painters from Robert Campin and Jan van Eyck to Hieronymus Bosch and, later, Peter Paul Rubens exploited the triptych’s thresholds, those physical distinctions between center and wings or inside and outside panels, to create often elegant narrative and symbolic programs. Jacobs has written a richly rewarding, indeed essential, book for anyone seeking to comprehend early Netherlandish art.”
“Jacobs’s fascinating book should reopen scholarly interest in these marvelous paintings with doors.”
“Through careful and insightful interpretation of the extant visual material and contemporary terminology, Jacobs’s argument offers a fresh perspective on this particularly Netherlandish altarpiece format.”

Lynn F. Jacobs is Professor of Art at the University of Arkansas.


List of Illustrations


Introduction: The Triptych as a “Painting with Doors”

Part I: Origins and the First Half of the Fifteenth Century

1 The Emergence of the Early Netherlandish Triptych I: Robert Campin (and His Associates)

2 The Emergence of the Early Netherlandish Triptych II: Jan van Eyck

Part II: The Second Half of the Fifteenth Century

3 The Triptych Reformulated: Rogier van der Weyden

4 The Triptych Popularized: Painters of the Second Half of the Fifteenth Century

5 The Triptych Unified: Memling, David, and Later Fifteenth-Century Painters in Bruges

Part III: The Sixteenth Century and Beyond

6 The World Triptych: Hieronymus Bosch

7 The Triptych in the Age of the Renaissance and the Reformation

8 Coda: The Triptych in the Age of Rubens





The Triptych as a “Painting with Doors”

The pictorial worlds created by fifteenth- and sixteenth-century Netherlandish artists often extended beyond one panel to inhabit three separate panels hinged together, a format that we today call the triptych. Many of the canonical works of Flemish painting—Robert Campin’s Mérode Triptych, Rogier van der Weyden’s Columba Altarpiece, Hugo van der Goes’s Portinari Altarpiece, and Hieronymus Bosch’s Garden of Earthly Delights—along with countless other works by anonymous artists, were executed as triptychs rather than as single-panel works. The popularity of the triptych as a format for Northern Renaissance painting has made it so familiar that we tend to read through it, as if it were transparent.1 To be sure, scholars have noted the practical functions served by the format: how the wings protect the center from dust and damage (especially helpful during transport); how the structure allows for changing imagery (particularly useful in accommodating changing liturgical purposes); and how the multiplication of panels permits greater narrative elaboration or the inclusion of donors (the latter a plus in this age of increasing private patronage).2 But aside from an almost tacit acceptance that the format was symbolic of the Trinity,3 there has been little sustained assessment of the role the triptych format might have played in the construction of meaning.

The best-known book on triptychs, Shirley Blum’s Early Netherlandish Triptychs: A Study in Patronage, of 1969, largely sidestepped questions of meaning. Blum saw the triptych functioning simply as a means of re-creating experiences previously found in medieval architecture, that is, as providing a way to take the total thought realm expressed within the architectural and sculptural programs of Gothic cathedrals and condense it onto a multipanel painting—not as carrying its own distinctive meanings.4 In this way Blum provided a naturalistic (that is, historically descriptive) explanation whereby the triptych arose as part of the shift in media in the late Gothic period away from architectural sculpture and toward panel painting. In so doing, she made format a contingent feature, just one of many ways in which this medieval mind-set expressed itself, a mere happenstance of the changing roles of the different art media rather than something that by its very nature structured meaning.

A more obscure German study, published a decade earlier, Klaus Lankheit’s 1959 Das Triptychon als Pathosformel, did present, in a nascent form, a theory of how the format contributed to meaning. Lankheit argued that the triptych format was a “pathos formula,” analogous to the gestures (studied by Aby Warburg, who coined the term pathosformel) that Renaissance artists adopted from ancient art to heighten the expressive force of their works. In particular, Lankheit believed that the triptych’s emphasis on the center (and on the subordination of the wings)—which he referred to as a “subordinating center”—functioned like Warburg’s gestures: because the “subordinating center” was derived from antique triptychs, in which it heightened the grandiosity of cult images of the Roman emperors, this motif increased affective power and sacred effect when incorporated into Renaissance triptychs.5 Lankheit’s argument pushed the significance of format into the realm of expressive content, a noteworthy advance, especially compared to the approach of his German predecessor, Jakob Burckhardt, whose 1886 paper “Format und Bild” argued that format simply delimited the beautiful from the remaining space.6 Although Lankheit’s work was already out of fashion by the time Blum wrote her book (Blum does not include him in her bibliography), his work has enjoyed a revival of late: two recent German studies were based on his book—one, Antje Neuner’s dissertation on early Netherlandish triptychs, focused on the compositional aspects of the “subordinating center,” thereby skirting issues of meaning, but the other, Karl Schade’s book, modernized Lankheit’s interpretation of meaning by arguing that within small-scale triptychs the “subordinating center” made the center a focus for devotion and contemplative prayer.7 The most sustained study of the meaning of the triptych format, however, is found in Marius Rimmele’s 2010 book, Das Triptychon als Metapher, Körper und Ort: Semantisierungen eines Bildträgers. Rimmele’s arguments—some of which are similar or complementary to ones I advance in this book—focus on the way in which the triptych, with the opening of its wings, structurally embodies the concepts of epiphany and revelation.8

Ultimately, any study of the construction of meaning within the triptych format needs to be grounded on documentary evidence. Unfortunately, the sources say very little about how people thought about triptychs. But they do provide insight into the conceptual framework that controlled the way in which people at the time comprehended the format—albeit indirectly, through the terminology used to refer to triptychs. The documents reveal that during the late Gothic and Renaissance periods triptychs were most commonly designated not by a term reflecting their tripartite nature (as in the current locution) but rather as “paintings with doors.” If we use the terminology of the times, that of “paintings with doors”—not Lankheit’s pathosformel and “subordinating center”—as the starting point for an analysis of the triptych, then a quite new and somewhat unexpected picture of the format and its role in the creation of meaning emerges.


In the fifteenth and sixteenth centuries there was no specific term for triptych, nor, it seems, was there such a term in earlier periods when triptychs were in use, be it ancient Rome or medieval Byzantium.9 Instead, Netherlandish triptychs were most frequently denoted in the documents in the manner of a 1480 record that refers to a Last Judgment by Dieric Bouts as a “small panel with its doors of the Judgment” (eenen cleinen Tafelnelkene met synen dueren van den Ordele)—using the Dutch term dueren (doors) to designate what we today would call the wings.10 Similarly, a 1489 inventory of an inn in Leuven lists “in the dining room . . . a painting, with two doors, standing on the altar there” (In de eetcamere . . . een taefferneel, met 2 doren, staende op den outaer aldair);11 and a 1505 inventory of the house of Cornelis Haveloes lists “a painted painting [made] of wood with two small doors of our beloved Lady” (een geschildert tavereel van houte met twee doerkens van onser liever Vrouwen).12 The equivalent term in Latin documents is janua (doors), as in the contract for the Holy Sacrament altarpiece commissioned in 1513 from Jan de Molder by the abbey of Averbode (a sculpted, rather than painted, work).13 Other related Latin terms used in the documents to designate the wings of triptychs include valvae (double doors or folding doors) and portae (doors): so, for example, the 1504 inventory of the abbey of Clairvaux refers to a “tabula cum portis.”14 French documents often use the term huisses or huissieres (doors), as seen in the 1540 contract between the painter Christophe and the officials of the Church of Saint John the Baptist in Namur, which includes a request that the artist “gild the frame of the doors of the aforementioned panel as much inside as outside” (dorer les bordeur des huysse de ladite table tant pardedens que dehors).15 This French term is found also in the 1319 payment for “another gilded panel, full of relics, opening with two doors” (une autre tavle dorée, plaine de reliques, ouvrant a deus huissieres).16 Some French documents on occasion use a related term, porte (door): thus, the 1420 inventory of the possessions of Philip the Good lists a triptych of precious materials, described as “a painting of gilded sliver, opening in the manner of [a] door” (Un tableau d’argent doré, ouvrant en façon de porte).17

