Jerry Saltz’s recent autopsy of Pace Gallery, “How Pace Gallery Broke Itself,” makes for uncomfortable reading in the current climate. His portrait is of a mega-gallery caught in the machinery of its own scale: global ambition, brand expansion, endless growth, then contraction. Around the same time, The Art Journal published a piece about a dispute between Modigliani scholar Marc Restellini and the Swiss firm Art Recognition, after Art Recognition announced that a painting it had analyzed with AI had also been included in Restellini’s newly published Modigliani catalogue raisonné. These two stories may appear unrelated. One concerns the business model of a mega-gallery. The other concerns the misinterpretation of the wording of an authentication announcement. But together they raise a larger question that the art market prefers to whisper about rather than discuss openly: who benefits when a catalogue raisonné redraws the boundaries of an artist’s oeuvre?
The question is newly urgent because Restellini’s Modigliani catalogue raisonné, a six-volume box set distributed by Yale University Press and priced at around $2,500, is not just another scholarly reference. It reportedly adds more than 100 works to the accepted Modigliani corpus while excluding only a smaller number of works that had appeared in previous catalogues (15). For an artist as valuable, and as notoriously forged, as Modigliani, this is a market event. It is also a scholarly event, of course, and one can be both at once. A single Modigliani can be worth many millions. Inclusion in a major catalogue raisonné can make a work saleable, insurable, lendable, exhibitable and bankable in a way that no press release or dealer’s opinion can.
Before publication, rumors circulated in the art world, including in conversations I overheard at TEFAF Maastricht, suggesting that Pace Gallery was somehow involved behind the scenes. The reasoning, as it was whispered, was that Pace was said to own or represent some of the newly included works, or to stand to benefit from their new status. Pace has stated officially that it was not behind the publication of the catalogue. In the absence of evidence to the contrary, that statement must be accepted. There is also no basis for implying that the works Restellini has included are inauthentic simply because their inclusion may increase their value. That would be an unfair and unsupported leap. Restellini is widely presented, including by the catalogue’s own promotional materials, as one of the leading living authorities on Modigliani. So all that must be set aside.
The more useful question is structural. What if a gallery, collector, estate, foundation, dealer or other interested market participant sponsors or promotes a catalogue raisonné that includes works they own, represent or intend to sell? That is not proof of wrongdoing. Many catalogues raisonnés require private funding because they are ruinously expensive to produce and unlikely to turn a profit through sales alone. Academic presses rarely underwrite such projects entirely, especially when the audience is small and the production costs are high. Archives must be assembled, images licensed, works examined, scientific tests performed, staff paid, and years, sometimes decades, of labor absorbed. The money has to come from somewhere.
The issue is transparency. If the catalogue’s conclusions may materially benefit a party involved in funding, promoting or shaping the project, readers deserve to know the relationship. The problem is not that the market cares about catalogues raisonnés. It must. The problem arises when scholarship, commerce and authentication become so closely intertwined that outsiders may reasonably wonder whether the catalogue is functioning only as a neutral map, or also as a market gate.
This was the point raised in my podcast, Is It? The Art Mystery Podcast, in the episode “Is It a Modigliani?” with Georgina Adam and Carina Popovici. The episode discussed the new Restellini catalogue, the inclusion of more than 100 works, and one painting in particular, Jeune Femme au Col de Fourrure, which Art Recognition had analyzed using AI and which Restellini had also included in his catalogue. Art Recognition’s point, as I understand it, was not that Restellini had used its AI analysis. This was never stated nor implied in any text or conversation to which I was privy. (In the interest of full transparency, please note that I am the art history advisor for Art Recognition, and so my understanding of this situation comes with knowledge both behind the scenes and in the public eye). It was that a painting previously excluded from the leading Modigliani catalogues had now been accepted independently by both an AI image-analysis firm and the leading human Modigliani scholar. Art Recognition had analyzed the painting on behalf of a client and determined it to be authentic. Restellini’s inclusion of the painting appears to have been independent of Art Recognition’s analysis, and I have seen no evidence that he relied on the AI finding. The leading AI image analysis system and the leading Modigliani scholar had agreed.
The inclusion of an AI authenticated painting in a catalogue raisonne is a remarkable moment. It shows how computational image analysis and connoisseurship, too often placed in opposition, can converge. They should: the best AI models are built on catalogues raisonne. Firms like Art Recognition teach their AI how to recognize an artist’s hand based initially on one or more catalogues raisonne. These were assembled by the leading human scholars and should combine connoisseurship, provenance research, and material forensic findings. The AI then becomes a fourth pillar on which authentication can stand even more firmly.
