This report draws on a combination of primary institutional data, peer-reviewed research, major art market surveys, and journalism. Where possible, figures are cited to their original published sources. Readers are encouraged to verify current statistics independently, as the art market and workforce data evolve year on year.
One source that informed this report’s genesis deserves specific acknowledgment. Art historian Katy Hessel discussed the institutional exclusion of women from the Royal Academy of Arts in a recorded conversation with Pandora Sykes. The interview is available to subscribers of Intelligence Squared Plus. Because it is paywalled and no official transcript exists, it has not been quoted directly; however, the documented history of the Royal Academy fully corroborates Hessel’s central claims, and primary and secondary sources establishing those facts are cited throughout Part I.
This report occupies unusual scholarly territory. No systematic account of mosaic workshop practice across cultures and periods currently exists. Art historians have studied individual commissions in depth — the mosaics of Ravenna, the Norman cycle at Monreale, the Byzantine achievement at Hagia Sophia — but the mechanics of how those works were made, how labor was divided, and how (or whether) credit was assigned within the workshop have remained largely subordinate to iconographic and formal analysis. This report draws those mechanics into the foreground.
The principal sources are: Pliny the Elder’s Natural History for ancient workshop practice; Byzantine art history scholarship (including Demus’s foundational work on Byzantine mosaic programs and the Britannica synthesis of medieval mosaic); primary chronicle sources including Leo of Ostia’s account of Desiderius’s Montecassino program; Smarthistory’s peer-reviewed materials on Norman Sicily; archaeological research on sinopiae from the southern Levant; and art market and contemporary studio documentation including the MyArtBroker analysis of Renaissance and contemporary workshop practice, documentation of Sol LeWitt’s attribution methods, and press coverage of Damien Hirst’s studio production.
For contemporary community attribution practice, the primary sources are the Chicago Public Art Group’s own documentation, the CPAG archive as documented by Sepia Archival Management, and the primary documentation of the Community Harmony Through Song and Play project at Manly Career Academy High School (CPAG, 2006), which is treated throughout as a primary case study. The Manly project documentation is available at rachaelquevargas.com.
Where scholarly consensus is absent or contested — particularly on the question of whether Byzantine mosaicists worked in organized workshops or as looser teams of craftsmen — this report acknowledges the uncertainty and proceeds with the weight of the evidence. The absence of systematic attribution data for Byzantine and medieval mosaic production is itself a finding, and is treated as such.
In the apse of San Vitale in Ravenna, completed around 547 CE, stands the Empress Theodora. She is rendered in Byzantine smalti — tiny cubes of colored glass fired with metallic oxides, some backed with gold leaf — her gaze fixed on eternity, her hands extending a golden chalice toward the altar. To her left and right, her court attendants process in the stiff, frontal formality of Byzantine iconography. Light catches the tessellation at oblique angles, making the gold surfaces shimmer as a viewer moves through the space.
Someone decided how that shimmer would work. Someone chose the andamento — the directional flow of the tesserae — that makes the mosaic feel alive under candlelight. Someone made the tens of thousands of individual decisions about which piece of glass to place where, at what angle, with what degree of irregularity, to produce the visual effect that has lasted fifteen centuries. That person’s name does not exist in any record.
The work is attributed to Justinian I, who commissioned it. The program is attributed to Bishop Maximianus, who dedicated it. The style is attributed to the Byzantine imperial tradition. The hands that made it are not attributed to anyone. They were craftsmen brought from Constantinople — the best in the empire, presumably, given the scale and ambition of the commission — and they are invisible in the historical record with a completeness that is structural rather than accidental. The attribution system of their era had no mechanism for naming them. Ours has barely improved.
This is the problem The Workshop Tradition addresses. Not as an art-historical curiosity, but as a load-bearing institutional question. The Invisible Workforce commits the MME to developing “the first mosaic-specific attribution standard for workshop and collaborative production.” The Outsider Art System builds an acquisition ethics framework that depends in part on the integrity of attribution claims. Neither commitment can be made fully implementable without the historical foundation this report provides — because that standard, to be specific rather than aspirational, must understand what workshop production in this medium actually looked like across the periods of its highest achievement.
The central argument is this: the genius-artist attribution model the art world currently uses — one named conceptual author, invisible labor — was never the model under which mosaic’s greatest works were made. Ravenna, Hagia Sophia, the Norman mosaics at Palermo, Monreale: all workshop productions. The medium’s highest achievements are structurally incompatible with the attribution framework the contemporary art world employs. That is not a gap in the research. That is an argument. And it has direct consequences for how the MME will credit artists, train students, commission work, and present its collection.
A secondary argument runs alongside the first: the problem of attribution in mosaic workshop production is not only historical. The art world has not solved it in the contemporary period. Damien Hirst concedes publicly that the best spot paintings that bear his name were painted by Rachel Howard. Rachel Howard’s name appears in interviews. It does not appear in catalog entries. The field has a word for the person whose name goes in the catalog. It does not have adequate words for anyone else. The Workshop Tradition proposes those words.
There is one mosaicist whose name survives from the ancient world. Pliny the Elder, writing in the first century CE, declared that “in [mosaic], the highest excellence has been attained by Sosus, who laid, at Pergamon, the mosaic pavement known as the Asarotos Oikos.” The Unswept Floor — a trompe-l’oeil pavement depicting the debris of a banquet, scattered as if left by accident — was, Pliny wrote, a marvel of illusionistic skill. Sosus also made a mosaic of doves drinking from a bowl, one of which cast the shadow of its head upon the water.
Neither of these works survives in the original. What survives are copies. The most famous copy of the Doves was excavated from Hadrian’s Villa at Tivoli in 1737 and is now in the Capitoline Museums in Rome. It is made of tiny cubes of colored marble — opus vermiculatum at its finest — and it is signed by one Herakleitos. Herakleitos made a copy of Sosus’s work, signed his copy, and exists in art history as a footnote. Sosus received the credit. Herakleitos provided the survival.
The fact that Pliny recorded Sosus’s name at all was exceptional enough that it has been noted by every art historian who has written about ancient mosaic. Sosus is the only mosaicist named in ancient literature — which means the thousands of other craftsmen who made the mosaic floors of Pompeii, the hunting scenes of the Villa Romana del Casale, the narrative cycles of North African villas, and the Nile mosaic at Palestrina are unnamed not because they were unskilled but because the attribution conventions of their era did not extend to them. Skill was recognized. The individual craftsman was not.
This was not universal even in antiquity. The Stag Hunt mosaic in the House of the Abduction of Helen in Pella, Greece — considered the finest Hellenistic floor mosaic known — is signed by Gnosis. Mosaics from Palace V at the Pergamon Acropolis bear the signature of Hephaistion. Signatures existed as a convention for the finest individual emblema work — the small, high-quality panel produced in the workshop and transported to the installation site. They were exceptional, not standard. The vast majority of ancient mosaic production was unsigned and unattributed.
The Roman mosaic industry was organized around a fundamental technical distinction that is directly relevant to the four-role framework The Workshop Tradition proposes. There were two primary modes of production: opus vermiculatum, using tiny tesserae, produced in workshops in relatively small panels which were transported to the site; and opus tessellatum, using larger tesserae, which was laid on site. The distinction was not merely technical. It encoded a division of labor.
The emblema — the fine, workshop-produced central panel — required the highest level of skill and the most concentrated artistic decision-making. Color matching, the management of gradients and shading, the illusionistic handling of light and shadow: these were skills developed in specialized workshops, some of which appear to have traveled with their products across the Mediterranean. Literary and archaeological evidence indicates that finished emblemata were transported from production centers in Pergamon, Ephesus, and Alexandria to installation sites throughout the Roman world.
The surrounding field work — the geometric borders, the plain tesserae that set off the central image — was executed in the coarser opus tessellatum technique by on-site craftsmen. The design decisions for the field work were less concentrated, the skill requirements less extreme. In modern terms, the emblema workshop held the Technical Translator and Design Author roles; the on-site team handled Fabrication. Both groups were anonymous. The attribution went, as always, to the patron.