Although “doors” was the predominate term used to designate the wings of Netherlandish triptychs, other locutions were employed at times. In rare instances the actual term “wings” does appear, as in the 1526 inventory of Saint Denis in Troyes, which uses the Latin alae (wings) in speaking of “a wooden panel, which is closed with two wings” (Tabellam ligneam, que duabus alis clauditur).18 But the term “wing” seems to have come into use mainly in the seventeenth century: Karel van Mander, writing in 1603–4, for example, refers to the Ghent Altarpiece at one point as having “two wings or double doors” (twee vleugelen oft dobbel deuren).19 The most common alternative to the terminology and conceptualization of the wing-as-door, however, was the designation of the wings as “leaves,” using the same term used for leaves of a book. This appears most often in French documents, which frequently designate wings as feuilles: for example, a will of 1412 refers to “one large painting, which opens with two leaves, showing the suffering of Our Lord” (j grant tavelet qui se oeuvre à deux fuelléz, figuré de la Souffrance Nostre Seigneur);20 and the inventory of Margaret of Austria lists a triptych by Rogier der Weyden, with wings by Memling, as “a small painting of a Man of Sorrows resting in the arms of Our Lady, having two leaves in each of which is an angel, and [on the] outside [of] the aforementioned leaves is an Annunciation in black and white” (Ung petit tableaul d’ung Dieu de pityé estant es bras de Nostre Dame; ayant deux feulletz dans chascun desquelz y a ung ange et dessus les dits feulletz y a une annunciade de blanc et de noir).21

The equivalence of “leaves” and “doors” for designating the wings is evident in the contract for the altarpiece of the abbey of Flines, which speaks of “leaves or doors” (feulletz ou huissetz).22 But while the two words can equally serve as references to triptych wings, the term “leaf” associates the triptych format more closely with books, linking the experience of the triptych conceptually with that of reading and more particularly with the page-turning that necessarily accompanies the act of reading.23 The analogies with page-turning seem to capture particularly well the unfolding of narrative sequences possible within the triptych format. However, to see wings as leaves of a book that are turned is also to understand wings as things to be moved, that is, opened and closed—a notion also central to an understanding of wings as doors.

Only in very rare cases does the terminology used for triptychs evoke the three-part character of these works. The 1379 inventory of the possessions of Charles V, king of France, refers to “a large painting of three parts, covered with silver outside and inside, closing, in which [there] are several relics inside” (Ungs grans tableaulx de troys pièces, couvers d’argent dehors et dedens, cloans, esquelz sont dedens plusieurs reliques).24 And the inventory of Charles the Bold mentions “an altarpiece of embroidery, in three pieces, of which there is, in the center piece, the story of the Three Kings making an offering, and the other piece has the death of Our Lady, and in the third [there] is the story of the nativity of our Lord” (Une table d’autel de broudure, en trois pièces, dont il a en la pièce du milieu l’istoire des trois rois faisant l’offrande, et l’autre pièce a le trespassement Nostre Dame, et en la tierce a l’istoire de la nativité nostre Sgr).25 But locutions of this nature are exceptionally rare in medieval and Renaissance discourse on the triptych. Instead, the discourse centers around the terminology of the door and in so doing associates the triptych with the complex of meanings surrounding the concept of the door.

Doors and Meaning

The designation of triptychs as “paintings with doors” exposes important aspects of the conceptualization of the triptych within the period.26 In particular, it reveals that an essential aspect of the triptych for its users and viewers is the way in which its wings function like doors to create boundaries, or thresholds, between the various zones of the triptych—that is, between center and sides of the opened triptych, and between the outside and inside of the closed one. Indeed, the language of the documents is replete with references to these thresholds, using such phrases as “on the one side” and “on the other side” (in de een zyde and in de ander zyde in Dutch) and “inside” and “outside” (binnen and buyten in Dutch, perdedens and pardehors in French).27 The terminology of opening and/or closing very frequently forms an intrinsic part of the way a triptych is defined or identified: for example, one document speaks of a “small painting, closing with two small doors” (tafelkin, sluutende met twee duerkins),28 and another, conversely, of “a . . . gilded panel . . . opening with two doors” (une . . . tavle dorée . . . ouvrant a deus huissieres).29

This language indicates that the addition of wings to a single panel to form a triptych is more than merely incidental. Rather, the inclusion of wings reconfigures a painting into a structure that maps out boundaries between its different panels and between imagery that takes places in different times and spaces, depicts figures on different levels of status (male and female, holy and not, or less, holy), and/or shows different states of being (statues and living creatures, man and God). The format thus uses doors to structure imagery in ways analogous to the role of doors in ritual. Beginning with Arnold van Gennep, anthropologists have recognized that doors are key elements in rites of passage from one status to another; in these rites of passage, crossing through the openings marked by the doors, that is, going across the thresholds, is, as Gennep puts it, “to unite oneself with a new world.”30 Indeed, in late medieval and Renaissance northern Europe, marriage was one of a number a key rites of passage that often took place at a doorway.31 So, too, the rite of passage from the lay world into the cloistered one also centered around the door: in the Benedictine rule, entry into the cloister involved the ritual of pulsans (knocking), in which a person wishing to enter the monastery and take on a religious life had to knock on the door to the monastery for four or five days—and thereby demonstrate his commitment to the cloistered life—before being admitted.32 Moreover, an especially crucial passage, that between life and death, was conceived in terms of doorways between different worlds. As is evident in paintings of the Last Judgment, death involved passage through the porta paradisi (the gate, or door, of paradise) or, for the less fortunate, the mouth of hell; the latter possibility confirms Mary Douglas’s observations on how, within ritual, orifices of the human body often mark boundaries in a manner similar to door posts.33

But the concept of the door carried additional, more specific meanings and associations within late medieval religious culture—all of which had implications for the meaning structure of the triptych. Most significant, the central figures of medieval Catholicism (and frequent subjects of early Netherlandish triptychs), Mary and Christ, were closely associated with the notion of the door. Among the many symbols and appellations for the Virgin Mary, one of the most common was the porta paradisi. This designation drew on the theological notion that Mary opened the gate to heaven, which had been shut by Eve—an idea expressed in the office of Terce in the Hours of the Virgin with the line “The gate of heaven was shut by Eve, but by thee, blessed Virgin Mary, it was opened once again” (Paradisi porta per Evam cunctis clausa est, et per beatam Mariam virginem iterum patefacta est).34 The Dominican liturgy of the Office of the Virgin evokes this same analogy in the following lines: “Yes you are the Door that leads to the King most high. You are the Door sparkling with light.” And the famous Marian hymn “Ave Maris Stella” calls Mary “happy gate of heaven” (felix coeli porta).35

Mary was very strongly associated not just with the open doors of paradise but also with its opposite, the closed door, or porta clausa. This association developed out of Saint Ambrose’s interpretation of Ezekial’s vision of the closed door as an Old Testament type for the perpetual virginity of the Virgin: Ambrose wrote that “Mary is the door which was closed and not to be opened”—an image that was perpetuated in later medieval authors, such as Honorius of Autun, who included Ezekial’s porta clausa in his compendium of Old Testament prefigurations of the Annunciation.36 This imagery was also popularized in medieval Annunciation plays, notably East Anglian Annunciation plays, which were closely related to their Netherlandish counterparts; these English plays—in a rather direct reference to Ezekial’s porta clausa—often include a scene in which Joseph, returning to the house after the Annunciation, has to knock three times before gaining entrance.37 Thus during the fifteenth and sixteenth centuries the door was a very well entrenched Marian symbol—and a particularly potent one—because its dual possibilities of opening and shutting were especially suited to the paradoxical, liminal position of Mary as both mother and virgin, both human and the bearer of God. Indeed, in the Middle Ages such concepts were very literally conveyed in the so-called Vierge ouvrante, a type of Marian statue in which the body of the Virgin is constituted by doors that open up to reveal an image of the Trinity within.38 The Vierge ouvrante statues express a full nexus of ideas about the Virgin—the Virgin as an open door to salvation and as a closed door of inviolate purity—as well as ideas about the door and the body as markers of crucial boundaries.

Christ too, during the Middle Ages, was conceived as a door. This notion—which reflects the liminal nature of Christ’s body as both dead and alive, human and divine, open and closed39—has direct roots in Scripture. In John 10:9, Christ, as part of a series of self-definitions that dominate this Gospel, says, “I am the gate. Anyone who enters through me will be safe.”40 The concept of Christ-as-door was elaborated in a sermon by Augustine, who stated that Christ was the door and that this door was opened when his side was pierced by the lance.41 In this way Augustine associated the opening of the door to salvation with the opening of the boundaries of Christ’s body. These concepts had continued resonance in the Lowlands well into the sixteenth century. A sixteenth-century Flemish devotional text, Den tempel onser sielen, describes the soul as a temple that was broken by original sin and had to be remade (through the Incarnation of Christ): describing the remade (now redeemed) temple of the soul, the text quotes God as saying, “I shall open the door, and the gates shall not be closed again. I shall show you hidden treasures, and hidden secrets shall I reveal in you” (Ic sal die dore opluycken, ende die poorten en sullen niet weder ghesloten werden. Ic sal di toonen verborghen schatten, ende verborghen heymelicheden sal ic in di openbaren).42 The text goes on to say, “This door or gate he [Christ] opened when he let us pierce his holy side, and let us open his godly heart, then the hardness of our heart was broken and the closed darkness removed from it” (Deze dore oft poorte dede hi op, doe hi ons zijn heylige side liet doorsteken, ende zijn godlike hert openen: doe wert ghebroken die verherdicheyt ons ghemoets, ende die besloten duysterheyt daer-van ghedaen).43 In this way the text continues the Augustinian association of the piercing of Christ’s side with the opening of the doors of salvation but also relates the image of Christ-as-door to the spiritual state of the inner soul.

