A Flemish trace for the Madonna Cagnola’s master (Zanetto Bugatto)

A few years ago, while consulting the folder of the anonymous fifteenth-century Provençal at the Roberto Longhi Foundation in Florence, I had the occasion to come across an image of a painted panel of unknown location, depicting the Cristo morto sorretto da un angelo, which an inscription in the file reported to have once belonged to the Bartolini collection in Florence. It is likely that, in interpreting the painting’s strongly Flemish component, Longhi had been led to link it to the Provençal area, partly because of the particular engraved ramages decoration of the gold background, characterized here by subtle plant and floral whirl motifs.

A decorative pattern that is certainly not uncommon in the Lombard and Ligurian, as well as Franco-Provençal, spheres, but which in this case turns out to be completely consistent with that which appears in the series of panels referred in his time by Longhi himself to that “very noble Lombard anonymous” who goes by the name of Maestro della Madonna Cagnola, in whom even recently the elusive Zanetto Bugatto has been sought to be recognized.  We know that, between 1461 and 1463, he served an apprenticeship in Brussels at the workshop of Rogier van der Weyden. An identification questioned by those who, rejecting the Lombard relevance of the various paintings, preferred instead to discern possible similarities with the expressly Ligurian and Provençal area.

In the case of our panel, the affinity with the works referred to the latter master is certainly not limited to the ornamental treatment of the background, but more generally affects the figurative part, marked by a similar and precise transposition of Flemish motifs, always interpreted with a lucidity and acuity that, in certain subtle passages of light, nevertheless betrays a Lombard or at least a Po Valley origin. Although partly altered by the obvious repainting, Christ’s face bears strong similarities to that of the Madonna now preserved at Villa Cagnola in Gazzada, whose extreme regularity of features, starting with the nose and mouth, as well as the shape of the eyebrows that overhang the wide half-closed eyelids characterized by the same cut, is repeated. The hand, with tapered fingers, also appears entirely identical to those of the Madonna Cagnola, while the subtle doses of light that characterize the accurate anatomical definition of Christ’s body recall similar solutions found in the San Giovanni Battista now in a private collection, which originally belonged to the same complex. The same can be said for the wholly graphic sharpness with which the intricate, manifestly Rogerian draperies of the Angelo’s robe as well as the shroud are depicted, as also appears in the various plates of the series referred to the anonymous artist, not least the one with the Sant’Orsola disclosed by Andrea De Marchi, where we singularly find a very similar motif of the hand with the long little finger holding up the mantle.

The latter element, moreover, refers explicitly to a Flemish prototype, as attested by a number of paintings that have come down to us that should certainly be placed in relation to our panel, which can therefore be imagined to bear the memory of a lost Flemish version perhaps executed by Rogier van der Weyden himself. In fact, the composition of the painting presented here is known to us through a number of Flemish panels, as evidenced by both the version in the Musée Royaux des Beaux-Arts in Brussels, referred to an artist from the southern Netherlands active in the late 15th and early 16th centuries, and the lesser-known one in the Monastery de Pedralbes in Barcelona, the latter part of a diptych attributed to a follower of Rogier depicting in the other panel the Vergine con il Bambino. Except for the variation of Christ’s left hand, placed in both versions at the bottom to collect blood from the wound of the side, the compositional pattern is in essence entirely coincidental, so much so as to suggest that a higher common prototype may have existed before these versions, although no trace of it remains.

From this point of view, however, it is interesting to note the affinity that can be discerned with the different iconographic version of Maria che sorregge il Cristo sofferente, of which some ancient versions painted by Memling and his workshop have come down to us, starting with the certainly autograph one in the National Gallery of Victoria in Melbourne, which Friedländer had already hypothetically attempted to relate to a lost triptych referred to Rogier van der Weyden, recorded in an inventory of the collection of Margaret of Austria. Although no works of this subject are known to be directly attributed to van der Weyden, it is nevertheless possible to imagine that a version of this composition also existed in origin, possibly characterized by some slight variations.