This division of labor — between the specialized workshop that made the fine work and the generalist team that executed the surrounding field — is the earliest documented instance of what the four-role framework describes as the separation of Technical Translator from Fabricator. It was structural, not incidental. The medium’s technical demands required it. The attribution system ignored it.
Before tesserae could be laid in any ancient or Byzantine mosaic, a foundation was prepared. Multiple layers of mortar were applied to the wall or floor surface. On the final, still-moist layer, artists drew the design — the composition, the outlines of figures, the spatial relationships. This underdrawing, painted in red ochre or black and known as the sinopia, was then covered by the tesserae. It disappeared from view at the moment the mosaic was completed.
Recent archaeological research on Byzantine mosaics in the southern Levant has documented polychrome sinopiae of remarkable sophistication — detailed preparatory drawings in multiple pigments, applied using fresco technique, that reveal a level of design skill and planning that is completely invisible in the finished mosaic. The sinopia at the Armenian Chapel Birds Mosaic in Jerusalem shows features that differ from the finished mosaic above it, indicating that the tessellation introduced new decisions not present in the original design. Two layers of authorship existed. Only one was visible.
The sinopia is the literal buried layer of authorship in mosaic. The person who drew it — who made the compositional and figural decisions that the tessellation then translated into stone and glass — is separated from the person who executed the tessellation by the most fundamental division of labor the medium contains. In Byzantine practice, the evidence suggests these were sometimes the same person (the master who both drew the sinopia and directed the tessellation) and sometimes different people (the master who drew, journeymen who executed). Either way, the sinopia’s author was never credited separately from the mosaic’s maker. The design layer was invisible by definition: it was buried under the work.
The sinopia is not a metaphor. It is the origin, in the physical structure of mosaic, of the distinction between Design Author and Technical Translator that the four-role framework names. That distinction was encoded into the medium’s production process from antiquity. The art world never found language for it.
The evidence for Byzantine mosaic workshop organization is more limited than the scale of the achievement would suggest. Current scholarship is cautious: recent research replaces the term “workshops” with “teams of craftsmen,” emphasizing collaborative artistic effort without clear hierarchical organization. Large-scale Byzantine mosaic programs required coordinated effort across multiple roles — preparation of the mortar ground, drawing of the sinopia, selection and cutting of tesserae, tessellation itself, and installation — and these roles were not necessarily held by the same individuals.
The commissioning structure is clearer. Major programs were funded by imperial or ecclesiastical patrons: Justinian hired ten thousand artisans to build and decorate Hagia Sophia, and established workshops in Constantinople for icon painting, enamel metalwork, mosaic, and fresco. The scale required organization. The Byzantine art world, in the words of the National Galleries of Scotland, recognized “no division between artist and craftsperson, with makers often extending their skills into more than one discipline” — a formulation that, while accurate in principle, obscures the actual specialization that complex workshop production required.
What we know about the technical process illuminates the labor structure. Before tesserae were laid, the mortar ground was prepared in multiple layers. On the final layer, the sinopia was drawn using strings, compasses, and calipers to establish geometric relationships, and then the figurative outlines were drawn freehand. The tesserae were then embedded while the mortar was still workable — a time-constrained process that required coordinated effort across multiple workers simultaneously. At the largest scales, such as the upper walls and domes of major churches, scaffolding was required, and different sections of the composition might be executed by different hands working on separate sections simultaneously.
Morphological analysis of Byzantine mosaics in the Levant has shown that individual artists can be identified through stylistic consistency in tesserae size, direction of inlay, and figural treatment across multiple sites. This confirms that individual practitioners existed as identifiable artistic entities within the larger production process — they had distinctive ways of working, recognizable across multiple commissions — but their names were almost never recorded. The distinction between the craftsman as a person with a distinctive hand and the craftsman as a named entity in the historical record is the exact gap the attribution system failed to bridge.
The relationship between design and execution in Byzantine mosaic — between the cartoon or sinopia and the tessellation that translated it into permanent form — is the most important structural feature of workshop production for the purposes of attribution.
In Byzantine practice, the cartoon established the composition, iconography, and color relationships of the finished work. It was a skilled creative act: the person who drew the cartoon made the formal decisions that determined what the mosaic would look like. But the tessellation was not mechanical transcription. The mosaicist who translated the cartoon into tesserae made decisions that the cartoon could not specify — the choice of andamento (the directional flow of the tesserae), the exact palette within the color families the cartoon indicated, the size and irregularity of individual tesserae, the angle at which each piece was set to manage light reflection. These tessellation decisions are not minor refinements. In the most sophisticated Byzantine mosaics, the andamento is the primary vehicle of form — it conveys the volume of a figure, the texture of a robe, the spatial recession of a background — in ways that go far beyond what any flat design drawing could specify.
The mosaicists of Ravenna tilted individual gold tesserae at slightly varying angles, so that no two caught the light in exactly the same way at the same moment. A viewer moving through the space experiences the gold surfaces as alive — shimmering, not static. This effect was not in the cartoon. It was in the tessellation. The person who decided how to achieve it — the Technical Translator in the four-role framework — was making an aesthetic judgment of the highest order. That person’s contribution was entirely subsumed into the attribution “Byzantine workshop” or, more often, “commissioned by Justinian.” The cartoon designer and the master mosaicist could be — and frequently were — different people. The attribution system collapsed them into a single unnamed entity.
Justinian I, who reigned from 527 to 565, is the most comprehensively memorialized figure in the history of mosaic. His portrait, in blue robe and golden crown, occupies the north wall of the apse of San Vitale in Ravenna — a church he never visited. His empress Theodora faces him from the south wall. The dedication inscription names the Bishop Maximianus. No craftsman is identified. The work is, in every institutional sense, Justinian’s.
The Ravenna mosaics were made by Byzantine craftsmen brought from Constantinople. This much is established. Who precisely those craftsmen were, how many there were, how they were organized, who directed the iconographic program, who drew the sinopiae, who made the tessellation decisions: none of this is recorded. The most important mosaic commissions of the 6th century — San Vitale, Sant’Apollinare Nuovo, Sant’Apollinare in Classe — are documented only at the level of patronage. The labor that produced them is invisible.
This was not unique to Ravenna. Hagia Sophia, the defining structure of the Byzantine achievement, was built by ten thousand artisans under Justinian’s commission. Not one artisan’s name is preserved. The dome, the mosaics, the shimmering gold surfaces that caused Procopius to write that the dome seemed “not to rest upon solid masonry, but to cover the space with its golden dome suspended from Heaven” — all of it attributable to a building program, to an era, to an emperor. The craftsmen who made that impression possible: gone.
This was the Byzantine attribution model. Credit traveled to the patron. The commission was the work. The craftsmen were the instrument. The model was not questioned because it had no competitors. There was no Vasari in the Byzantine world writing Lives of the Mosaicists, no tradition of individual genius attached to the anonymous practitioners of a collaborative medium. Mosaic was, by the convention of its own greatest period, a medium without artists’ names.
The specific technical contribution that distinguishes the Technical Translator role from simple fabrication deserves its own account, because it is the locus of the attribution problem in mosaic in a way that has no direct parallel in painting or sculpture.
Andamento — from the Italian word for “movement” or “flow” — refers to the directional arrangement of tesserae in a mosaic. The andamento is not merely aesthetic decoration: it is the primary means by which mosaic creates the illusion of form, volume, depth, and surface texture. Opus vermiculatum — the “worm-like” technique — places a single row of tesserae following the exact contour of a focal figure, creating a halo of attention around the subject. Opus musivum extends this principle, making the entire background flow around the central figure rather than simply filling space. Different andamenti create radically different visual effects from the same compositional design: the same cartoon rendered by a mosaicist who understands andamento will produce a work of dynamic visual life; rendered by one who does not, it will produce a flat, mechanical transcription of the design.