Indeed, in late medieval spirituality, the spiritual relation between the worshipper and Christ was often seen through the metaphor of the door.44 Much of this imagery drew on Song of Songs 6, in which the Bride hears her beloved knocking and opens the bolt of the door to the beloved. Additional imagery derives from Revelation 3:20, “Look, I am standing at the door, knocking. If one of you hears me calling and opens the door, I will come in to share his meal, side by side with him.” These textual sources formed a basis for religious practices, particularly in female spirituality (examined in detail by Jeffrey Hamburger), in which devotion to Christ was conceived as opening the doors of one’s heart to Christ when he comes knocking; thus the codex with the legend of Hedwig of Silesia, a thirteenth-century saint, states that “she continually awaited the coming of the Consoler, so that on his arrival and knocking at the gate of her heart, she could quickly open the door.”45

The notion of the door, of course, had resonance for an understanding not only of Mary and Christ but also of the church building itself. The doors of medieval churches were the focus of much decoration and often bore inscriptions marking their significance; these inscriptions frequently drew on the imagery of Jacob’s vision of a ladder into the skies, repeating Jacob’s exclamation, “This is the house of God and the door to heaven” (Haec est domus dei et porta caeli).46 In this way the doorway of the church was literally as well as figuratively inscribed as the entrance into another world, not the mundane, profane one, but the heavenly, paradisial realm of God. Of course, all of this was consonant with the interpretation of the church door as symbolic of Christ, as in Durandus of Mende’s thirteenth-century treatise on the symbolism of the church, since Christ’s redemption was the means by which mankind was finally allowed entrance into heaven in the aftermath of the Fall.47 The significance of the door of the church is also manifest in consecration rituals for new churches: as part of this rite, the bishop would circle the exterior of the church three times, sprinkling it with holy water, and after each circuit would strike the door with his crosier and recite from Psalm 23, “Lift up your gates, ye princes, and be ye lifted up, ye everlasting doors, and the King of Glory shall come in.”48 Byzantine Eucharistic rites are also closely bound up with doors, with ritual activities centering on two entrances, the first through the outer portal into the church interior, the second through the chancel screen (or templon) into the altar area.49

The motif and concept of the door reverberated throughout the whole church structure, not just its entrance. Like its predecessor, the Jewish temple, the medieval church was structured less around space for its own sake and more around the control of space through attention to thresholds and internal divisions.50 One key division, between the space of the clerics and that of the laity, was created through the regular use of choir screens (with doors) in medieval churches.51 Further divisions could be established by various other furnishings, such as the choir tapestries, which could enclose the choir on feast days and separate the space of the canons from that of the laity, in the nave, and the bishop, at the altar.52 Not only were the spaces of the church divided and marked by thresholds, but also the church was filled with furniture and objects that themselves marked thresholds with their own sets of doors. Church sacristies usually contained storage cabinets for vestments and Mass implements—and sometimes relic cabinets—which had real doors that could be closed and locked to protect the precious and sacred contents.53 So, too, church altars and treasuries accommodated wooden winged altarpieces, metal reliquary triptychs, and sculptures within tabernacles, whose doors protected against thieves and dust, even as they provided fields for pictorial or sculptural imagery. Like the shutters of altarpieces, doors also protected the pipe organs in churches.54 Even manuscripts had doors of a sort, for the covers of books were equivalent to doors: they had to be opened to gain access to the text and were locked with clasps when the reading was concluded.55

Of course, the door also defined limits within domestic as well as religious space. Entrances into the home, and private spaces within the home, were marked by doors and thresholds to be crossed. Objects of special value were locked and enclosed within cabinets with doors (to be discussed more fully below), which were common and important furnishings in medieval homes. Major public spectacles were also often located within the framework of the door: for example, mystery plays and other dramatic productions could be staged within arched frameworks, which sometimes were closed off with curtains or triptychlike doors.56 Many of the tableaux vivants set up for the joyous entry of Charles V into Bruges in 1515 were situated within sets outfitted with doors, which had inscriptions on their outsides and opened to display painted images on their insides.57 The opening of these doors revealed living actors in staged scenes designed to convey specific political message.

All of this indicates that to call triptychs “paintings with doors” was to vest upon them a variety of associations from a variety of contexts—all of which were replete with meaning. Putting doors onto paintings—and conceiving of artworks as “paintings with doors”—thus was an intrinsically value-laden activity. The response of late medieval viewers of “paintings with doors” would have incorporated the same sort of awareness (conscious or subconscious) of the role of the door that we ourselves would experience if a stranger came knocking at the door of our house in the middle of the night. One notion seemingly not contained within the concept of the triptych as a “painting with doors” is that of the Trinity. Although allusions to the Trinity are often considered basic to its three-part format, the conceptualization of the triptych as a painting with doors, that is, as one unit with some accessories, tends to deny or at least ignore the format’s triune nature.

Openings and Closings of the Doors

Above all, viewers of “paintings with doors” would have noted whether the doors were closed or open. Most Netherlandish triptychs allowed for the folding of the wings, which had imagery painted on their reverses, giving the triptych two distinct views, interior and exterior. In church contexts the opening and closing of the triptych—though not well documented—was closely linked to the celebration of the Mass: thus, for example, an Ordinary from the cathedral of Laon (twelfth to thirteenth century) states that the altarpiece should be opened on feast days, except that of the Annunciation during Lent.58 Similarly, the instructions to the sexton of the Oude Kerk in Delft (1539) provide lists of the festivals when the high altarpiece should be opened and half-opened, as do instructions for sextons of churches in Nürberg, Lübeck, Tergernsee, and Freisung.59 Other documentation suggests, albeit indirectly, that Rogier van der Weyden’s Beaune Altarpiece was opened specifically on feast days. According to this account, on feast days a parement showing an Annunciation was placed on the main altar below the altarpiece; the addition of the parement is probably a sign that the altarpiece was opened, since it makes little sense to have displayed a parement of the Annunciation beneath a closed altarpiece that also depicted the Annunciation.60 Whether these openings occurred prior to the ritual or as part of the ritual is not clear. But John Knox’s 1559 description of an iconoclastic act in the Parish Church of the Holy Cross of Saint John the Baptist (Saint Johnston, present-day Perth, Scotland), tells of a congregation that was moved to destroy an altarpiece on the high altar after it was opened up by a priest who was going to say Mass—a story that suggests that at least in this instance the opening of the altarpiece was a public act and thus likely part of liturgical ceremony.61

The smaller triptychs in a church, whether on altars or over tombs, could also be opened and closed in relation to the Mass. Thus, for example, a 1570 document provides money for the maintenance of a candle in front of the epitaph of Roelants van Wynde, which was located in the chapel of the Holy Trinity of Saint Peter’s in Leuven; the act also states that “the candlelighter shall also be responsible every Friday, at the time that they shall have the Mass of the Holy Cross in the aforementioned church, to open the aforementioned painting and, the aforementioned Mass being finished, to shut and close [it]” (sal de ontsteker oyck schuldich zyn alle vrydagen, ten tyde als men de misse van den heyligen Cruyse inder voers. kercke doen sal, tvoersc. tafereel open te doen ende voers. misse gedaen zynde, wederom toedoen ende sluyten).62 So, too, the 1582 testament of Marthe Oliviers states: “Likewise [she] desired to be placed above her and her husband’s tomb her painting of our beloved Lady, standing on the sun, [with] Saint John the Baptist and Saint John the Evangelist on the doors, with a metal candlestick standing in front of it, whereupon [on] all four feast days one shall place a wax candle of one-half pound and [shall] light the same candle during the High Mass [on] all Sundays and feast days; for the candle to be lit and [for] the painting to be open and shut . . . the sexton of the church every year will receive six pennies” (Item, begeert boven haer en haers mans sepulture gestelt te worden haer taeffereel van onser liever Vrouwen, in de son staende, op de doere Sint Jan Baptista en Sint Jan Evangelista met eenen metalen candelaer, daer voer staende, waerop men, alle vier hoechtyden sal setten een wasse kersse, van eenen halven ponde, en de selve kersse ontsteken, onder die hoemis, alle sondage en hoochtytdagen, voer welcke kersse te ontsteken, en het taeffereel op en toe te doen . . . die koster van de kercke jaerlycx hebben sal ses stuyvers).63 Although this document is not fully explicit on this point, the clear implication is that the painting was open and shut in conjunction with the lighting of the candle, which, in turn, was done in conjunction with the celebration of the Mass.

