The current conservation of Rembrandt's Night Watch
In 2019, the Rijksmuseum launched Operation Night Watch, a highly ambitious effort to research and conserve what is likely the most famous 17th century Dutch painting in the world. This is the biggest conservation project in the history of the Rijksmuseum, and it involves a multidisciplinary team of more than thirty people, conservators, scientists, curators, data analysts, photographers and general staff, among others. In this photo here I'm the good-looking guy on the far right in the second row and I don't know who that man in the center is. But anyway, the work is carried out in front of the public inside a specially constructed glass box.
A variety of state-of-the-art imaging technologies were employed: macro-XRF, macro-XRPD, RIS, OCT to name a few. These added extra insights to those that had already been discovered with older technologies. Older technologies such as X-radiographs that were made in the 1970s.
That research phase of the project was concluded just last month and the results were presented in our own symposium on 8th December. Other symposia are planned for the future, the first of which has been announced for April, the 11th to the 14th. This evening I would like to briefly discuss some of the new insights that have been made, specifically, those regarding the genesis of the painting. In the spirit of the current exhibition in Frankfurt, which examines Rembrandt not in splendid isolation but in the greater context of the Amsterdam art world of the 1630s and 1640s, I will devote the greater part of my talk to The Night Watch as one of seven group portraits, group portraits executed for the great hall of the Kloveniersdoelen in the 1640s. One of the most fascinating discoveries made by the Operation Night Watch team is the presence of a light colored sketch throughout the painting. This compositional sketch was executed with chalk rich paint and is visible in the macro-XRF calcium map.
A compositional sketch in this medium has never previously been found in Rembrandt's work. It is this compositional sketch that is responsible for the ghostly appearance of the dog at the lower right. Due to abrasion, the sketch has become more visible than intended. The sketches also become exposed in the faces of some of the civic guardsmen. The chalk rich paint was also used to define the forest of pikes on the right side of the composition, three of which did not make it into the final composition. There is also a sword present in the calcium map between the figures of Frans Banninck Cocq and the lieutenant Willem van Ruytenburch that was subsequently painted out.
Especially intriguing are the forms Rembrandt sketched in the upper middle of the canvas. Inside the arch, some of these shapes might be alternative positions of the company's banner. But what are these squiggly lines supposed to define? If anybody has suggestions, we would be very happy to hear them. In addition to this, light colored compositional sketch some of the elements have been sketched in dark brown. Unlike the light colored chalk based drawing, such dark brown ones have been found in a number of Rembrandt's paintings.
These vary in composition, but most often contain bone black and red lake. The position of the feet of the Musketeer to the right of Willem van Ruytenburch was initially indicated with these dark lines. The final position of the feet was altered, but the sketched-in feet are still visible to the naked eye and very distinctly in the RIS-SWIR image on the right.
RIS is short for reflectance imaging spectroscopy and SWIR is short for shortwave infrared. Another instance where a dark sketch was used as the metal collar or gorget of the pikemen directly above Willem van Ruytenburch in the picture. The lance held by the civic guardsman next to him, was also initially sketched in this dark brown paint. The use of these sketches in The Night Watch, the light one as well as the dark ones perhaps sheds light on the fact that no preparatory drawings on paper have come down to us. Dozens of small changes Rembrandt made to the composition of The Night Watch were unearthed by the Rembrandt Research Project on the basis of the X-radio graphs that were made in the 1970s.
A number of other changes were discovered using the macro-XRF scanner, and this will perhaps come as a bit of a déjà vu for you. For example, the vivaciously colored feathers that once adored the helmet of Claes van Cruijsbergen, the Randassier next to the standard-bearer. From the iron, copper and mercury maps We know that these feathers must have been composed of red, yellow and azurite. Karin Groen published a groundbreaking essay, and yes, the pun is intentional. In 2005, in the fourth volume of the Rembrandt Corpus on Rembrandt's innovative use of quartz grounds, beginning with The Night Watch. A quartz ground is a natural earth ground layer remarkably rich in quartz and with a high amount of clay minerals.