The andamento decisions in Byzantine gold-ground mosaics are particularly consequential. The gold tesserae in the great Byzantine cycles were set at slightly varying angles — not randomly, but with deliberate attention to the angle of incident light in the specific space. A skilled mosaicist calculated how each section of the gold field would be seen from the nave at the angles worshippers would occupy, and tilted the tesserae accordingly. This is not transcription. It is site-specific, light-specific, viewer-specific artistic judgment, made by a person with deep technical knowledge and acute perceptual intelligence. In the history of art, no attribution framework has ever attempted to name that judgment. The four-role framework calls the person who made it the Technical Translator, and credits them as a creative contributor distinct from both the designer and the fabricator.
In the 1060s, Abbot Desiderius of the Abbey of Montecassino sent envoys to Constantinople. He was rebuilding the great basilica of his abbey — the most ambitious building campaign in 11th-century Italy — and he needed mosaicists. The Chronicle of Leo of Ostia, a monk at Montecassino during Desiderius’s tenure, records what happened: “Meanwhile he sent envoys to Constantinople to hire artists who were experts in the art of laying mosaics and pavements. The [mosaicists] were to decorate the apse, the arch, and the vestibule of the main basilica; the others, to lay the pavement of the whole church with various colors.”
The Byzantine masters executed their work. The church was consecrated in 1071 by Pope Alexander II in one of the most celebrated events of the 11th century. The mosaic program was, by contemporary accounts, magnificent. None of it survives. The earthquake of 1349 destroyed the basilica; later reconstructions removed what remained. Two small fragments depicting greyhounds now in the Monte Cassino Museum are all that exist of what was considered among the finest mosaic work in the Western medieval world.
What does survive — and what makes the Montecassino commission genuinely significant for the attribution history of the medium — is a phrase in Leo of Ostia’s Chronicle that Britannica identifies as “of particular importance”: Desiderius “took care to see that young local artists were trained by the foreigners.” This was deliberate policy, not incidental outcome. Desiderius understood that the real value of the Byzantine mosaicists was not only the work they made but the knowledge they carried. He secured both: the immediate commission and the long-term transfer of expertise.
The young local artists who trained under the Byzantine masters went on to influence the development of Cosmatesque work throughout southern Italy. The Sicilo-Byzantine school that produced the Norman mosaics at Palermo and Monreale drew on this transmission. The mosaic tradition in Venice, whose craftsmen later decorated the Baptistery of Florence and the facade of St. Mark’s, was built on similar transfers. The knowledge of how to make Byzantine mosaics traveled westward through a chain of training relationships — none of which involved named credit to the individuals who held the knowledge. What Desiderius bought from Constantinople was expertise. What he gave his young monks was the opportunity to acquire it. Neither the Byzantine masters who trained them nor the Cassinese monks who learned from them are named in any record that has survived.
The Montecassino model — expertise imported, knowledge transferred, labor anonymous — is the structural template for how the Byzantine mosaic tradition traveled through the medieval West. It is a model of institutional investment in artistic knowledge and technique that produced some of the most ambitious art of the medieval period, and that left none of its practitioners’ names behind.
The Norman Kingdom of Sicily, under Roger II and his successors in the 12th century, produced the most extensive surviving medieval mosaic cycle outside Constantinople: the 6,500 square meters of tesserae that cover the interior of Monreale Cathedral, completed around 1180 to 1190. Adding the Cappella Palatina at Palermo, the Cathedral at Cefalù, and the Church of the Martorana, the Norman building program represents an investment in mosaic at a scale unmatched in the medieval West.
The production of these mosaics involved a level of multicultural collaboration that is extraordinary even by the standards of Norman Sicily’s cosmopolitan court. The Cappella Palatina alone combines Byzantine-style gold-ground wall mosaics, an Islamic-style painted muqarnas ceiling, Cosmatesque floor pavements, and Norman architectural structure. Craftsmen of multiple religious and cultural traditions worked in the same building simultaneously. Who made the mosaics? The answer depends on which part of the program is being discussed, and scholarship has not settled the question.
For the Cappella Palatina and Cefalù, the quality and purity of Comnenian style strongly suggests imported Byzantine craftsmen — a conclusion supported by the historical pattern of Byzantine expertise being brought to the Norman court. For Monreale, the situation is more complex. The mosaic program shows a more agitated and expressive style than the earlier commissions — what Demus characterized as “late Comnenian style” that had developed in Constantinople and then reached Sicily — yet is executed with enough variation to suggest multiple hands working at different times across a program that was not completed until after the death of its commissioner, William II, in 1189.
The workshops established by the Norman program had lasting consequences: Sicilian craftsmen trained in the Byzantine technique went on to export their expertise to other centers. The mosaics at Salerno (c.1190) and Grottaferrata near Rome (c.1200) are attributed to the Sicilo-Byzantine school. The apse decoration of Ravenna Cathedral and the dome of the Baptistery of Florence were executed by Venetian mosaicists working in the same tradition. The Norman investment in Byzantine expertise created a self-replicating production system that shaped European mosaic for a century after the original Norman commissions were complete.
None of the craftsmen who made Monreale are named. The work is attributed to William II, who commissioned it. The style is attributed to Constantinople. The hands are invisible, as they had always been.
The Montecassino and Norman examples confirm a pattern that holds across the medieval Western mosaic tradition: attribution traveled to the patron. The commissioner was the work. The craftsmen — however skilled, however mobile, however consequential their technical decisions — were not named in the records of their era.
This was not, it should be noted, because the medieval world was incapable of naming craftsmen. Medieval building accounts, guild records, and monastic chronicles name architects, sculptors, and illuminators with regularity. The invisibility of the mosaicist was not a general anonymity but a specific one — the same anonymity that Rozsika Parker documented in the history of embroidery, the same pattern that Designed to Fail identifies across media coded as collaborative, functional, or decorative. The medium’s association with commissioned, site-specific, patron-directed work placed it structurally in the category of anonymous art rather than named individual art — even when the level of artistic decision-making its production required was fully comparable to the most celebrated work of the same era.
When the Renaissance changed the attribution model for painting, it did not change it for mosaic. The genius-director credit structure that elevated Raphael and Michelangelo was built for a medium that lent itself, however inaccurately, to the fiction of the single autographic hand. Mosaic, by its nature, could not sustain that fiction. It remained anonymous. And because attribution drives value, and value drives institutional attention, mosaic’s anonymity contributed directly to its displacement from the canon that the Renaissance was simultaneously constructing.
The Renaissance bottega — the workshop organized around a named master — was the organizational form that created the modern attribution problem in full. Its structure is important to understand precisely, because the contemporary art world’s attribution conventions are its direct descendants.
In the Florentine bottega, the master handled the initial conception of a work and, typically, the most critical passages of execution: the faces, the key figural relationships, the parts of the painting that viewers would scrutinize most closely. Apprentices ground pigments, prepared surfaces, transferred cartoons to walls or panels, and executed the less demanding passages. Journeymen — trained craftsmen who were not yet masters — handled more complex sections under the master’s direction. The division was not always clean: a highly skilled journeyman might execute passages that were, by any reasonable judgment, among the finest work in the finished piece. The attribution went to the master regardless.
What made the bottega distinctive from what came before was the codification of the master’s name as the attribution. The guild system required each master to have a recognizable individual style — the works emanating from a workshop had to look like the work of that particular master, because the master’s name and reputation were the market’s primary quality signal. Apprentices were trained to suppress their own emerging styles and to work in the master’s manner. The resulting attribution convention — the master’s name attached to works substantially executed by others, in a style deliberately indistinguishable from the master’s own — was not fraud in its own terms. It was the system.
Leonardo da Vinci apprenticed in the workshop of Andrea del Verrocchio around age 14. Verrocchio had trained under Donatello. At Verrocchio’s workshop, Leonardo worked alongside Domenico Ghirlandaio and Pietro Perugino — each of whom later opened their own workshops and trained Michelangelo and Raphael respectively. The chain of training and attribution runs: master trains apprentice, apprentice becomes master, master takes credit for workshop production, master’s name enters the canon. The labor within the chain is invisible at every link.