Thus, within a church, some triptychs, whether functioning as altarpieces or as epitaphs, were opened in relation to liturgical ceremony. But the practices of opening and closing are likely to have been more complicated than the documentary evidence suggests. Quite likely altarpieces were opened outside of liturgical celebration also—especially those altarpieces on secondary altars and in private chapels, which could have been opened more often at the discretion of the viewer and in response to the needs of personal, as opposed to ritual, devotion.64 Moreover, in domestic settings the time and occasion of the opening of triptychs were based solely on the devotional activities—and/or the interior-decorating proclivities—of the owner. Certainly the triptych that, according to a 1489 document, stood in the dining room of an inn in Leuven would have been opened and closed in accordance with a less rigid schedule or—if the proprietors were overtaxed or inattentive—could have been displayed permanently in an open (or closed) position.65

The situation is further complicated by the fact that the opening of triptychs admits of degrees. Wings do not have to be opened flush with the center but can be opened only partially, so that they end up placed at an angle to the center. Although no firm documentary evidence has been found,66 some visual evidence supports the claim that triptychs, like diptychs, were sometimes displayed with angled wings.67 Images of triptychs within paintings often show an angled placement of the wings,68 and certain triptychs themselves supply evidence of angled display. In some cases, physical evidence points to such a display configuration; in others, formal elements within the triptych imply an angled display of the wings, such as a system of perspective that “works” only when the wings are placed at a particular angle or a variation in the figure scale between center and wings that appears to make better sense when the wings are placed at an angle.69 Specific examples are considered later in this book.

The opening and closing of the triptych—while it might vary in circumstance and degree—establishes a hierarchy in the format. Because the inside at many times is hidden and protected by the closed wings, it assumes a higher status vis-à-vis the exterior, often (at least in religious works) a more sanctified, holy character.70 In addition, the opened view itself has a hierarchical structure, with the larger central section, partitioned off from the wings, standing as the focal point of the work. In this way the “painting with doors” is indeed marked by a “subordinating center,” as Lankheit observed.71 But equally important, the “painting with doors” is a structure that establishes thresholds between exterior and interior, as well as between center and sides. Thus, part of the “essence” of the triptych is to raise questions about the relations between the representations on either side of each threshold and about how open or closed these thresholds are. These questions, and the answers provided within the triptychs, have important implications for the meaning of the works.

But the questions (and answers) also have implications for how the viewer engages with triptychs. Paul Philippot has argued that the experience of viewing works with thresholds to cross heightens the viewer’s identification with the imaginary reality created within the work.72 And indeed it does seem that the necessity, imposed by the format, for the viewer to understand how the various zones of the triptych relate to one another makes the experience of viewing a triptych particularly engaging. The level of viewer engagement also was heightened by the mutability of the form: since triptychs could not only be opened or closed but also be opened to different angles, they provided a wide variety of views. Moreover, at times they even could offer the viewer a kinetic (or, for the actual manipulators of the artworks, haptic) experience of the very process of shifting from one view to another.73 The format of a “painting with doors” thereby allowed fully pictorial works to take on a characteristic of architecture and sculpture and occupy real space (not just create illusionistic space). In so doing, the triptych format could address and manipulate another threshold—that between the artwork and the audience—in ways that differed significantly from single-panel paintings.

Internal Evidence

While the language of the documents speaks to the ways in which the audience and consumers of art understood triptychs, visual evidence within the triptychs indicates that the artists too conceptualized the triptych as a “painting with doors.” A telling example appears in one of the best-known Netherlandish triptychs, the Mérode Triptych (color plate 1), whose date of around 1425 places it among the earliest known Netherlandish painted examples and whose panels depict the Annunciation in the center, the donors at the left, and Joseph in his workshop at the right.74 In this work, the frames between the left and center panels separate the representation of an opened door on the one side (behind which the donors appear) and a barely visible door frame shown within the center panel, to the left of the angel Gabriel. A door thus is depicted within that part of the triptych called the “door,” and—thanks as well to the door frame depicted in the center—a threshold is represented between center and wing.75

Such a confluence between imagery and format suggests that the triptych is self-consciously commenting on its own nature.76 Indeed, the conceit of the opened door is reiterated in the form of other hinged panels shown throughout the triptych in various stages of opening or closing. Most notably, the left panel has a second door—this one held open by a man (whose identity is a matter of iconographic controversy)—which forms a threshold between the city in the background and the garden where the donors kneel. But also, the center and right panels have a number of window shutters, which look like minidoors, flapping open and closed in various ways. For Michael Ann Holly, these visualizations of the theme of opening and closing stand as metaphors for the process of interpretation itself.77 But the depictions of openings and closings also allude to the openings and closings of the triptych. The doors and windows shown in the Mérode Triptych thereby serve as meta-images of the format wherein they appear.

The door in front of the donor (in the left panel) has been opened with keys, which still hang out of the keyhole (fig. 1) and form an additional indication that what the Mérode Triptych is about (at least in part) is the triptych format itself. Although largely ignored within the plethora of symbolic decodings of this work, the keys, as Holly has noted, are very prominently displayed, and the artist renders them with special attention: indeed, the keys are among very few objects in the scene that cast shadows.78 These keys raise many questions, not just those about the possibilities of interpretation noted by Holly, but also more basic questions about the story line of the work. Whose keys are these, and who has opened the door? Did the donor insert the keys into a keyhole on the left side of the door (as he faced it), open the door, and then swing it around so that the keys now face the center panel? Or did the angel or the Virgin open the door from the inside and push it slightly ajar to allow the donor a peek into the inner sanctum? And why were the keys left in the door? The artists who created this work—which likely was a collaborative product from the shop of Robert Campin—leave these questions unanswered, perhaps deliberately. But as shown, the keys emphasize the presence of the door and more specifically the process of opening the door to create an open threshold between wing and center; they also have temporal implications, suggesting that the Virgin’s chamber only just previously had been locked up tight.79 The real reason why the keys hang out of the door in clear view of the viewer may be less important to the story line of the work than to the need to show the audience how the image of a recently unlocked door on the left panel is a metonym for the triptych format itself.

The self-consciousness about format seen in the Mérode Triptych is similar to that in a later work, Pieter Aertsen’s 1552 Christ in the House of Mary and Martha (fig. 2), a painting that to a significant degree is about painting.80 The foreground of this work is dominated by life-sized still-life elements—mainly kitchen utensils, flowers, and food—while the background shows an opening into the next room, framed as if a painting within the painting, where Christ, Mary, and Martha are situated. The work thus can be seen as a “split painting,” in which trompe l’oeil still life is set against pictorialized religious imagery.81 The two areas coexist in marked contrast, both in terms of style—the Flemish illusionism of the still life versus the Italianate rendering of the religious scene—and iconography, kitchen items versus religious history. As analyzed by Victor Stoichita, the background can be seen as the text (translated into a painting) that is sacred and rendered through methods of traditional painting, whereas the foreground is outside the text, a nonsacred world that comments on the text, encouraging the spectator to look beyond the imagery of terrestrial food to understand its true meaning, the spiritual food offered by Christ.82 In making this commentary, the foreground projects itself right out of the painting into the real world in front of the painting, that is, the kitchen where such a work could have hung. The work thus involves questions about reality versus fiction, particularly about where the boundaries between them are to be found, and hence becomes a discourse on the nature of representation itself.

One of the significant motifs in this discourse is reminiscent of the imagery seen in the left wing of the Mérode Triptych, a cupboard door with keys in the lock, which swings open toward the viewer in the right foreground of Aertsen’s work. Stoichita emphasized the role of this door in aggressively encroaching on the spectator’s space, helping to bring the foreground, in some sense, out of the painting.83 But within the context of a painting that is clearly concerned with representational theory, the door also functions as an overt introduction of the viewer into the fictional world of the painting, perhaps even alluding to locked cabinets or rooms within which sixteenth-century paintings were often housed. The dangling keys left in the lock provide an allusion to the artist, whose hand is missing (or at least not represented holding the keys) but is present in the work, and whose art unlocks and opens up the world of representation to the viewer. Aertsen’s door thus participates in the commentary on the nature of painting, just as Campin’s door participates in the commentary on the nature of the triptych.

In these two works, then, the representations of doors express the artists’ self-awareness of what their art is doing. Typically, this sort of artistic self-consciousness is associated more with sixteenth-century than with fifteenth-century Northern art.84 But Campin’s early fifteenth-century Mérode Triptych shows a clear awareness of the analogy between the door depicted in the scene (a door of a house) and the door structure of the triptych—along with a specific and self-conscious desire to manifest this awareness.85 Indeed, the deliberate inclusion of doors, which form literal illustrations of the thresholds spanned, seems as appropriate for the Mérode Triptych as for Aertsen’s Christ in the House of Martha and Mary, since both works stand at the threshold of new directions in art, the early Netherlandish triptych and the early modern genre scene.