Restellini objected to any suggestion that he had relied on Art Recognition’s findings (to my knowledge this was never stated or implied anywhere), and The Art Journal reported that the wording of Art Recognition’s announcement was later revised. The revision clarified the distinction between Art Recognition’s analysis and Restellini’s independent catalogue decision. But the broader significance remains: this appears to be a landmark case in which AI authentication and traditional expertise arrived at the same conclusion on a work earlier scholars had not accepted. In my experience, this convergence happens more often than the public debate around AI authentication might suggest. It has been covered in multiple episodes of my podcast, Is It?, and the results announced in podcast episodes have been reported in major media, including The Times und The Guardian (twice).
That should be good news. It should also make us ask harder questions about the mechanisms by which works are accepted or rejected. A catalogue raisonné is, in theory, an instrument of clarity. It attempts to list the accepted works of an artist, often with provenance, exhibition history, bibliography, technical information and critical commentary. In practice, it can also be a market switch. On one side is a painting “attributed to” Modigliani, difficult to sell and shadowed by doubt. On the other side is a painting “included in Restellini,” or “included in Ceroni,” or “included in Wildenstein.” The object may not have changed. The price has.
History gives us several cautionary tales.
The most notorious early example is the Otto Wacker scandal. In the late 1920s, Wacker, a Berlin dealer, sold a group of fake Van Gogh paintings into the market. Several were accepted by respected experts and some were included in Jacob-Baart de la Faille’s 1928 Van Gogh catalogue raisonné before the fraud was exposed. De la Faille later withdrew them. A rival catalogue, published by the other leading Van Gogh scholar, H.P. Bremmer, was also consulted. The two experts could not agree (de la Faille couldn’t even agree with himself). Wacker was eventually convicted. The scandal was devastating because it showed how quickly a catalogue could confer legitimacy, even when the underlying evidence was weak. Once a work appeared in the definitive printed record, it could move more easily from dealer to collector, from doubt to confidence, from suspicion to invoice. The catalogue did not create the forgeries, but it helped them breathe.
Die Debatte um die Wacker case also shows why catalogues raisonnés are never just books. They are infrastructures of belief. They provide the language through which dealers sell, collectors buy, museums borrow, insurers insure and lawyers argue. De la Faille was not a villain in this story. He was an expert who was deceived, as other experts were deceived. That is precisely why the case matters. The danger does not require corruption. It can arise from pressure, incomplete information, confidence, ambition, proximity to dealers, or the simple human desire to solve a problem neatly.
A more recent and artist-specific warning comes from the tangled world of Modigliani authentication itself. Christian Parisot, long associated with Modigliani scholarship and with the Modigliani Institute, published or oversaw Modigliani-related cataloguing projects and issued opinions that affected the market. His name later became entangled in legal proceedings and fraud allegations involving works attributed to Modigliani and to Jeanne Hébuterne. In 2010, he was convicted in Italy in a case involving fake Hébuterne drawings. In later reporting, investigators described a broader concern that questionable works were being legitimized through certificates, exhibitions, expert authority and institutional-seeming structures. The details are messy, as Modigliani always is. The lesson is cleaner: when an expert system becomes too close to market circulation, it can begin to function as a mechanism for conferring value.
Again, the point is not to smear all Modigliani scholarship. It is the opposite. Modigliani’s market needs serious scholarship because his oeuvre has been so damaged by forgery, poor documentation and competing authorities. He died young, poor and ill in 1920. He traded works for food, rent and drink. His provenance trails are often fragmentary. His style is recognizable enough to be imitated and elusive enough to keep experts arguing. For decades, Ambrogio Ceroni’s catalogue was treated as the market’s benchmark, even though everyone knew it was incomplete. Restellini has long argued that a new, more scientifically grounded catalogue was needed. That argument is reasonable. The question is how such a project should be funded, disclosed and insulated from market pressure when its conclusions can create or destroy enormous value overnight. The simplest solution is to be entirely transparent about funding sources. Then people can come to their own conclusions.