Giorgio Vasari’s Lives of the Artists (1550, second edition 1568) is the founding document of the attribution convention the contemporary art world still uses. Vasari established the template: a canon of named individual geniuses — Leonardo, Michelangelo, Raphael — whose transcendent creative intelligence distinguished them from the merely skilled craftsmen of earlier eras and the lesser practitioners of their own. The Lives created not just a canon but a framework: art was what geniuses made; craft was what craftsmen made; and the division between the two was the division between named individuals and anonymous labor.
Mosaic appears in Vasari’s Lives primarily as a vehicle for the masters he celebrated: he discusses the mosaics of Giotto and Ghirlandaio in terms of the painters’ genius finding expression in an unusual medium. The mosaicists who executed those programs are not mentioned. Vasari’s framework had no place for the Technical Translator — the person whose medium-specific skill was the actual vehicle of the vision Vasari attributed entirely to the named master.
This was not merely a historiographic choice. Vasari’s framework shaped institutional practice for four centuries. Museum acquisition committees, auction house attribution decisions, grant applications, and curatorial programs all proceeded from the assumption that art had one author — the conceptual originator — and that everyone else who contributed to its production was an instrument of that author’s vision. The medium through which the vision was transmitted was largely irrelevant; the name on the work was everything.
The genius-director attribution model that Vasari codified for painting was built specifically for a medium that could, at least fictionally, be executed by a single person. A canvas is portable. Paint is divisible. A painter could, in principle, work alone — even if in practice the great painters of the Renaissance never did. The fiction of the solitary genius was sustainable for painting because the medium did not visibly contradict it.
For mosaic, the fiction was never sustainable. No single person can execute a major mosaic commission alone. The scale, the physical demands, the time constraints imposed by working in wet mortar, the logistics of scaffolding for wall and ceiling work — all of it required teams. But when the art world’s attribution conventions were generalized from painting to all media, mosaic was assigned the same model: one named conceptual author, invisible labor. The difference was that in painting, the fiction could be maintained with some plausibility. In mosaic, it could not.
The result was that mosaic was doubly penalized: it was excluded from the fine arts canon because it was collaborative and therefore (by the logic of the attribution model) not autographic; and it was excluded from adequate credit frameworks because the attribution model that did apply — the single-name model — assigned credit to the patron or, later, to the conceptual director, rather than to the practitioners whose skills were the work. The medium that required the most sophisticated collaborative structure got the least credit for that structure.
The stakes of the attribution model are not academic. They are financial, at a scale that makes the consequences concrete.
The Salvator Mundi — a painting of Christ as “Savior of the World” — was sold at Christie’s New York in November 2017 for $450.3 million, making it the most expensive work ever sold at auction. Christie’s attributed the work fully to Leonardo da Vinci. Many scholars believed — and subsequent scholarship has increasingly confirmed — that the painting was executed with significant workshop participation. The attribution was contested before the sale. Christie’s omission of workshop participation and the full attribution to Leonardo is widely regarded in the field as having been directly responsible for a significant portion of the auction price: the price of a Leonardo is not the price of a Leonardo workshop piece by any reasonable market measure.
The Salvator Mundi case is not an isolated example of attribution error. It is an extreme example of what the attribution model does when it functions as designed: it concentrates value in the named artist and erases the labor that made the work possible. The Rubens workshop produces a fundamentally different market result from Rubens alone — estimates of £15,000–20,000 versus £4–6 million for comparable works. Attribution is not a scholarly nicety. It is a financial determination that transfers wealth between parties on the basis of decisions about who gets credited with what.
In the mosaic context, this dynamic plays out at every level of the market and the commission system. A mosaic “by” a named artist is valued differently from a mosaic “produced in the workshop of” a named artist — even when the same hands, making the same decisions, did the same work. The four-role framework is not merely a scholarly refinement of attribution terminology. It is an argument about who gets paid for what, and whose economic interests the attribution system currently serves. The Invisible Workforce‘s commitment to a mosaic-specific attribution standard is, in economic terms, a redistribution argument.
The 20th century produced new studio models that were, in structural terms, the bottega system with its fictions stripped away. Andy Warhol’s Factory made the mass-production of art explicit and celebrated: silkscreens produced by assistants, works that existed in editions, the deliberate erasure of the autographic mark as aesthetic principle. Jeff Koons employs industrial fabrication teams to produce technically demanding large-scale works to exact specifications. The concept belongs to the named artist; the execution belongs to anonymous or semi-anonymous labor.
The market accommodated this. Warhol’s Factory works are attributed to Warhol. Koons’s fabricated sculptures are attributed to Koons. The fiction of the single autographic author was replaced with an explicit conceptual-director model — the artist as architect, to use Hirst’s own metaphor — and the market accepted the substitution without significantly reducing the attribution credit that flowed to the named artist’s name. The labor that made the work possible remained as invisible as it had been in Vasari’s model.
Sol LeWitt’s wall drawings represent the most intellectually honest version of the contemporary collaborative model. LeWitt argued explicitly that “the idea becomes a machine that makes the art”: the concept is the work; execution is “a perfunctory affair.” By this logic, the fabricators who realized his wall drawings could be anyone who could follow instructions, and indeed they were, ranging from art students to museum staff to trained fabricators commissioned for specific installations.
What distinguished LeWitt’s practice from Warhol’s and Hirst’s was a deliberate crediting choice: LeWitt insisted that the names of his fabricators appear on museum and gallery labels and in all written materials accompanying his projects. He was explicit about this. The fabricators were not erased; they were named as part of the work’s identity. The Landmarks program at the University of Texas at Austin describes LeWitt as “a one-man WPA program of the postwar period” in terms of the employment of trained labor — and notes that his insistence on crediting that labor was a deliberate act, not a courtesy.
LeWitt’s model has a significant limitation, however: it credits the Fabricator role while maintaining an absolute hierarchy in which the Conceptual Originator’s idea is the art, and the fabricator’s execution is, in LeWitt’s own word, “perfunctory.” The fabricator is credited — which is better than invisibility — but the credit is for execution of instructions, not for creative contribution. In the four-role framework, the LeWitt model names Fabricators but does not create a category for Technical Translators: those who make judgment calls within the parameters of the instructions, who bring their own knowledge of the medium to bear in ways the original instructions could not specify.
LeWitt acknowledged this gap in practice even if not in theory. The Dia Art Foundation’s documentation of his work notes that fabricators’ physical characteristics — their height, arm length — sometimes became embedded in the finished piece, and that the role of assisting drafters could be “as influential to the piece’s fabrication as the architectural quirks of its site.” Technical knowledge and physical presence shaped the work in ways that exceeded instruction-following. LeWitt’s attribution model was better than Hirst’s. It was not yet adequate to the full range of contributions collaborative production involves.
The Rachel Howard case is the most documented contemporary instance of the attribution problem The Workshop Tradition addresses, and it warrants specific attention because Hirst himself has named it.
Rachel Howard was Damien Hirst’s first studio assistant, working for him from 1992 to 1995. During that period, she painted more than 300 of Hirst’s spot paintings. Hirst has said publicly, in his autobiography and in interviews: “The best person who ever painted spots for me was Rachel. She’s brilliant — absolutely fucking brilliant. The best spot painting you can have by me is one painted by Rachel.”
Rachel Howard painted over 300 works. Hirst by his own admission painted five. The best of the works that bear his name were made by her. She is named in interviews. She is not named in catalog entries. The catalog entries say: Damien Hirst.
Howard went on to have a significant independent career. In 2017, Hirst gave her a solo exhibition at his Newport Street Gallery — the Art Newspaper described her as “Hirst’s most coveted spot painter” — where she showed work that had nothing to do with spots. She is “considered to be one of the most influential British Contemporary artists of the 21st century,” according to auction house records. She is also the best-documented case in the contemporary art world of a Technical Translator whose credit was systemically erased while the Conceptual Originator publicly acknowledged her contribution and then returned the catalog entry to his own name.