The analogies between the opened doors of Aertsen’s and Campin’s paintings are indicative of the ways in which the split image of the later sixteenth century re-creates aspects of the triptych within a single panel. In Aertsen’s panel, as in the Mérode Triptych, there are splits and separations between zones of reality (that is, between zones of sacred history and of the here and now)—although those in the Mérode Triptych are created by literal splits between the panels, whereas those in Aertsen’s painting are embedded within the single pictorial surface. But neither work is purely about separation, since they also use the door to forge a connection between zones, in one case linking the space of the spectator to that of the painting (Aertsen) and in the other linking the space of the donor to that of the object of his prayers (Campin). The unlocked doors of these two works thus say a lot about how these works convey meaning.


This book focuses on demonstrating how the triptych format structures and generates meaning. My goal is not to argue that the format creates only one meaning, but rather to examine how each triptych manipulates the format—by structuring the thresholds inherent in it—to create different meanings. The meanings created within triptychs are in no way exclusive to the triptych format, for single-panel works can (and do) construct and manipulate thresholds much as triptychs do.86 However, unlike the single-panel format, the triptych format—or, more precisely, the format of a “painting with doors”—requires the artist to deal with multiple boundaries in constructing meaning within the work. Hence questions about relations across thresholds, while optional in the construction of a single-panel work, are essential and inescapable parts of the meaning structure of triptychs.

The book’s attention to the role of format in the construction of meaning places its methodology outside the paradigm of “disguised symbolism.” Since the concept of disguised symbolism was first introduced by Erwin Panofsky in his 1934 article on Van Eyck’s Arnolfini Wedding, iconographic studies of early Netherlandish painting have often been dominated by an obsession with decoding symbols.87 And even after several decades of criticism of the notion of disguised symbolism largely eliminated the decoding mania, scholars of Northern Renaissance art still tend to cast questions of meaning in terms of the opposition between symbol and reality.88 This book argues that Netherlandish paintings can communicate meaning in other, nonsymbolic ways and that format is one of these. For this reason, the book is as much about how art means as it is about what art means.89

The book’s treatment of Netherlandish triptychs as “paintings with doors”—and its unpacking of the implications of the treatment of format for meaning—results in new ways of interpreting and understanding these complex artworks. These new modes of interpretation are not intended to supplant traditional iconographic studies but rather to expand them beyond the limitations of purely symbolic interpretation. My approach builds upon a general scholarly interest in the margins and framing of art (as, for example, in Michael Camille’s Image on the Edge: The Margins of Medieval Art), and its analysis of spatial passageways complements Alfred Acres’s study of temporal passageways in Netherlandish painting.90 The reinterpretations of Netherlandish triptychs in this book thus are often quite compatible and even cotenable with those proposed by earlier scholars. But, I believe, once one reconceptualizes Netherlandish triptychs as “paintings with doors,” these artworks never quite look the same or mean the same as they seemed to before.

This book deals primarily with painted triptychs of the fifteenth and sixteenth centuries. Within the book I classify works as triptychs only if they (1) are composed of separate panels, whose divisions are visible and thereby establish thresholds between the various sections, and (2) have a stressed center, or, to use Lankheit’s term, a “subordinating center” (that is, in some way mark off the inside center as the most important zone within the total structure and concomitantly make the other parts of the work subordinate to it). Under this definition, what counts as a triptych here is not only the standard type, with hinged, folding wings, but other variants as well. Triptychs with fixed (that is, nonfolding) wings still count, since they, though lacking any interior/exterior thresholds, nevertheless have thresholds and a stressed center within their fixed, open state.91 So, too, frieze-type triptychs, which have equal-sized panels (rather than the double-sized center of the standard type), also have thresholds and a stressed center, even if the center is stressed only by its centrality and not by increased size. In addition, triptychs—either with fixed or folding wings—that have one scene painted continuously across all the panels still meet my definition of a triptych, since the separation of the panels in and of itself provides thresholds and a resultant emphasis on the center.

My definition of triptychs also includes works typically called polyptychs, which are made up of more than three panels. Although they have more panels than the standard triptych, most of these are structured as a series of thresholds with a subordinating center and folding wings, just like regular triptychs. Indeed, the documents make no distinctions between triptych and polyptych, calling both panels or paintings with doors.92 For these reasons, this book treats “polyptychs” with thresholds and stressed centers, such as Jan van Eyck’s Ghent Altarpiece (however, a quadriptych, which has thresholds but no stressed center, would not qualify for inclusion here).93 Triptychs or polyptychs that combine sculpted centers with painted wings also satisfy my definition of “triptych” and indeed meet the criteria particularly well in that the shift in media heightens the sense of threshold and especially the differing statuses between the divided zones (with the sculpture of the center dominating over the wings to a degree not possible in single-media works). However, since the production of sculpted altarpieces was a complex industry that functioned fairly separately from that of painted works—and has been examined in my previous book94—sculpted works will only be considered here insofar as they relate to pictorial developments. So, too, diptychs are considered only peripherally here. To be sure, diptychs have thresholds in that they consist of two panels hinged together, and many documents refer to the panels of a diptych using the same terminology used for triptychs, that is, “doors” or “leaves.”95 But diptychs lack a subordinating center, and hence the dynamic governing the doors of the diptych differs somewhat from that governing the doors of the triptych. The specific dynamic of the diptych, of late, has received significant attention elsewhere.96

This book surveys Netherlandish painted triptychs from the early fifteenth century through the late sixteenth century, ending with a coda on the triptych in the age of Rubens. As such, it offers a more comprehensive study of the format than can be found elsewhere. The chapters are focused on individual painters or groups of painters (arranged chronologically) and treat the material both thematically and via detailed case studies. The case studies examine individual triptychs from a variety of points of view, considering their formal elements, iconography, physical structure, patronage, and historical context; but above all, the analysis centers on the ways in which the thresholds of the triptych are treated and on the impact of this treatment on the meaning of the work. First, however, I would like to provide some background regarding the functions and sources of the early Netherlandish triptych.

The Functions of the Early Netherlandish Triptych

In the late Middle Ages and Renaissance, triptychs were experienced in a variety of ways because of their multiplicity of functions. Modern scholars typically associate triptychs with the function of altarpiece, that is, with decorating the altars in churches and thereby forming a backdrop to the celebration of the Mass, as was the case for triptychs such as Bouts’s Holy Sacrament altarpiece (referred to in a 1464 contract simply as a “panel” [tafele]) and Jan van Scorel’s Crucifixion triptych for the New Church in Delft (referred to in a 1550 contract as a “high altarpiece” [hoogh autaer-stuck]).97 A 1390 document records a payment from the duke of Burgundy to Jean de Beaumetz “for a painting with two halves closing over, which has in the middle the Coronation of Our Lady and on one of the sides the Annunciation and on the other the Visitation of Our Lady and Saint Elizabeth, and all are gilded with fine gold—which paintings my lord has had placed on the altar of the chapel at the Chartreuse” (pour uns tableau a deux demis sur clouans esquels a ou milieu le couronnent Nostre Dame et en l’un des cotez l’Annonciacion et en l’autre l’Acolement Nostre Dame et de saincte Helisabeth, et sont tous dorez de fin or, lesquelz tableaus mon dit seigneur a fait mettre sur l’autel de la chappelle, aux Chartreus).98 So, too, a document from the Sisters of Notre Dame of Sion in Bruges records a 1509 donation of “a beautiful panel in oil, placed on the high altar . . . made and given by Master Gerard David” (een schoon tafele van olyvarwe, staende up den hooghen outaer . . . ghemaect ende ghegheven by meester Gheeraert David); the work’s status as a triptych is evidenced by the statement that “the doors of the aforementioned panel were unpainted on the outside and also inside” (De dueren van der vorscreven tafele waren buuten ongheschildert ende ooc binnen).99

But not all triptychs in churches were altarpieces; many served as epitaphs. For example, the Church of Saint Peter in Leuven had a number of epitaph triptychs, including that of Marthe Oliviers, noted in a 1582 testament stating that she “desired to be placed above her and her husband’s tomb her painting of our beloved Lady, standing on the sun, [with] Saint John the Baptist and Saint John the Evangelist on the doors” (begeert boven haer en haers mans sepulture gestelt te worden haer taeffereel van onser liever Vrouwen, in de son staende, op de doere Sint Jan Baptista en Sint Jan Evangelista).100 The 1505 inventory of the house of Cornelis Haveloes lists “a painted painting [made] of wood with two small doors [depicting] our beloved Lady and painted therein is the aforementioned late Cornelis Haveloes, which he wished to be set in the place near [where he] is buried” (een geschildert tavereel van houte met twee doerkens van onser liever Vrouwen, ende dair inne geconterfeyt is die voirschr. wylen Coernelys Haveloes, d’welck hy begeert heeft geset te worden ter plaetsen daer by begraven is).101