Up until 2001 they were found in forty-five paintings on canvas by Rembrandt and his circle and in no paintings by other artists. An advantage of the quartz ground, which was invariably one layer, is that the painting to which it was applied could be more easily rolled up. Is this an indication that The Night Watch was not painted on location? I will say more about this later. It has also been pointed out that a quartz ground would have been an economical solution for a painting the size of The Night Watch.
Only one layer was needed and the ingredients sand and clay were cheap. Rembrandt deliberately left the quartz ground layer exposed here and there, using it as a middle tone. The sword held by Randassier Claes van Cruijsbergen can serve as an example of this. As quartz grounds contain titanium dioxide, the macro-XRF titanium map is a good indicator of its presence here. Another interesting example is sergeant Rombout Kemps extended right hand. The remarkably unfinished appearance of which is due in part to this use of the ground layer as a middle tone.
But not everywhere was it intended to be so much in evidence as it now is. Abrasion has caused more of the quartz ground layer to be exposed in the dog, for example than was initially intended. Not only did Rembrandt use the quartz grounds for the first time in The Night Watch the Operation Rembrandt team has also discovered a new pigment in the master's pallet. This is an arsenic containing pigment that was primarily used in Dutch still lifes.
It was also common in Italian painting, but has not been found elsewhere in Rembrandts oeuvre. It was discovered in The Night Watch in the impasto rich area of the embroidery of lieutenant van Ruytenburch's buff coat and his sleeves. Before I continue, I would like to thank the Operation Night Watch team and in particular Anna Krekeler, Katrien Keune and Annelies van Loon. This slide that I have just presented were taken from earlier presentations by them, with their permission, of course. I hope. In addition to applying every conceivable method of technical examination to The Night Watch, we are also examining this icon of 17th century Dutch art in its original context, the great hall of the Kloveniersdoelen one of one of three civic guard headquarters in Amsterdam. At the time of its construction in 1630, it was the largest reception room in the metropolis and would remain so until the town hall on Dam Square was completed in 1655.
Initially undecorated, seven monumental group portraits, six of them civic guard pieces, were installed in the first half of the 1640s, transforming the great hall into the most impressive assembly space in the Dutch Republic. The three placed on the long wall opposite the windows of the hall in 1642, The Night Watch, a painting by Nicolaes Pickenoy and one by Jacob Backer, were the largest ever executed. The Night Watch would remain Rembrandt's largest painting until he executed the Conspiracy of the Batavians under Claudius Civilis in 1662.
Some scholars have argued that there was another reason the great hall was exceptional. Rather than coming together in a piecemeal fashion over time, as they had in the other Amsterdam civic guard headquarters, the decorations of the great hall followed a predetermined plan. This is the argument. Among the arguments put forward for this theory is the contention that the civic guard paintings filled the walls of the great hall from floor to ceiling.
In a 2015 article Gwen Tauber and I demonstrated that this was definitely not the case where the southeast wall where the first and last canvases of this series hung. Joachim von Sandrart's portrait of the guardsmen of district 19, which originally hung to the left of the fireplace on the southeast wall is dated 1640 two years prior to the paintings on the longwall. But it had been assumed since 1986 that this painting was originally conceived for another location, either another civic guard headquarters or another place in the Kloveniersdoelen and was later radically altered in order to fit into the scheme of the great hall. Based on the X-radiographs taken during the paintings restoration in 1983 and on the hypothesis of their Rijksmuseum conservators at the time, it was claimed in the literature between 1986 and the publication of our article in 2015 that Sandrart's portrait originally had a horizontal format. One scholar claimed that it was in 1642 that the painting was altered into a vertical format.
Another scholar claimed it was as late as 1645 that this supposedly occurred. The figures on the cut off pieces were supposedly added to the new part above the lowest seen. And others were inserted into the lower original part of the composition. Re-examination of the technical evidence, however, resulted in a very different interpretation.