Hirst’s use of assistants is controversial in a way that Warhol’s or Koons’s was not. Part of what makes it uncomfortable is the frankness: Hirst did not merely use assistants and stay quiet about it. He talked openly about which paintings were better when Rachel painted them. The attribution system continued regardless. The catalog entry absorbed the admission. The name on the work remained his.
The genius-director model — one named Conceptual Originator, invisible labor — was built for painting and then applied by default to all media. But the article cited in the session discussion that proposed The Workshop Tradition made a crucial observation: “Byzantine workshops did not use a genius-director model. The craftsmen were not executing a master’s vision; they were the tradition.” This is the exact distinction that makes the mosaic attribution argument structurally stronger than the painting argument.
In the painting studio, the fiction of the genius-director could at least point to a real creative act: Hirst conceived the spot paintings, even if he executed very few of them. The concept — the grid of pharmaceutical-named spots — was his, however simple. In Byzantine mosaic production, the situation was different in kind, not degree. The craftsmen were not realizing a single master’s vision. They were the carriers of a technical tradition that could not be separated from its practitioners. The knowledge of how to cut smalti, how to set tesserae at the angles that produce Byzantine shimmer, how to read a sinopia and translate it into the medium’s specific visual language — this knowledge was embodied knowledge. It lived in the craftsmen’s hands. The patron provided the program. The craftsmen provided the tradition. The tradition was the art.
This is not a romantic argument about anonymous craft. It is a precise argument about where the creative decisions in mosaic production were located. They were located in the Technical Translator role: in the tessellation decisions, the andamento choices, the palette management, the light calculations that were specific to the medium and could not be delegated to a cartoon or a patron’s brief. The medium required those decisions. The attribution system ignored them. The Workshop Tradition‘s framework names them.
The contemporary art world has one word for everyone who is not the named artist: assistant. Rachel Howard was Hirst’s “assistant.” The Byzantine craftsmen who made the Theodora mosaic were Justinian’s “craftsmen.” The journeymen of Raphael’s workshop who painted the peripheral angels of the Transfiguration were Raphael’s “assistants.” The nine youth artists at Manly Career Academy whose drawings and collages became the visual content of Community Harmony Through Song and Play were, in most systems of public art documentation, “participants” or “youth artists” — a category that fails to distinguish between those who designed the imagery (a genuinely creative act) and those who fabricated it (a skilled but different act).
The word “assistant” collapses contributions of fundamentally different kinds — each with distinct creative content, distinct skill requirements, and distinct credit obligations — into a single invisible category. The category’s invisibility is structural: it exists to make the named artist’s credit appear total. This is the field’s dominant attribution problem, and it is most acute in media like mosaic, where the tessellation decisions are both unmistakably skilled and unmistakably distinct from the conceptual program.
Part of the problem is terminological; part of it is structural. Research across community mural and mosaic project documentation reveals that the field already uses several role titles consistently — Lead Artist, Assistant Artist, Mentor Artist, Teaching Artist, Community Partner — but applies them without a coherent attribution standard beneath them, and without distinguishing between types of community contribution that carry different credit obligations. The MME’s framework builds on field language where it exists and introduces new terms only where the field has a genuine gap.
The framework operates in two tiers. The first contains roles present on every project; the second contains roles that apply when they are present. Both tiers use field terminology where it exists. The analytical categories used throughout Parts I through V of this report — Conceptual Originator, Design Author, Technical Translator, Fabricator — describe the type of contribution being made. The attribution taxonomy below describes the practical roles that hold those contributions in community mosaic practice.
| Role | Description | Historical Parallel | MME Credit Obligation |
|---|---|---|---|
| Lead Artist | The primary commissioned artist. Holds the Technical Translator role (tessellation, andamento, palette decisions) and typically also the Conceptual Originator role. The term is standard across the field. | Byzantine: the master who directed the workshop and drew the sinopia. Renaissance: the named master of the bottega. Community art: universally the term for the contracted professional artist. | Named first in all records, on the plaque, and in all press materials. Role explicitly identified. |
| Assistant Artist | A professional artist supporting the Lead Artist. Holds Technical Translator and/or Fabrication roles depending on the specific contribution. Field term used consistently in community art documentation. | Byzantine: journeymen executing under the master’s direction. Manly (2006): Caswell James, credited as Assistant Artist in CPAG documentation. | Named in all attribution records and on the plaque. Specific contribution described in the living record. |
| Student | An MME Studios student working on a live commission as part of their training. Holds Technical Translator or Fabrication roles depending on contribution. The MME does not use “Apprentice” or “Intern.” | Desiderius at Montecassino explicitly trained local monks under Byzantine masters. Norman Sicily trained local artisans who became the next generation of practitioners. | Named in all attribution records. Role identified as Student, connecting credit to their MME training. Never unpaid. |
| Design Contributor | A community member whose drawings, imagery, or design decisions became the visual content of the work. A genuinely creative contribution — co-authorship of the image, not assistance. The field currently collapses this into “participant”; the MME distinguishes it. | Manly (2006): the nine youth artists whose drawings and collages became the final panels. In Byzantine practice: the unnamed designers whose sinopiae were the design layer buried beneath the tesserae. | Named individually in all attribution records and on the plaque when the number permits. Where numbers exceed plaque capacity, counted and stated: “With [N] Design Contributors.” Full individual names in the living record. |
| Fabrication Participant | A community member who engaged in the physical making of the work — cutting material, placing tesserae, assisting with installation — without holding a design authorship role. Skilled, valued, and distinct from Design Contributor. | In Byzantine and Norman workshops: the teams responsible for large-format background work and installation logistics, distinct from the design and tessellation specialists. | Named in the living record. Counted on the plaque: “With [N] Fabrication Participants.” Never erased, never collapsed into Design Contributor. |
| Role | Present when | Historical parallel | MME credit obligation |
|---|---|---|---|
| Teaching Artist / Mentor Artist | Any project with an instructional dimension — school residencies, youth programs, community workshops. When the Lead Artist is also teaching, this role may be held by the same person or a separate practitioner. | Manly (2006): Nina Smoot-Cain, credited as Mentor. Her suggestion of cut-paper collage shaped the entire design methodology — an instructional contribution with direct creative consequences. | Named in all records and on the plaque. Contribution described specifically: what they taught, what they introduced, how it shaped the work. |
| Cultural Advisor | Projects involving specific cultural expertise that shaped the conceptual frame — ensuring imagery is appropriate, accurate, and genuinely representative of the community depicted. Recognized in indigenous art and high-stakes commission contexts. | Norman Sicily: the Byzantine iconographic program required practitioners who held specific theological and visual knowledge. The Lyons Family monument (NYC DCLA) used a named curator to facilitate community engagement sessions that shaped the conceptual program. | Named in all records and on the plaque. Nature of contribution explicitly described. Not collapsed into “community consultation.” |
| Community Historian | Projects involving documented community research — local history, oral history, archival research — that shaped subject matter or imagery. Distinct from Cultural Advisor, whose expertise is cultural rather than historical. | Not named in historical practice. The concept is present in contemporary place-based public art that involves deep historical research, but the role is almost never credited separately. | Named in all records. Named on the plaque when the historical contribution was central to the work’s subject matter. The first instance of this credit in the field may be an MME project. |
| Project Manager / Coordinator | Large-scale commissions with distinct administrative functions — logistics, permitting, community outreach coordination, budget management. Recognized as a distinct professional role by CPAG, San Francisco Arts Commission, and major public art programs. | The role has existed implicitly in every major commissioning context — Desiderius’s “envoys to Constantinople,” the unnamed administrators of Norman building programs. Never credited. | Named in all records. Not typically on the plaque unless the project management function was central to the work’s community dimension. Always credited in press materials and institutional documentation. |
| Community Partner | The commissioning or host organization for community projects — a school, a community organization, a cultural institution. An organizational credit, not an individual one. | In Byzantine practice: the commissioning patron (bishop, abbot, emperor). In Norman Sicily: the royal court. In every era, the institutional commissioner received the attribution the craftsmen did not. | Credited as Community Partner in all records and on the plaque. Named specifically, not as “the community.” The distinction matters: the Community Partner is an institution; the Design Contributors are individuals. |
The analytical framework — Conceptual Originator, Design Author, Technical Translator, Fabricator — is not a contemporary invention imposed on historical practice. It describes divisions of labor that existed in workshop production throughout the history of the medium, even when the attribution system of each era collapsed those contributions into a single name or into anonymity.