Other triptychs in churches commemorated the donors and served as foci for prayers for their souls without appearing directly over their tombs. The records of the convent of the Sisters of Notre Dame of Sion show a donation of “a little painting in which there is the Holy Trinity and in the first door Saint Judoc and in the second Madam Van Hecke, the esteemed widow of Sir Wouter van Hecke, the late alderman of the Freemen; she asked to have this aforementioned painting hung in the choir of the sisters so that they would be urged to pray for her soul” (een taeffereelke daer in staet de Heleghe Drievuldicheit, ende in de eerste duere Sinte Joos, in de tweeste joncffrauwe van Hecke de eersaemme wedwe van d’heer Wouter van Hecke, wylen schepen van de Vrie; dit vorscreven taeffereel begheerde zy up der susteren choor ghehanghen te hebben om dat zy te bedt over haer ziele bidden zouden).102

But this Holy Trinity triptych functioned quite differently from another Trinity triptych in the same convent (and most likely in the convent church), a reliquary triptych made in 1555, which was described as “a large painting with two doors painted with gold letters, [with] the Holy Trinity inside, and decorated with relics of different saints and also with silk flowers” (een groot taeffereel met ij dueren bescreven met gulde letteren, binne staet de Heleghe Drievuldicheit, ende verchiert met helichdom, van diversche heleghen ende ooc met zyde blommekins).103 Such a work would function within the relic cult, rather than for the benefit of the donor’s soul. This work may also have functioned within the realm of private or personal devotion,104 for the gold letters mentioned in the document could reflect the relatively common practice of writing devotional prayers, sometimes indulgenced prayers, on the wings of triptychs.105 Among the surviving examples of triptychs with such texts are a half-length version of Rogier van der Weyden’s Descent from the Cross in Madrid, with Saint Gregory’s Seven Prayers on the wings, a prayer that was indulgenced if read in front of the Man of Sorrows;106 a Madonna and Child by a follower of Rogier in Tournai, with wings bearing a prayer to the Virgin, “Salve Regina”;107 and a Virgin and Child in London, now attributed to a follower of Hugo van der Goes, with wings (originally from a triptych with a central indulgenced image of the Virgo in Sole) that contain a prayer about the Immaculate Conception.108

Not all triptychs, however, hung in churches. In the convent of the Sisters of Sion, at least one triptych was located in the cell of a nun, presumably for her personal devotions: the convent records list “a little painting of Mary, painted [in] oil, with two little doors, on one of which is Saint Anthony, and on the other Saint Petronilla, which image or painting has been given to the church by our dear fellow sisters, Sister Marie and Sister Ursula van der Houve, two sisters. Sister Marie has [it] in her cell” (een beeldeke van Maria, olie varruwe, met ij duerkins, in ’t dat een staet Sinte Anthuenis, ende in ’t dat andere Sinte Petronelle, welck beelde of taeffereeke ghegheven es gheweest in de kercke van onse beminde mede susteren, suster Marie ende suster Ursula van der Houve, ij ghesusteren.—Suster Marie heeft up haer celle).109 Still others were located in secular settings such as town halls. Documentation of Dieric Bouts’s triptych of the Last Judgment indicates that in 1468 “this Judgment [hung] in the councilmen’s room in the town hall of Leuven” (hetwelcke Oordeel hancht in de schepene Camere opt Stadthuys te Loven).110 And some triptychs were located in the home: a 1418 testament speaks of “a small panel with two leaves that is in the room above the washing place, in which there is, in painting, the Crucifixion” (ung tavelet à deux foellés, lequel est en le salette deseure le lavoir, où est en pointure le cruchefiement).111 The inventory of Cornelis Haveloes of 1505 lists in the dining room (eetcamere) “a small painting of an image of Mary, painted, with two little doors” (een cleyn tavereelken van eene Marien beelde, geschildert, mit twee doerkens).112 Another document, of 1489 (discussed above), records the presence of a triptych in the dining room of an inn in Leuven, where it stood on what was designated an altar.113 Within homes, triptychs were often displayed on top of cabinets called dritsoer or dritshoer in Flemish, dressoir or trésor in French. Netherlandish fifteenth- and sixteenth-century paintings testify to this practice—as well as to the presence of triptychs in the home—in their depictions of triptychs within home settings: Joos van Cleve’s Annunciation in New York (fig. 3), for example, stages the Annunciation within a domestic environment that includes a wooden cabinet, covered with a white cloth, on which stands a partially opened triptych with a grisaille exterior and a polychrome interior.114 The inventory of Cornelis Haveloes also suggests that triptychs could be displayed on top of—or be stored within—the dritshoer in the house: one listing in the inventory appraises the “wooden cabinet” (scrynhouten dritshoer) and then lists items “found on top of and inside [it]” (daer op ende inne bevonden).115 So, too, the inn in Leuven included “one round cabinet, closed, on top [of which] one painting of Saint Christopher” (1 rondt tritscer gesloot daerop 1 taverneel van Sinte Christoffel).116

The location of a triptych, however, was not necessarily fixed. Triptychs could migrate from homes into churches after the deaths of the owners. This was the case for the two epitaph triptychs discussed above: the Marian triptych, which Cornelis Haveloes had in his house and wanted placed near his grave, and the Virgin in the Sun triptych of Marthe Oliviers, which the 1582 document calls “her painting,” a sign that the work was in her possession in her home at the time she requested its future placement over her and her husband’s grave.117 In addition, the Baptism triptych by Gerard David (figs. 67, 68), as recorded in a document of 1520, was transferred by the heirs of Jan des Trompes from his personal effects in Jan’s home to the Brotherhood of the Sworn Clerks of the Tribunal (a lawyer’s confraternity) “on the condition that the same sworn clerks shall be obliged to have placed a stone or brass panel in the wall next to the altar in the aforementioned chapel and therein have [the panel] engraved for the consecrating given the aforementioned painting” (met condicien dat de zelve gheswooren clercken ghehouden zullen werden te doen stellene eenen steen of lattoenen tafle in den muer neffens den outaer in de voorseide cappelle ende daer inne doen graveren by wien de voorseide tafle ghegheven).118 Thus one and the same triptych could be moved from one context to another and thereby experienced in a very different ways: the simple act of transport, without any additional alterations, could turn a home furnishing used for private devotion into a public ritual object or a tomb decoration used for memorial prayers.

Triptychs not only could change location but also could mutate in their very format, with wings altered, added, or eliminated. Thus, for example, alterations were made to the wings of the triptych that Gerard David donated to the Sisters of Sion in 1509, as is evident from the record noting that in 1536 the wings, which were unpainted on the interior and exterior, were “taken off in order to paint and to color” (afghedaen om te schilderen ende te varwen).119 So, too, physical evidence of an altarpiece in the Utrecht Catharijneconvent—which combines a central panel from the shop of Pieter Coecke van Aelst with donor wings by the Amsterdam painter Dirck Jacobsz—indicates that the triptych was first furnished with wings painted in a solid dark color, and at a later date donor portraits were painted on top of the monochrome layer. Most likely this reflects a market practice of selling triptychs with unfinished wings on the open market; after purchase, the wings could then be completed according to the buyer’s specifications.120 On the other hand, a document from the convent of the Sisters of Sion in Bruges attests specifically to the addition of wings to a single panel: the document cites “a panel, the Pietà made there in oil paint, which came from Sister Melchie Heindericx in 1518, whereon our reverend father confessor, Brother Lievin de Vos, made a leaf or door in order to close [it], etc., therein representing Saint Albert and Saint Elizabeth, before whom is kneeling a brother in memory of our very beloved father, Brother Yzenbart de Bru” (een bardt, de Noodt van Maria daer in ghemaect olyvarwe, ’t welc quam van suster Melchie Heindericx anno XVc xviij; daer an heeft ons eerweerdich pater confesseur, broeder Lievin de Vos, een blat of duer an doen maken om te sluuten, etc., daer in staende Sinte Albertus ende Sinte Lisbette, daer vooren knielende een broeder ter ghedynckenesse van onsen zeer hertelicken beminden pater, broeder Yzenbart de Bru).121 Physical evidence also can point to later additions of wings, as, for example, in the case the Mérode Triptych (to be considered more fully in chapter 1), where overpainting and dendrochronological findings suggest that the wings likely were added later to a preexisting central panel.