The painting had a vertical format from the very beginning. But only included wether now the lower and middle pieces of canvas. These do have different weave patterns, but as can be seen in the angle map, there is no cusping on either side we've seen. The lower piece of canvas is approximately 205 centimeters or three ells high. The maximum canvas width.
This and the lack of cusping here makes it highly unlikely that the lower piece of canvas was ever cut down before the middle stroke was added. Such a strip would have had to measure at least 20 centimeters. There is vague cusping on the left side of the lower piece at regular intervals of 10 to 12 centimeters, which continues in the middle piece of the canvas. Indicating that both pieces were stretched as a whole, rather than individually, on the right side there is no cusping visible in the angle map, but again, in a consistent fashion that is neither on the lower nor the middle pieces of the canvas.
The upper border of the painting would have been located where what is now the upper scene, as cusping is very much present under this scene, as can be seen in the angle map. The uppermost strip of canvas was probably added to the painting when it was transferred with the civic guard portraits by Flinck and Pickenoy from the Kloveniersdoelen to the Town Hall on Dam Square where they hung together on one wall in the burgemeesters room. This would have occurred in 1753, the same year the city's paintings restorer Jan van Dyck was paid 866 guilders for relining and repairing the three canvases. Significantly, canvas from the same bolt was used for the 35 centimeter wide addition to Sandrart's painting. And this was also used to repair the portrait at the bottom under Captain Bicker's feet. It was also used to the right of the table with the bust of Marie de' Medici, but I haven't indicated that on this slide.
A very narrow strip of the same canvas can also be found at the bottom of Flinck's Portrait of the Guardsmen. The addition to Sandrart's painting was probably made so that it would have had the same height as the portraits by Flinck and Pickenoy. All three works, which hung as an ensemble in the burgomasters' room, where altered to a corresponding height of around 340 centimeters, the same height they now have.
Sandrart's portrait was originally only three meters and eight centimeters high, whereas the wall on which it hung in the great hall was as much as 4.53 meters high. Govert Flinck's picture on the other side of the fireplace now measures just under 340 centimeters, and we believed in 2015 that it was originally not much higher than that. In the meantime, X-radiographs have been made of the Flinck and the relatively deep cusping apparent in the upper edge of the angle map confirms our suspicions. It can be concluded, therefore, that these two paintings, at any rate, did not cover the entire surface of the southeast wall and there heights even differed from each other. From contemporary copies after The Night Watch in Jacob Backer's painting on the southeast or longwall we know that they have been cut down. X-radiographs
of Backer's and Pickenoy's paintings have now also been made, which was quite the feat given their enormous size. But these X-ray radiographs still await stitching and analysis. Alas, hopefully this will provide evidence for the original dimensions of these works. I can assure you, however, that the original dimensions of The Night Watch will be revealed in the very near future. Discovering the original dimensions of the civic guard portraits is not the only reason for making and analyzing these X-radiographs. By the way, Bartholomeus van der Helst's contribution to the great hall has also been included in this X-radiography campaign.
But we are also of course extremely eager to discover on the hand of these these X-rays, among other things, whether or not the six artists involved in the decorations responsible used the same materials. Pigment samples have also been taken from all seven paintings, to this end. And although further analysis is needed, it can be concluded that Rembrandt was the only one to use a quartz ground. The others all used a double ground, commonly found in paintings executed in Amsterdam in the first half of the 17th century, which consisted of a layer of red baked clay covered by a grey base layer. Research into the ratio of the pigments in the mixtures, the size of the grains and their distribution, and the presence of small additions of pigments other than the primary components may reveal that Pickenoy's and Backer's grounds, both from 1642 after all, came from the same batch, although I don't think this is very likely.
In the remainder of this evening's talk, I would like to examine two, shall I say, art historical notions associated with the hypothesis that the decorations of the great hall followed a predetermined plan. The first of these suggests that Hendrick Uylenburgh may have coordinated these decorations, and at any rate secured the commissions for Rembrandt, Govert Flinck and Jacob Backer, who were all newcomers to Amsterdam and not well known to the civic guardsmen of the Kloveniersdoelen. The second notion considers the six civic guard portraits, or at least those by Bartholomeus van der Helst on the northwest wall and the three including The Night Watch on the longwall to be a continuous frieze, with one painting flowing into the other. In my consideration of these two claims, location plays a pivotal role.