This is what would be implied by a graphic translation by a Flemish engraver, the so-called Monogrammist F, in which some details present in Memling’s different versions return, although as far as the composition is concerned the subject of the engraving bears more similarities to the panel published here, a sign that there must have been a different version of the Cristo morto sorretto dalla Vergine, on the right of the scene, which is also known from a number of late replicas .The iconographic coincidence with our panel is indeed compelling, just observing the body and face of Cristo, as well as the gesture of the Vergine holding up the shroud with her left hand, characterized by the same wide draperies that cover, as in our panel, part of the lid of the tomb also placed in the left part of the depiction. Obviously, without other clues, it is difficult to establish whether the different version with the Engelpietà, could have derived from this particular model, reproposed by the engraving, which was then well spread in the Flemish sphere as confirmed by the various replicas mentioned above, to which the more modest one already preserved in Madrid in the collection of don Josè Lazaro can also be added.

What has been said can only confirm the importance of our panel, which in fact comes to return us further evidence of the artist’s pro-Flemish preferences, which can be assumed in some cases to have developed directly in contact with some of the most illustrious examples of “ponentine” painting, as attested by the oft-mentioned case of the Columbaaltar of the Alte Pinakothek in Munich, painted by van der Weyden around 1450-1455, or other works that came out of his workshop. With regard to the panel presented here, alongside a possible origin from a Rogerian model, it seems to me that elements can also be grasped that go back more directly to the production of Dirk Bouts, to which the somewhat elongated and pointed form of the angel’s face clearly refers, comparable, in the minute definition of the features and the small fleshy mouth, to that of certain Madonnas painted by the Dutch painter, as confirmed, for example, by the comparison that can be established with the panel in the National Gallery in London.

Certainly it is not possible to discern in our panel that “geometrizing lucidity” that has been caught in the master’s other panels, from which the Cristo in morto sorretto da un angelo differs also because of the more sketchy definition of the punched nimbuses. This could perhaps call into question its belonging to the same complex to which the Madonna Cagnola must have belonged, although this hypothesis should not be discarded in any case, as suggested in this same issue of the journal by Andrea De Marchi, who is oriented to consider this Engelpietà an apicle element of the mentioned polyptych. Alternatively, one could think, as Federico Cavalieri suggested at the time [link here; Ed], that the complex was surmounted by a panel with the Crocifissione or the Annununciazione. Only a direct observation of the panel, however, could provide us with other elements in this regard while allowing us to verify whether the painting in question may have undergone particular tampering or remaking over time, both in the punching of the nimbuses and in other parts. It must be said, however, that the decorative motif of the Cristo’s halo, characterized by five-stamped rosettes, does not appear to be very dissimilar to that used by a Lombard artist, whom some have wanted to identify with Ambrogio Bevilacqua, for the Madonna adorante il Bambino, from the Pinacoteca Malaspina in Pavia, which Cavalieri has suggested, albeit indirectly, to relate to the production of the same Master of the Madonna Cagnola.

This, too, is evidence of the master’s Lombard origin, which in more recent times has been confirmed thanks in part to the discovery of a fragmentary fresco with Magdalene embracing the cross part of a larger Calvary, certainly attributable to him, on the altar of the sacristy of the Milanese abbey of Chiaravalle, made known by Cavalieri himself upon the recommendation of Stefania Buganza. The panel conserved in Santa Maria di Dicinisio in Sormano in Valsassina has also proved to be no less probative in this respect, and it shows, as Andrea De Marchi has noticed, precise iconographic derivations from some of the panels from the same complex attributable to the Master of the Madonna Cagnola, which can therefore be assumed to have been located in Lombardy from the very beginning. As Cavalieri has also noted, the comparison becomes very close for the figures of the Battista, the Sant’Ambrogio and the San Gerolamo, although it cannot likewise be ruled out that other parts of the Sormano panel are derived directly from other lost panels in the same series. I am especially thinking of the San  Bernardo, which in the particular posture of the hand holding the crosier closely resembles the Sant’Ambrogio in a private collection, as well as the San Gregorio already in the Museum of Art in Toledo and the San Nicola in a private collection, the latter two probably part of a different complex, equally related to the Master of the Madonna Cagnola and his workshop. This is also the case with other details of the panel of Santa Maria in Dicinisio, as shown, for example, by the hand of the saint holding the book on the far right almost traced one would say to the strongly Rogerian hand of the San Lorenzo now in a private collection. On closer look, even the kneeling figure of San Giuseppe also placed on the right side of the Sormano painting shows, in the particular way of rendering the folds of the robe and mantle veiled, references to the Flemish tradition.