A third example involves Renoir and the still life Basket with Peaches and Grapes. This case is less a proven forgery scandal than a study in how even disputed inclusion can unlock a sale. The painting had been doubted by some experts and excluded by the Wildenstein side of Renoir authority. It was later included in the Dauberville Renoir catalogue raisonné with a caveat noting that its authenticity had been contested. According to Le Monde, that inclusion helped make possible a 2012 sale for €768,000. The work’s authenticity remained debated, but the catalogue entry had done enough. It did not end the argument. It changed the work’s commercial life.
That may be the most important lesson of all. Catalogues raisonnés do not need to be universally accepted to affect the market. They need only be accepted enough. A caveated inclusion may still be better than no inclusion. A new catalogue may compete with an older one. A dealer may cite the authority that helps and avoid the authority that hurts. Buyers often do not have the time, knowledge or appetite to parse decades of expert warfare. They want a name, a page number, a plate, a reference. “Included in the catalogue raisonné” becomes a phrase of comfort, even when the underlying reality is complicated.
The Wildenstein name hovers over any discussion of catalogue authority because the Wildenstein Institute produced some of the most important catalogues raisonnés in the modern art market, including for major French artists. The Wildenstein family were also dealers. That dual role has long fascinated and troubled observers. To be clear, a dealer-scholar is not automatically conflicted in a corrupt sense. The history of art expertise is full of dealers who knew more than academics because they handled the works, saw the archives, spoke to collectors and understood the market’s hidden routes. But when the same ecosystem sells works and helps define the authoritative corpus, disclosure and governance matter. The art world has often relied on reputation where other fields would demand formal separation.
There are also less scandalous but still revealing examples of market-funded scholarship. Gagosian’s support for Diana Widmaier Picasso’s work on a Picasso sculpture catalogue raisonné, reported years ago by The Art Newspaper, was discussed openly as funding from a major gallery for a project with obvious market implications. That does not mean the project was improper. Openness is precisely what makes the example useful. A gallery may support scholarship; indeed, some scholarship would not happen without gallery money. But the relationship should be visible so readers and buyers can weigh it accordingly.
This is where the Restellini-Pace situation becomes interesting, without requiring any accusation. Pace publicly celebrated the catalogue’s release through programming and has announced a future Modigliani exhibition curated by Restellini. Pace has publicly collaborated with Institut Restellini around the catalogue and has announced a 2027 Modigliani exhibition curated by Restellini. It has said it was not behind the catalogue. These facts can coexist. The ethical question does not depend on proving secret patronage. It is enough to ask what the best practices should be when a commercial gallery promotes an authentication authority whose conclusions may affect works in the gallery’s orbit.
A sensible system would require three things. First, full disclosure of funding. If a catalogue raisonné is supported by a gallery, collector, estate, foundation, owner or group of owners, that should be stated plainly in the publication. Second, disclosure of any works in the catalogue that are owned, represented, consigned by or otherwise commercially linked to funders or major promoters of the project. Third, independent governance: an advisory board, published criteria for inclusion and exclusion, and a clear account of what technical, documentary and connoisseurial evidence was used. This would not eliminate disagreement. It would make disagreement more honest.
The old model depended on authority. The new model must depend on transparency. Authority will always matter in art authentication. No algorithm can replace looking, and no scientific test can answer every question. But authority without disclosure is no longer enough, especially in markets where one scholarly decision can move eight figures.
The irony is that the catalogue raisonné began as a tool against confusion. It was meant to bring order to the chaos of studios, estates, dealers’ records, exhibitions, inheritances, lost works, destroyed works and fakes. At its best, it remains one of the great instruments of art history. But because the market has learned to monetize certainty, the catalogue raisonné has also become a financial instrument. It can rescue a work from limbo. It can consign another to oblivion. It can make collectors rich. It can wipe out value. It can provide scholarship. It can provide cover.
That is why the current Modigliani debate matters, even if every work in Restellini’s catalogue proves entirely correct and every rumor about Pace proves baseless. The stakes are larger than one scholar, one gallery or one artist. The Modigliani market is simply the sharpest version of a problem that runs through the upper end of the art world. When scholarship and commerce meet, they should do so in daylight.
A catalogue raisonné should be trusted because its methods are strong, its evidence is visible and its conflicts are disclosed. It should not require blind faith. Modigliani, who spent much of his short life outside institutions of power, would probably have found it grimly amusing that the fate of his paintings now depends on institutions so powerful that a page number can be worth millions. But that is the world we have built. The least we can do is read the entries carefully, and ask who benefits when a disputed painting receives its place in the accepted record.