All these changes often involved corresponding changes in function. Such a change would surely have followed if the terms of the 1470 testament of Anselme Adornes had been fulfilled. Before his dangerous trip to Jerusalem, Adornes drew up a will (which has been associated with Eyckian paintings of Saint Francis in Turin and Philadelphia) that states, “Likewise so do I give each of my beloved daughters that are cloistered, that is, Margriete the Carthusian and Lowyse Sint Truden, a painting in which there is Saint Francis in [a] portrayal by Master Jan van Eyck’s hand, and [direct] that there shall be made, in the little doors that close the aforementioned small painting, my person and my wife” (Item zo gheve ic elcken van myne lieve dochters die begheven zyn, te wetene Margriete, tSaertreusinnen, ende Lowyse Sint Truden, een tavereel daerinne dat sinte Franciscus in potrature van meester Jans handt van Heyck ghemaect staet, ende dat men in de duerkens die van de zelve tavereelkins beluicken, doe maken myn personage ender mer vrauwe).122 It is uncertain whether the testament is calling for donor portraits to be added to already existing wings or for wings to be added to single-panel works to convert them into triptychs.123 But in either case, the changes would have added a new, memorial function to the work, as well as new ways of structuring meaning. For when the work changed from a painting to a “painting with doors,” it communicated meaning in a new way. And, as I argue in my discussion of the Mérode Triptych, the artists involved in adding wings to single panels were quite aware of this phenomenon.

Of course, things also changed when wings were removed from triptychs. The practice of removing wings from triptychs is not much documented, but our museums provide ample evidence of how the component panels of triptychs could be separated from one another and dispersed. But this part of the story lies outside the scope of this study.

Sources of the Early Netherlandish Triptych

Although the first known triptychs go back to ancient Rome and Roman Egypt, the early Netherlandish triptych is rooted largely in traditions established in Byzantine art.124 To be sure, the Netherlandish triptych, like most other art forms, was influenced by a multiplicity of sources, many indigenous, such as northern European furniture, tabernacles, altarpieces, and reliquaries. Nevertheless, direct lines of descent lead from Byzantine triptychs to the first triptychs in the West (in the twelfth century)—and from these, ultimately, to the Netherlandish triptych, which emerged around 1400. So just as studies of Netherlandish devotional imagery have demonstrated a dependence on Byzantine icons, examination of the Netherlandish triptych reveals a parallel dependence on Byzantine triptychs.125

The triptych format is found in some of the earliest Byzantine artworks. A full third of the early Byzantine Christian icons found in the Monastery of Saint Catherine at Mount Sinai originally formed parts of triptych structures.126 One of the earliest surviving, fully intact Byzantine triptychs is a ninth- to tenth-century work (fig. 4) with the Ascension in the center, Saint Theodore and Saint George in the wings, and a yellow cross on the reverse (this latter, a standard decoration for the backs of icon triptychs).127 But fragments of earlier triptychs are among the Sinai icons, including examples going back to the sixth and seventh centuries.128 Kurt Weitzmann has argued that the small scale of these triptychs shows that they were intended for private worship; indeed, Weitzmann sees the triptych as particularly suitable within the context of personal devotion directed at holy images, since the format, opened only when being worshipped, created a very dramatic effect.129 Even larger-scale triptychs found in early Byzantine art—for example, the seventh-century Elijah, which once formed the wing of a triptych at least 60 cm high—would not likely have stood on the altar, but against the walls of the nave or chapel, since at this time only the implements of the Mass and the Gospel lectionary (and not panel paintings) would have been allowed on the altar table itself.130

Nevertheless, Byzantine icon triptychs did have links to the altar because of their relationship to the doors of the templon. The templon—an architectural screenlike structure that separated the nave from the altar—came to be outfitted with doors, of which the central and largest was called the “royal door.”131 These doors were normally closed, except during services, when they would be open and the priest would emerge from behind them to serve Communion; this, together with other openings of the royal doors into the sanctuary of the church, provides a liturgical parallel to the opening of the doors of the icon triptych for personal worship.132 Indeed, Byzantine triptychs and royal doors seem to have been closely related art forms, which shared a number of formal analogies: the shapes of the wings of icon triptychs were similar to those of the royal doors, with both forms shifting from rectangular to more circular forms around the twelfth century.133 So, too, both royal doors and icon triptychs often included the scene of the Annunciation in their upper sections. The Annunciation appears in icon triptychs as early as the sixth or seventh century, for example, on the upper inside part of the wings of a heavily damaged icon from Mount Sinai.134 Such parallels suggest that already in Byzantine painted icon triptychs, there was a definite association—on functional, formal, and iconographic levels—between wings and doors. And since the royal doors of the templon (and later of the iconostasis) were closely associated with both Mary and the doors to paradise, it is likely that the doors of icon triptychs carried similar implications as well.135

Another popular type of Byzantine triptych was the ivory triptych of the middle Byzantine period (ninth to twelfth centuries).136 One of the most common formats for the ivory triptychs included a central image of the Virgin Hodegetria (that is, the Virgin and Child depicted as in the original icon in the Church of the Hodegon), with angels and saints on the wings, arranged on several registers.137 Other ivory triptychs had Crucifixion scenes in the center—or sometimes the devotional image of Christ on the Cross flanked by Mary and John, rather than a fuller narrative—with a number of tiers of saints and angels on the wings. A particularly well known type is that represented in three famous examples—one in the Louvre (the Harbaville Triptych, fig. 5), one in the Palazzo Venezia, Rome, and one in the Museo Sacro in the Vatican)—in which all three panels of the triptych are divided into two tiers (unlike the Hodegetria or Crucifixion types, where the center is a single scene), with the top center tier showing the Deësis and the remainder presenting rows of saints. Typically these middle Byzantine ivory triptychs, like the early Byzantine icon triptychs, had crosses on the reverses of the wings (either single crosses that spanned both wings or crosses on each side), sometimes with plants, birds, or landscape elements placed in the negative spaces between the elements of the cross; on occasion these triptychs instead had figurative elements on the exterior.138 The exact conditions of display and use of these works remain uncertain, but these middle Byzantine ivory triptychs, like many Byzantine painted icon triptychs, probably served in the domain of personal worship.139 The middle Byzantine period also saw the use of the triptych for metalwork reliquaries of the True Cross. The True Cross formed the main reliquary type in the East, unlike the West, where saint reliquaries were favored.140 As far back as the eighth century, enamel reliquaries of the True Cross were made in Byzantium in a triptych format, but production of these reliquaries flourished particularly in the tenth and eleventh centuries.141

The influence of Byzantine triptychs on art in western Europe becomes evident beginning in the twelfth century in the Mosan region of the Lowlands, which was a center for the production of reliquaries of the True Cross. The earliest known Mosan True Cross reliquary is the Stavelot Triptych in the Pierpont Morgan Library (fig. 6), dated in the 1150s.142 The center of this metal-enamel work (whose wings contain Mosan medallions showing the Legend of the True Cross) actually incorporates two eleventh-century Byzantine enamel triptychs, which Wibald, abbot of the monastery of Stavelot, is believed to have brought back from the Byzantine court.143 By including these two Byzantine triptychs within a Western triptych, the Stavelot Triptych not only comments on its own format but also identifies the source of that format as Byzantium.144 The larger of the two Byzantine triptychs contains the relic of the True Cross. This relic, which constitutes the sacred core of the whole work, itself forms a cross at the middle of the center panel; the relic is flanked at the bottom by Constantine and Helena, and at the top by two archangels, while the wings show four Byzantine military saints on the inside and the four Evangelists on the outside.145 The smaller Byzantine triptych, above, depicts the Crucifixion in the center of its interior (with decorative motifs on the inner wings) and the Annunciation on its exterior (fig. 7). The appearance of the Annunciation on the exterior of this Byzantine triptych harkens back to the royal doors and the upper inner sections of the wings of icon triptychs. But it also points forward, standing as evidence that the depiction of the Annunciation on the exterior of early Netherlandish triptychs—one of the most characteristic features of the type—had sources in Byzantine art.146 These Mosan True Cross reliquaries also helped forge other connections between Byzantine art and early Netherlandish triptychs, for Mosan triptychs, most notably the True Cross reliquary in Liège (fig. 8), sometimes incorporated, in the wings and center, the arched profile of twelfth-century Byzantine royal doors and triptychs, which later was adopted in some early Netherlandish triptychs.147