It has long been known that Bartholomeus van der Helst included his own self- portrait in the civic guard portrait of District 8 of Amsterdam under the command of Captain Roelof Bicker, my great great grandfather. Van der Helst arranged the troop in front of the brewery De Witte Haan or The White Rooster which was located on Nieuwmarkt. Van der Helst self-portrait is in this small circle here on the screen. The identification of the building, The White Rooster brewery, has nothing to do with what little we can see of its architecture, but rather with the white rooster painted on the barrel on the right on the right, straddled by a guardsman dressed in black with a bandolier and powder cartridges.
Partially obscuring our view of the barrel is the halberd of sergeant Joachim Rendorp, the owner of The White Rooster. Three of his brothers-in-law are also present in the painting, including Dirck Joosten Rijskamp who hangs out of the window holding a rummer in his right hand. The White Rooster was where the district civic guardsmen socialized. It is number 1 on Balthasar Florisz.' map of 1625. This is a detail of the map. This map, by the way, and all the information I am telling you about van der Helst is derived from Judith van Gent's monograph on the artist.
Number 2 on the map is the location of van der Helst's own home. It is obvious, therefore, that van der Helst would have known the guardsmen he portrayed very well. Until now, van der Helst has been the only painter known to have lived in the district of the guardsmen that he portrayed for the great hall of the Kloveniersdoelen. This is Jacob Backer's portrait of the officers and some of the guardsmen of District 5. A large number of these men were herring merchants, including the man I have circled here.
He was one of the two sergeants of the company and had the very appropriate last name Visch. His full name was Teunis Jansen Visch. Four of the guardsmen lived on Nieuwendijk in District 5, which I have outlined in red here. The yellow line is Nieuwendijk and Teunis Jansen Visch lived at number 19, number 2 on my map. Number 1 on my map was the house on Nieuwendijk owned by Jacob Backer's brother Tjerck Backer.
As Jaap van der Veen has pointed out, Jacob Backer probably lived in the bakery at 6 Nieuwendijk owned by his brother, as no other address is known for him in Amsterdam at this time. It seems obvious to me that Teunis Jansen Visch and Tjerck Backer who lived in such close proximity to one another would have had a profound bond based on the fact that their last names reflected their occupations. Both men probably also made fun of Tjerck's brother Jacob because he was a painter. But his last name wasn't schilder or painter, but rather bakker or baker. At any rate, it is more than likely that Jacob Backer knew the guardsmen of District 5 very well.
They in turn, would have surely been aware of Backer's ability to execute the civic guard portrait because he had already executed a now last one for the Kloveniersdoelen, their headquarters. Rembrandt would have known at least one of the guardsmen of District 2 before he was given the commission. This was Jan Claesz. Leijdeckers, who's nephew Gerbrand van den Eeckhout probably trained with Rembrandt in the late 1630s and was later described by the artist biographer Arnold Houbraken, as Rembrandt's great friend.
But as Bas Dudok van Heel has pointed out Rembrandt undoubtedly knew someone much higher up in the hierarchy of the Kloveniersdoelen: Pieter Reael. He was portrayed by Govert Flinck in 1642 as one of the four governors of the Kloveniersdoelen in the painting that hung above the fireplace on the southeast wall of the great hall. And I have circled him in red here. Reael was Banninck Cocq's predecessor as captain of District 2.
Rembrandt portrayed Reael's favorite nephew Johannes or Jan Wtenbogaert in a 1639 etching known as The Gold Weigher. It seems very probable that Reael would have recommended Rembrandt to Banninck Cocq. Rembrandt may, in turn, had recommended his former pupil Govert Flinck as painter of the portrait of the four governors. Although Flinck would later become one of Rembrandt's principal rivals in Amsterdam and on the Amsterdam art scene he had not yet reached this pinnacle in the early 1640s. And while Rembrandt may have had a contemptuous regard for some of his clients, we also know that he was very generous and supportive of his pupils and fellow artists.