In this case, however, the most direct match seems to be found, rather than in the anonymous artist’s models, in the fragmentary fresco preserved in the church of Santa Maria degli Angeli in Vigevano, which has reliably been proposed to be linked to the name of Zanetto Bugatto. In fact, the fragment depicting the Adorazione del Bambino is what remains to us of a larger, lost pictorial cycle made at the behest of Galeazzo Maria Sforza in 1472 by Bonifacio Bembo, Leonardo Ponzoni and Zanetto himself, to whom the documents would seem to attribute a leading role. As Binaghi Olivari has rightly observed, the fresco certainly cannot be referred to the quite different Bonifacio Bembo, let alone, in the absence of more probative clues, to Leonardo Ponzoni; while, as far as Bugatto is concerned, the name could very well be spent also in view of certain undoubted Flemish connotations that have already been discerned pertaining the two surviving figures.

Although partly altered by repainting, the figure of San Giuseppe still presents in the best-preserved parts an evident quality of execution, which can be discerned, for example, in the definition of the face, vaguely Foppesque in taste but at the same time also Nordic in the precise definition of the character’s physiognomic and psychological features. Just as unquestionably Flemish appears to be the model from which the artist seems to have drawn the saint’s long, tapering hands, easily assimilated, even in the detail of the bent thumb, to those of certain donors appearing in the van der Weyden panels, from whose models he also seems to have drawn the motif of the narrow flared sleeves at the bottom covering part of the hands, which singularly returns identical also in the San Giuseppe of the Sormano panel, proving the existence of undoubted relations between the two works. After all, as Andrea De Marchi has pointed out, this type of “delicate and chiseled” hands, of clear Rogerian influence, constitutes a bit of a recurring signature in the works of the Master of the Madonna Cagnola, as both the Sant’Orsola, published by the scholar, and the San Gregorio, now accesible after the recent restoration, attest.

Similar considerations can also be made with regards to the Vigevano fresco, in relation to the admittedly more modest figure of the Adoring Virgin, profoundly altered by the remakes in the areas of the face and hands, but still clearly discernible in its precise translation from a Rogerian model, which Binaghi Olivari has correctly recognized in Pierre Bladelin’s triptych (Berlin, Gemäldegalerie der Staatlichen Museen), where the Virgin returns almost identical both in the posture and in the definition of the drapery. All these elements lead to the plausible assumption that the fresco was executed by Zanetto himself (perhaps in this last figure assisted, as De Marchi suggests, by Ponzoni). Thus it seems logical to think he was then involved in the creation of the various ducal portraits that as it is known completed the scene of the Adorazione del Bambino placed by the high altar. Despite the urgings that the very young artist must have received during his stay in Flanders, one can equally imagine that for the portrayals of the patrons he adhered to the model then in vogue for court portraiture, that is, placing the portrayed in profile.

Zanetto must have been, after all, accustomed to this kind of practice also because of his documented activity for the Milan mint, for which we know he supplied cartoons for medals and coins on several occasions. Correcting an earlier hypothesis by Syson, Marco Albertario, has in fact made it possible to link to his name the making of the first gold ducat, which features on one reverse the portrait of Duke Galeazzo Maria Sforza then in his early twenties. The coin had been minted in 1467, that is, when the young painter who had provided the design had returned from Flanders barely three years before. In my opinion, rather than the graceless portrait of Galeazzo Maria miniated on the frontispiece of Raffaele da Vimercate’s Liber iudiciorum of 1461, recalled by Albertario (Milan Biblioteca Trivulziana, cod. 1329, c. 2r), the profile of the coin finds precise correspondence with the one that appears in a fragmentary parchment (Milan, Trotti Bentivoglio collection) that was later readapted, together with another depicting perhaps his mother Bianca Maria, as a small picture, both of which were disclosed by Malaguzzi Valeri, when they were still in the collection of Guido Cagnola. Similar precision can be spotted in the definition of the profile, which stands out almost as if it were a real medal, thanks also to the shading of the background, as well as the accurate rendering of the iron mesh and armour, of almost identical make to that worn by the young duke on the coin. 