In ancient workshop production: the emblema specialist held both Design Author and Technical Translator roles. The patron held Conceptual Originator. The on-site team handled Fabrication. In the modern taxonomy: the emblema specialist is the Lead Artist; the on-site team are Fabrication Participants. Occasionally named (Sosus, Gnosis, Herakleitos) — but only as exceptions, never as standard practice.
In Byzantine workshop production: the patron (emperor, bishop, abbot) was the Conceptual Originator. The master who drew the sinopia was the Design Author; in modern terms, the Lead Artist. The lead mosaicists who made tessellation decisions were Technical Translators — in modern terms, also the Lead Artist and any Assistant Artists. The teams who executed at scale were Fabricators — in modern terms, Fabrication Participants. All four analytical roles existed. None was attributed separately.
In Norman Sicily: the Norman king was the Conceptual Originator. The iconographic program was directed by Byzantine masters — the Lead Artists of those commissions. The Technical Translator role was distributed across practitioners whose individual contributions are beginning to be separable through stylistic analysis; they would today be credited as Lead Artists or Assistant Artists depending on the nature of their contribution. The large execution teams were Fabrication Participants.
In Renaissance bottegas: the master held both Conceptual Originator and Design Author roles, and typically also Technical Translator for the most demanding passages. Journeymen held Technical Translator roles for secondary passages. Apprentices held Fabrication roles. The master’s name covered all of them in the attribution record.
The Community Harmony Through Song and Play project (CPAG, Manly Career Academy High School, North Lawndale, Chicago, 2006) is the primary documented precedent for the MME’s community commissioning model. It is cited throughout Foundations of Instructional Practice as the foundational case study for the MME’s instructional framework. It is cited here as the primary case study for the attribution framework in community mosaic practice — because it shows multiple roles operating simultaneously, with different contributors holding different roles, and the project documentation attempting, unusually for its era, to reflect that complexity.
The project’s formal credits as documented: Lead Artist: Rachael Que Vargas (credited under prior name). Assistant Artist: Caswell James. Mentor: Nina Smoot-Cain. Youth Artists: Sheena Barlow, DeJuan Birge, Katheris Ellis, Ashley Harvey, Kenyetta Howard, Ernest Johnson, Deanna McElroy, Tiana Solid, Andrew Section. Additional contributors: Elke Claus, Chris Silva, Julio Berlin, named with their specific contributions.
Mapping these credits to the expanded taxonomy:
Lead Artist: Rachael Que Vargas — held the Technical Translator role (tessellation decisions, andamento, palette management across three panels) and the co-Conceptual Originator role (bringing the specific educational and design methodology).
Assistant Artist: Caswell James — held Technical Translator and Fabrication roles.
Teaching Artist / Mentor Artist: Nina Smoot-Cain. Her suggestion of cut-paper collage as the design methodology is not courtesy credit; it is accurate attribution of an instructional contribution with direct creative consequences. The collage process was the formal solution that unlocked the students’ imagery. That contribution is now named by a role title the field already uses. In the MME system it goes on the plaque.
Community Partner: Chicago Public Art Group — the commissioning organization.
Design Contributors: the nine named youth artists, whose drawings and collages became the visual content of the final panels. Their imagery was not source material for someone else’s design — it was the design. The existing documentation names them individually and makes their design authorship legible, which is why this project stands as the standard. In the MME taxonomy, they are credited as Design Contributors, not “youth artists” or “participants.”
Fabrication Participants: some of the same youth artists, in their capacity as physical makers of the work, as distinct from their design contribution. In practice these roles often overlapped in the same individuals across different stages of the project, which is why the distinction must be documented at the process level rather than inferred from a single credit line.
The Manly documentation is unusually careful by the standards of its era and sector. It is the floor, not the ceiling, and Part VII makes the case for why careful is not enough.
The field’s current understanding of attribution — to the extent that community art practice has any systematic understanding of it — treats it as a completion obligation: the credits are established when the work is finished, and the record is made at that moment. This understanding is inadequate, and the failure modes are structural, not incidental.
Point-of-completion attribution fails because records are maintained by institutions that have their own lifecycles, and those lifecycles are not synchronized with the artist’s. Websites are rebuilt. Staff turn over. Archives are reorganized. Filing systems that were once comprehensive become disorganized over decades without the resources to maintain them. The CPAG archive — which documents hundreds of significant public art projects across five decades — was, by the organization’s own account for its 50th anniversary, “a hallway containing file cabinets of papers, 35mm slides, photographs, AV materials” whose contents were “a mystery” to the organization’s current staff. The archives “outgrew basic organization” because CPAG “prioritized spending on creating new artworks and assessing the condition of/restoring older murals.” This is not negligence. It is the structural reality of under-resourced nonprofit arts organizations managing more institutional history than they have resources to maintain — and making the understandable, even admirable, choice to put resources toward the work itself.
The consequence is predictable. Works that were carefully documented at completion are partially or fully missing from institutional records twenty years later. Artists whose contributions were credited at the time of the project are invisible in any search of the commissioning organization’s current web presence. The record was made. The record failed forward. The credit was real at the moment of completion. It does not exist in any practically accessible form today.
The name-change failure mode in attribution records is not a rare edge case. It is a structural vulnerability of any attribution system that uses a static credit line rather than a persistent identifier for the artist. Artists change their names for a range of reasons: marriage, divorce, transition, legal name changes, professional rebranding. The art world has no standard mechanism for updating attribution records when this happens.
For trans artists — and specifically for trans artists who transition after completing significant work — the failure mode is particularly damaging. Work done before transition is credited to a name that no longer belongs to the artist, and no mechanism exists to connect that prior name to the artist’s current institutional identity in a way that preserves both the historical record and the artist’s current identity. The artist is effectively split in two: a prior-name self whose work exists in the institutional record, and a current self whose institutional record does not contain that work. For purposes of grant applications, exhibition proposals, artist CVs, and institutional credentialing, the work under the prior name does not count — not because it was not done, but because the attribution system has no way to connect it to the current person.
The MME attribution standard addresses this directly. Every attribution record maintained by the MME will include a name-change field: “Also known as” or “Formerly credited as,” which preserves the historical record without stranding the artist in a prior identity. The connection between prior-name credit and current identity will be explicit, versioned, and publicly accessible. The attribution record will travel with the artist, not only with the work.
The absence of standard mechanisms for attribution maintenance and identity continuity is not accidental. It reflects the underlying assumption of the attribution model: that attribution is a completed act, established at the time of production, attached to the work and to no one else. The work has an author. The author is named. The name is fixed. When the work changes hands, the attribution travels with it. When the artist’s name changes, the attribution does not.
This model works reasonably well for a narrow band of art-world practice: the single-author painting or sculpture that is produced, attributed, sold, and enters a permanent collection where the record is maintained by an institution with conservation resources. It does not work for community art, where multiple contributors hold different roles; for long-term studio practice, where the artist’s name and identity may change; or for the specific situation of mosaic, where the medium’s highest achievements have always required collaborative production that the single-author model cannot accommodate.