The impact of the Byzantine triptych in the West became even more pronounced with its strong influence—both in style and usage—on Italian art of the twelfth and thirteenth centuries. So, for example, in the early twelfth century the cathedral of Tivoli acquired a triptych that had a life-sized image of Christ in the center and the Virgin and Saint John in the wings and was carried in procession and anointed during the feast of the Assumption, following Byzantine practices of icon worship.148 Similar uses of triptychs are found in the domestic realm: records of 1252 indicate that Roman women frequently owned, and venerated in their homes, icons outfitted with closable wings.149 At this time, too, the Franciscans adopted the painted triptych as their main altarpiece type; typically these altarpieces had an image of Mary in the center and folding wings that could cover the icon as needed.150 By the thirteenth and early fourteenth centuries triptych production also flourished in France, particularly Paris. These Parisian triptychs generally were small-scale works, produced either in ivory (e.g., the early fourteenth-century triptych of the Life of the Virgin in the Louvre, fig. 9) or in metalwork (e.g., the mid-thirteenth-century work from Floreffe in the Louvre, another True Cross reliquary in a triptych format).151 Parisian ivory triptychs differ from their Byzantine counterparts in their inclusion of Gothic forms (pointed-arch profiles, trilobed framing of scenes), but may have a direct source in middle Byzantine painted and ivory triptychs that depict narrative scenes associated with Orthodox feasts.152 The large-scale production of ivory triptychs in Paris (and also in Amiens, another key center at the time) points to a growing use of the triptych format in the Gothic North.153

By the time of the production of the first early Netherlandish painted triptychs, around 1400, the triptych was very well assimilated into western European artistic production. In Trecento Italy, which still retained strong Byzantine influences, some triptychs were used to display Byzantine-derived cult images for personal devotion, such an early fourteenth-century Venetian Man of Sorrows triptych (fig. 10);154 others, such as Pietro Lorenzetti’s Birth of the Virgin triptych (fig. 11) and his Arezzo polyptych, were large-scale triptychs or polyptychs without moving wings and functioned as altarpieces.155 In the later part of the fourteenth century many small folding triptychs, often serving as reliquaries, were produced in Siena.156 The reliquary in Baltimore, a triptych by an anonymous Italian artist (fig. 12), is a typical example: the work holds relics in its pointed gable and lower frieze, while depicting the Virgin and Child with saints in the center, two saints on each of the wings, and, in another carryover from Byzantine traditions, the Annunciation (with Gabriel on the left wing and Mary on the right) in the triangular tops of the upper register of the wings.157 Other Trecento Sienese triptychs, following other strands of Byzantine tradition, combined devotional imagery in the center with several tiers of narrative imagery in the wings.158 Small triptychs were also produced in Venice, for example, the Madonna of Humility triptych (fig. 13), which has a central image of the Madonna and Child surmounted by an image of Calvary; the wings show standing saints and an Annunciation at the top.159 In this work, the Madonna of Humility and the standing saints are all enclosed within gilded arched frames, which are rendered in actual relief, not just pictorially.

In the North, around 1400, there were a number of different triptych types, many combining both painting and sculpture. These included joyaux (small items in precious metals and gems), such as the tiny Man of Sorrows gold and enamel reliquary triptych probably produced in Paris (fig. 14), the Passion Altarpiece of the Chartreuse de Champmol sculpted by Jacques de Baerze with wings painted on the exterior by Melchior Broederlam, and the Mosan Cardon Tabernacle and Mayer van den Bergh Tabernacle (each with a central statue and painted wings, ca. 1400).160 Among the fully painted Netherlandish triptychs produced around 1400 is the Calvary and Saints triptych, a small-scale reliquary triptych (fig. 15).161 This triptych, which, when opened, measures 54 by 50.5 cm (21" × 20"), has three tiers of arched compartments in the center and wings, each of which contains a depiction of a saint, presumably the saint whose relics were originally embedded into the wood below each depiction. The upper section of the triptych is triangular, with saints depicted in the wings and a Calvary group in the center. This triptych is similar to Trecento Sienese, Byzantine-inspired triptychs—like that of figure 12—in its reliquary function and its triangular profile, although the triangular type is also found in thirteen- and fourteenth-century Parisian triptychs.162 The arched enclosures around each figure of the Calvary and Saints triptych are also similar to framing elements found in both large- and small-scale Trecento Italian triptychs (e.g., fig. 13), as well as Mosan True Cross reliquary triptychs (e.g., fig. 8) and, via these intermediaries, link to Byzantine triptychs of this type.163

Another very early Netherlandish painted triptych is the Norfolk Triptych (color plates 2, 3), which was probably produced in the Mosan town of Maastricht around 1415.164 Like the Calvary and Saints triptych, this is a small-scale work (33 × 58 cm, opened, or 33" × 58") that also served as a reliquary—specifically, since it has a wooden plug inserted into the oak, a True Cross reliquary. This function links the work to Byzantine, Italian, and, perhaps most important, local True Cross reliquary triptychs, that is, those produced in the Mosan regions during the Romanesque period.165 The Norfolk Triptych’s interior (color plate 2) is arranged in two registers, with some additional imagery in the top extensions, a format that likely derives from sculpted altarpieces.166 Each tier is divided up into compartments, not unlike those of the Calvary and Saints triptych (fig. 15), and each compartment is framed with simulated architecture rendered in gray monochrome, or grisaille. The double-arched frames enclosing many of the compartments here are particularly reminiscent of the double-arched profiles of Mosan Romanesque triptychs, such as that in Liège (fig. 8). Most of these compartments—again like Byzantine examples and Byzantine-inspired Western triptychs—portray standing saints, mostly arranged in pairs. At the very center of the Norfolk Triptych are two compartments that depict the Coronation of the Virgin at the top and the Angels’ Pietà on the bottom. This program derived directly from metalwork examples that combined these two themes, such as the Parisian gold and enamel reliquary triptych in the Rijksmuseum (fig. 14).167 But the Man of Sorrows theme itself, in the metalwork sources and the Norfolk Triptych, is one that first emerged in Byzantine icons.168

The Norfolk Triptych’s exterior (color plate 3) is devoted, on the upper tier of the two primary registers, to the Adoration of the Magi, which is depicted across both panels. The left half of the scene actually includes a figure thought to be a donor, standing behind the stable of the Nativity, with a gift in his hand.169 The bottom tier reverts to images of standing saints. In the topmost tier of the closed triptych is the Annunciation, with the Virgin at the left and the angel at the right. This element, as we have seen, is closely linked to Byzantine traditions (e.g., fig. 7) and to Byzantine-inspired, Trecento Italian reliquary triptychs (e.g., figs. 12, 13), although in the Italian works the Annunciation tended to appear on the upper interior, rather than the upper exterior, of the wings.

But perhaps the most significant influence that the Byzantine triptych exerted on the Norfolk Triptych, and on succeeding Netherlandish examples, was in fostering an association between triptychs and doors. The Norfolk Triptych is clearly structured as a painting with doors. Its interior, visible when the doors are opened, is the most holy zone, with the imagery sanctified through the use of gold-leaf backgrounds and the placement of framing elements around the devotional imagery.170 The center of the center is reserved for the most sacred imagery, showing the divine Christ in heaven crowning the Virgin above the suffering human Christ as Man of Sorrows below.171 The imagery on the doors, or wings, is distinctly separated from the imagery in the center. On the inside of the wings, the pictorial frames around each group of saints are supplemented by actual frames dividing the three panels; indeed, the presence of the frame is emphasized by its elaborate coloring (blue, gold, and red) and its decoration with both pictorial elements and sculpted rosettes.172 This creates a sealed threshold between center and wings of the opened triptych.

The threshold between the exterior and interior of the “painting with doors” is especially strongly marked and distinguished in the Norfolk Triptych. On the exterior (color plate 3), the figures do not appear within sealed compartments, but within open spaces. Whether part of narrative scenes—the Annunciation and the Epiphany (on the top extension and the upper tier, respectively)—or groups of saints (on the bottom tier), they are set within natural landscapes with blue skies that contrast with the architectural niches and gold-leaf backgrounds of the interior. As a result, the figures of the interior seem to exist within a heavenly structure, in eternal time, whereas those on the exterior exist within an earthly, outdoor zone, which includes narrative elements that suggest a more temporal dimension (especially if the contemporary donor is in fact included in the Epiphany). The compartmentalization of the interior asserts sanctity, while the less divided nature of the exterior makes the figures more accessible: though the exterior is divided pictorially into registers and is painted on two separate panels, the landscapes appear to link across the horizontal tiers, thereby downplaying the divisions of the format.

Thus the Norfolk Triptych manipulates the structure of a “painting with doors” to make the thresholds inhering in the format—both within the opened view itself and between the open and closed views—a key part of meaning, specifically, the way in which the work distinguishes the holy from the less holy. The Norfolk Triptych thereby sets the stage for the manipulation of the zones within Netherlandish triptychs of the 1420s and beyond. These works, considered in the chapters that follow, develop the triptych/door analogies rooted in the Byzantine triptych tradition to more fully exploit the nature of the triptych as a painting with doors.