There is no doubt that this beautiful miniature portrait, so far strangely ignored by studies, may be a useful clue to better assess the characteristics of Sforza court portraiture during the seventh decade of the century, that is, when Zanetto’s activity in this particular field is widely documented. Moreover, it is precisely the particular stylistic connotations of this portrait that suggest the possible intervention of a painter rather than a miner. This has also been observed with regard to another miniature portrait depicting Duke Francesco Sforza, annexed to codex 786 of the Biblioteca Trivulziana in Milan, which shows strong similarities in both technique and execution with our cutout. As difficult as it is to assess the quality of the colour texture of the Trotti Bentivoglio sheet through a black and white photograph alone, one can equally imagine that this may reflect, in the rendering of the complexion, the pearly and tender tones that characterise the portrait in the Trivulzio codex, with which it shares the richness of the chiaroscuro effects, which also contribute here to giving the image a sense of greater volume. Nor does the meticulous technique with which the lustre of the sumptuous worked brocade of Duke Francesco’s robe is rendered on the Trivulzio sheet, through the use of subtle touches of light, in an entirely Flemish taste, appear too distant.

The close similarities between the two different portraits, as well as suggesting a possible execution by the same artist, could lead one to think that they had originally been part, together with the one mentioned with Bianca Maria, of the same codex illustrated at the beginning by a number of full-page portraits of the most famous members of the ducal family, as appears in other important Lombard Renaissance manuscripts. Of course, this is not intended to prove that it could have been Zanetto himself who painted these portraits, although such a hypothesis might seem, on the basis of what has been observed so far, highly suggestive. Obviously, in this case, one would have to think of an activity carried out by the artist also in the field of miniature painting, which is not shown by the numerous documents that have come down to us. While waiting for other clues, it would therefore be best to limit ourselves, in the light of the observations made, to considering these miniatures simply as possible reflections of the portraiture put into practice in these same years by Zanetto, whose artistic personality we imagine to be of a very different calibre.

Therefore, the most likely hypothesis is the one aired at the time, on the basis of Longhi’s observations, by Ferdinando Bologna and more recently accepted by Andrea De Marchi and Federico Cavalieri, i.e. to recognise in the Master of the Madonna Cagnola, due to the unquestionable Flemish and Lombard influences, the same Zanetto. A hypothesis that is now confirmed by the panel presented here, since, as we have said, it probably bears the memory of a lost prototype by Rogier van der Weyden, which the artist, as demonstrated in other cases, may have met directly in Flanders, perhaps by travelling to Brussels, as in fact happened to Zanetto. Moreover, as has been noted, even the lacunose fresco in Santa Maria degli Angeli, the only certain evidence of Bugatto’s activity, presents in its best preserved parts similarities with the well-known production of the Master of the Madonna Cagnola. This is demonstrated, for example, by the carefully studied head of San Giuseppe, worthy of inclusion, due to its “pathetic intensity” as well as the accuracy of certain details, alongside that of the Sant’Ambrogio from a private collection in New York, or that of San Gerolamo, also made known for the first time by Roberto Longhi. Similar considerations can also be made, albeit in a less direct manner, with regard to the face of the Cristo morto presented here, which can also be compared, due to certain precise Morellian details, to that of the San Giuseppe in Vigevano, which, together with the fragmentary Maddalena in the abbey of Chiaravalle and perhaps the Crocifissione in Buccinasco (assuming it is the work of the same artist), are all that remains of his activity as a fresco painter, which he often carried out alongside other well-known Lombard painters.

Moreover, it is precisely on the basis of the fragmentary Adorazione del Bambino in Vigevano that Stefania Buganza has come to identify as the possible author of the preparatory cartoons for two stained-glass windows in the Certosa di Pavia, those with the Natività and San Bernardo, the same Zanetto Bugatto, who probably painted them in the last phase of his activity. By this time, as the scholar notes, the memories of the experience in Flanders seem to be distant in time, replaced by the more modern stimuli from Padua and Ferrara or those that came to him from Vincenzo Foppa, with whom he had the opportunity to collaborate. Yet even in these possible last works, the artist still shows, especially in the Natività, that he has not completely forgotten that first approach with the Nordic world, whose models he now reinterprets according to a new monumentality that clearly betrays the comparison with the most up-to-date Foppa experiences, which he had to look at while still having in mind, it seems to me, certain Franco-Fouquettian models that he was able to get to know during his stay in France in 1468. These are the same observations that were made about the Madonna Cagnola and the various panels connected to it, which respond perfectly, in the subtle combination of Flemish and Lombard suggestions, to the figurative path we imagine for Zanetto Bugatto, within which the panel presented here also fits perfectly, further evidence of his activity prematurely interrupted due to the sudden death that seized the artist early in 1476.

April 15, 2024