The MME’s attribution standard is an argument that the model must expand. The expansion requires not only new language — the expanded role taxonomy — but new infrastructure: records that are permanent, versioned, publicly accessible, and updatable. Records that travel with both the work and the artist. Records that include a maintenance obligation with a budget line. Records that have a governance mechanism to ensure they are updated rather than abandoned. And records that are connected to the physical work not by assumption or goodwill but by a structured policy: the plaque, the living record, and the link between them.
The commitments in this section are operational. They specify what the MME will do, not what it aspires to do. They are grounded in the historical and analytical foundations of Parts I through VII, and they will be implemented through the institutional instruments developed in The Invisible Workforce (the attribution standard) and The Outsider Art System (the acquisition ethics framework). They are recorded here as the analytical foundation for both.
Every work produced with MME involvement — commissioned, collected, or produced through the MME Studios program — will carry a full attribution record using the role taxonomy defined in Part VI. Roles held by the same person will be attributed to that person in multiple categories. Roles held by different people will be separately attributed. The record is not a formality. It is the institutional argument, made specific.
The credit obligation differs by role and tier. Lead Artists and Assistant Artists are named in all records, all press materials, and on all plaques. Students are named in all records and press materials; whether they appear on the plaque depends on the scale of the project and their specific contribution, determined project by project. Design Contributors are named individually in the living record and on the plaque where the number permits; where it does not, their count is stated explicitly. Fabrication Participants are named in the living record; their count appears on the plaque. Teaching Artists, Mentor Artists, Cultural Advisors, and Community Historians are named on the plaque and in all records when present. Project Managers and Community Partners are named in all records and press materials.
One distinction matters enough to state explicitly: the MME does not use unpaid intern labor, and the term “Intern” does not appear in the attribution taxonomy for the same reason it does not appear in the MME’s employment practices. Students working on live MME commissions as part of their training at MME Studios are credited as Students. The credit is accurate, connects their work to their education, and is never unpaid.
The MME’s attribution commitment spans three distinct surfaces, each with different affordances and different limitations. Understanding what each surface can and cannot do is the precondition for making the commitment real rather than aspirational.
The living digital record is the authoritative attribution document. It contains the full role taxonomy — all roles, all names, all contributions described — versioned, timestamped, and updatable. It carries the name-change field. It is the document that travels with the work when the work changes institutional hands. It is the record the MME is committed to maintaining through a policy of permanent URL structure that does not change with website rebuilds, organizational restructuring, or domain changes.
The physical plaque is permanent, highly visible, and space-constrained. Once fabricated, it cannot be corrected without replacement. It will always name: the work’s title and year; the Lead Artist(s); the Teaching Artist or Mentor Artist if present; the Cultural Advisor if present; and Design Contributors individually when the number permits legibility. When the number of participants exceeds what the plaque can carry legibly, it states the total count explicitly — “With [N] Design Contributors and [N] Fabrication Participants.” The plaque is honest about what it cannot contain. It does not suppress the existence of those it cannot name.
The QR code bridges the two surfaces. Printed on the plaque, it links directly to the living record, where the full attribution appears. This is a present-moment solution, not a permanent one. QR code technology may become obsolete. Digital links break. No digital format is archival in the sense that engraved stone or bronze is archival — a fact that anyone who has tried to open digital files created twenty years ago already knows. The MME makes no claim that the QR code is a permanent solution. What it does is make the full record accessible at the point of encounter with the work, for as long as the technology and the institutional commitment behind it hold.
The permanence of the living record depends not on the QR code but on the institutional governance beneath it: a policy of permanent URL structure, maintained through transitions of all kinds. This policy will be codified in The Invisible Workforce and included in the MME’s founding governance documents. The record’s URL will not change when the website does. Redirects will be maintained. The budget line for attribution maintenance will cover that maintenance explicitly. The commitment is to governance, not to technology — which is the only commitment an institution can honestly make.
The gap the field has not addressed — and that the MME’s taxonomy fills — is the distinction between community members who authored the design and those who participated in the physical making. The field currently uses “youth artists,” “community participants,” or “volunteers” for both, without distinguishing between a participant whose drawing became the imagery of the work and a participant who helped cut tile on a Saturday. The credit obligation to these two people is different. The taxonomy names that difference.
Community members whose drawings, imagery, or design decisions became the visual content of a commissioned MME work are credited as Design Contributors in all attribution records, named individually in the living record and on the plaque when numbers permit. This is not a gesture of inclusion. It is an accurate account of where the design came from. The credit belongs to them because the image belongs to them.
Community members who engaged in the physical making of the work without holding a design role are credited as Fabrication Participants. Named in the living record. Counted on the plaque. Never erased, never collapsed with Design Contributors, and never credited as something they were not. The distinction is one of role, not of value — fabrication is skilled, important, and genuinely collaborative. It is credited honestly.
Cultural Advisors whose expertise shaped the conceptual or iconographic program — ensuring imagery is appropriate, accurate, and genuinely representative — are named in all records and on the plaque. Their contribution is intellectual and creative. It is currently invisible in attribution conventions. The MME names it because it is real.
Every attribution record maintained by the MME will be versioned and timestamped. When a record is updated — for any reason — the previous version is preserved and the update recorded with a timestamp and an explanation. No version of an attribution record is destroyed. The current version is always publicly accessible; prior versions are available on request.
Every attribution record will include a name-change field: “Also known as” or “Formerly credited as,” updated whenever a name change is communicated to the MME. The MME will not require documentation of why a name changed. It will require notice that a change has occurred. The historical credit — the prior name, the date of the prior credit — will be preserved and connected to the current identity, not severed from it. An artist who worked with the MME under a prior name retains full credit for that work in a form that connects to who she is now.
Attribution maintenance will be a budget line in all commissioning contracts and in the MME’s annual operating budget. The CPAG archive failed forward not because the intention was absent but because the resources were not. The MME’s standard addresses this directly: the framework is specific enough to implement, and the budget line is explicit enough to fund.
The record will travel with the work when it changes institutional hands, through attribution documentation included in every transfer agreement. The MME will require that institutions receiving works it has commissioned or deaccessions maintain the full role taxonomy and the name-change field as a condition of transfer. The standard does not stop at the MME’s door.
The following table maps the analytical framework — Conceptual Originator, Design Author, Technical Translator, Fabricator — against the principal production models documented in this report. The final row describes the MME standard using the expanded role taxonomy from Part VI.
| Production Model | Design Author | Technical Translator | Credit Assigned To |
|---|---|---|---|
| Medieval / Byzantine | Patron / master (unnamed) | Workshop craftsmen (unnamed) | Patron or era |
| Renaissance bottega | Master (named) | Journeymen (unnamed) | Master alone |
| Warhol Factory | Warhol (named) | Factory assistants (unnamed) | Warhol alone |
| LeWitt wall drawings | LeWitt (named) | Fabricators (named on labels) | LeWitt; fabricators credited on label |
| Hirst spot paintings | Hirst (named) | Rachel Howard and others (named in interviews; not in catalog) | Hirst alone in catalog |
| MME standard | Lead Artist + Design Contributors named individually | Lead Artist / Assistant Artist / Student named by role | All roles named; plaque + living record + QR code |
The argument of The Workshop Tradition runs from antiquity to the present, and it arrives at the same place at every stop. Mosaic was never made the way the art world’s attribution model says art is made — by a single named genius whose creative act is the work. It was made by teams, by workshops, by craftsmen who held different roles at different points in a production process that was, by the medium’s physical nature, collaborative from the first layer of mortar to the final tessera.
The craftsmen who tilted the gold tesserae at San Vitale to catch the light of oil lamps are not named. The mosaicists Desiderius brought from Constantinople to build the finest basilica of 11th-century Italy are not named, and their work no longer exists. The Byzantine masters who taught their knowledge and technique to Norman kings’ workshop teams and whose technical tradition traveled from Montecassino to Sicily to Venice are not named. Fifteen centuries of the medium’s greatest achievements are attributed to the people who paid for them, not the people who made them.
The contemporary model is not better. Rachel Howard painted over 300 of Damien Hirst’s best spot paintings by his own admission. She is named in interviews. She is not named in catalog entries. The attribution went, as it always has, to the person whose name is the market signal.
The Workshop Tradition does not recover those names. That loss is permanent. What it does is establish that the system that erased them is a choice — a historical construction, not a natural fact — and that the MME is choosing differently. The expanded attribution taxonomy names the roles that collaborative production actually involves: Lead Artist, Assistant Artist, Student, Design Contributor, Fabrication Participant, Teaching Artist, Cultural Advisor. The plaque names those it can; the living record names all of them. The QR code bridges the two for as long as the technology holds. The governance policy holds for longer. The maintenance budget holds it honest.
The Manly project’s documentation — nine youth artists named, the mentor credited, the design process made legible — is the floor, not the ceiling. The MME will not leave its projects in that position.
Mosaic’s greatest achievements were structurally incompatible with the attribution framework the art world uses. That is the argument. The MME builds its attribution practice on the premise that the artists — historical and contemporary, named and unnamed — who made the work deserve to be seen. The attribution system has never made that possible. The MME will.
Designed to Fail: Gender Inequality and the Art/Craft Divide. The foundational argument that The Workshop Tradition extends into the specific history of workshop attribution and collaborative labor.
The Mosaic Record. The comprehensive account of mosaic’s history to which The Workshop Tradition contributes the workshop production dimension.
Made by Hand. Mierle Laderman Ukeles’s Maintenance Art as feminist answer to the devaluation of skilled, collective, maintenance labor. The Workshop Tradition‘s workshop analysis is the historical version of the same argument.
The Invisible Workforce. The institutional commitment that The Workshop Tradition provides the analytical foundation for. The expanded role taxonomy, the plaque/living record framework, and the maintenance obligation specified in The Workshop Tradition are the content of the The Invisible Workforce standard.
Foundations of Instructional Practice. The Manly Career Academy project is the primary case study in both The Workshop Tradition and Foundations of Instructional Practice. The Workshop Tradition uses it as attribution evidence; Foundations of Instructional Practice uses it as instructional evidence. The Participation and Attribution stage of the MME’s instructional framework (Foundations of Instructional Practice, Stage 6) teaches the role taxonomy developed in The Workshop Tradition.
Pliny the Elder. Natural History, Book XXXVI. On Sosus of Pergamon and ancient mosaic.
Bryn Mawr Classical Review. Review of Roger Ling, Ancient Mosaics. Analysis of Sosus, emblemata, and ancient workshop conventions.
Invaluable.com. What Roman Mosaics Reveal About Ancient Art. Overview of Roman workshop production and the emblema system.
Wikipedia. Sosus of Pergamon. Summary of ancient sources.
National Galleries of Scotland. Byzantine Art. On workshop organization and the absence of the art/craft divide in Byzantine practice.
TheArtStory.org. Byzantine Art and Architecture Overview. Justinian’s workshop organization and the scale of Hagia Sophia.
Wikipedia. Byzantine Mosaics. Technical process including sinopia preparation and tessellation methods.
Wikipedia. Byzantine Art. Iconoclasm, patronage structure, and the major surviving programs.
Rogue Art Historian (Substack). Tesserae of Power, Pattern, and Prayer. On Byzantine craftsmen in Islamic monuments and Norman Sicily.
Crostini, Barbara. “Gypsum and Mortar: Constructing Byzantine Builders and Craftsmen between Late Antique Practice and Medieval Exegesis.” Different Visions 10 (2023). On the organized collectives of Byzantine craft labor.
Research on Armenian Chapel Birds Mosaic. “Mosaic Technology in the Armenian Chapel Birds Mosaic, Jerusalem: Characterizing the Polychrome Hidden Sinopia.” MDPI Heritage 7/10 (2024). On sinopiae as documented design layers distinct from tessellation.
Morphological analysis of Byzantine mosaic artists. “Mosaic Artists in the Byzantine East.” Academia.edu (2022). On identifying individual artists through tesserae size, direction of inlay, and stylistic consistency.
Britannica. Mosaic Art — Medieval Mosaics in Western Europe. On Desiderius and Montecassino; the training imperative; Norman Sicily; Venetian mosaicists.
Smarthistory Guide to Byzantine Art. “The Visual Culture of Norman Sicily.” On the debate over imported versus local craftsmen at the Cappella Palatina.
Wikipedia. Monreale Cathedral Mosaics. On the extent of the program and the Byzantine style.
Sabidius Classics Blog. The Byzantine Mosaics of Norman Sicily. On Demus’s stylistic analysis and the Comnenian style at Monreale.
Smithsonian Associates. Medieval Mosaics in Norman Sicily: An Artistic Convergence of Empires.
MuseoFacile. The Pavement of the Abbey Church of Desiderius (1066–1071). On the Constantinople craftsmen and the Cosmatesque legacy.
Leo of Ostia. Chronica Monasterii Cassinensis. Primary chronicle account of Desiderius’s commission; the training imperative (“he took care to see that young local artists were trained by the foreigners”).
MyArtBroker. Damien Hirst’s Assistants vs. Renaissance Workshops (January 2026). On the bottega system, the Salvator Mundi, and contemporary studio models.
World History Encyclopedia. Life in a Renaissance Artist’s Workshop. On the bottega structure and division of labor.
Serenade Magazine. Were the Old Masters Really Working Alone? On Raphael’s workshop and the collaborative reality of Renaissance production.
Landmarks, University of Texas at Austin. Sol LeWitt, Wall Drawing #520. On LeWitt’s crediting practice.
Wikipedia. Rachel Howard. Primary documentation of her work as Hirst’s studio assistant and spot painter.
The Art Newspaper. Damien Hirst Gives Exhibition to Former Spot Painter Rachel Howard (2017). Context on Howard’s career.
Wikipedia. Damien Hirst. Hirst’s quoted admission on Howard’s spot paintings.
Chicago Public Art Group. cpag.net. Mission, project documentation, and program overview.
Wikipedia. Chicago Public Art Group. History and founding mission.
Archival Outlook, May/June 2024. Recovering a History of Public Art in Chicago. On the CPAG archive condition and the 50th anniversary rebuild.
Smithsonian Institution, Archives of American Art. Chicago Public Art Group Archive. On the 35 linear feet of paper records organized by project.
Foundations of Instructional Practice at MME Studios (v2.0, March 2026). Primary documentation of the Manly Career Academy project, including attribution records, design process documentation, and the institutional commitments that follow from it.
Rachaelquevargas.com. Community Harmony Through Song and Play. Lead artist’s documentation of the project.
Pompano Beach Arts. Mosaic Team Program documentation. On Lead Artist / Apprentice / Lead Apprentice role structure in a funded mosaic program.
University of Leeds Sustainability Service. Apply for the Community Mural Project 2024/25. On “Mentor Community Artist” and “student artist” as distinct credited roles.
Design and Review Criteria for Public Art (Project for Public Spaces). On the artist as “collaborator, interpreter, visionary, teacher, mentor, and liaison.”
CPAG (Chicago Artists Coalition job listing). Public Art Project Manager. On Project Manager / Coordinator as a distinct professional role in community public art.
The Mosaic Store. Design Fundamentals — Andamento. On opus types and the role of directional flow.
Merrin’s Art. Mosaic Techniques: Direct, Indirect, Opus Styles and Modern Innovations. On andamento as the primary vehicle of form.
Gary Drostle. Are You Ready to Create a Larger Mosaic. On andamento, the cartoon, and professional large-scale production.
This report was developed through an iterative, fact-checked, and edited collaborative research process between Rachael Que Vargas and Anthropic’s Claude (in two roles — long-form research and document operations). The questions, institutional framework, and editorial judgment are the author’s; the research synthesis and structural development are collaborative.
© 2026 Rachael Que Vargas / Museum of Mosaic Environments. Licensed under Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0). You may share and adapt this work for non-commercial purposes with attribution. Full license: https://creativecommons.org/licenses/by-nc/4.0/