It is also possible that Johannes Wtenbogaert also introduced Flinck to his uncle. Houbraken informs us that it was Flinck's habit on Sundays after church to visit some of the most important private collections in Amsterdam, among them that owned by Johannes Wtenbogaert. Unfortunately, we do not know whether this was before or after Flinck portrayed the governors of the Kloveniersdoelen. Be this as it may Backer, Rembrandt and Flinck were probably very well acquainted with the guardsmen they portrayed in the great hall of the Kloveniersdoelen and didn't need introductions from Uylenburgh. According to one recent scholarly treatment of the great hall, the room was conceived as a unified sequence of civic guardsmen flowing from one painting to the next.
The flight of steps in Backer's painting, for example, leads into van der Helst and the building on the right of Pickenoy's picture continues into Backer's painting. A closer look at the locations depicted in the four paintings on the longwall and van der Helst's painting on the northwest wall reveals that each painter had a different point of departure. While it seems likely that the artists involved agreed to, or were instructed to paint their figures life size, there was no such agreement in place when it came to the architectural backdrops. As for the steps in Backer's painting, running up to van der Helst, any illusion that the guardsmen in one of the paintings could step into the other is negated by the position of the steps further back in Backer's painting. Because of their position in space, the figures on those steps had to be less than full length.
And the figures in van der Helst's painting seem therefor to be giants in comparison. Well, both van der Helst and Pickenoy chose to include a landmark in their portraits. That left no question as to the district that was represented while the architecture in Backer's and Rembrandt's pictures is purely fictional. Van der Helst, as we have seen, made reference to The White Rooster brewery, which was a gathering place for the guardsmen of District 8. The architecture in Backer's painting has been described as Flemish classicizing in the past, but it is at any rate the product of the painter's imagination.
The building on the right of Pickenoy is also fanciful. But in the left background we see the no longer extant Jan Ruytenburch Tower, which occupied the southeast corner of District 4 and served as its clock tower. Pickenoy's guardsmen are assembled at some point along the west side of Singel, one of the major canals in the district.
The notion that the building in front of which Pickenoy's guardsmen are standing I'll start that again. The notion that the building in front of which Pickenoy's guardsmen are standing continues into Backer's painting is, well, simply incorrect. A close look at Backer's painting reveals a classical porch jutting out from either the front or the back of the structure, depicted by him. It is impossible that this porch would have sprouted out of the side of the building and Pickenoy's paintings. Moreover, why would Backer's guardsmen of District 5 want to portray themselves as living in District 4? The district represented by Pickenoy.
The architectural disunity would have continued for contemporary viewers when they moved on from Pickenoy's painting to Rembrandt's. The gigantic arch in the latter has been compared to the temporary arches that were erected for the joyous entry of Maria de' Medici in 1638. A more convincing comparison has been made to the Sint Anthonis Gate, one of the gates leading into and out of Amsterdam. While this comparison, too, is not 100% convincing, the structure does seem closer to a city gate than the temporary arches erected for Maria de' Medici. A city gate also makes more sense in the context of Rembrandt's painting. The inscription on the page opposite a drawing after The Night Watch in Banninck Cocq's family album describes what's going on in the painting: "The young Lord of Purmerlant, as captain, orders his lieutenant the Lord of Vlaerdingen to march out his company of civic guardsmen" In the past it has been suggested that Banninck Cocq's troop is marching out to welcome Maria de' Medici to Amsterdam or to take part in a parade.
These seemed to me to be rather banal activities for Rembrandt to have immortalized, and they don't explain the gigantic backdrop of a fictitious city gate. The only explanation that fits is that the troop is marching out to defend the city against its enemies. In actuality, Banninck Cocq and his guardsmen never did such a thing, but then it is also unlikely that they kept company with little girls dressed in fantastic outfits and carrying chickens at their sides or shrunken Musketeers from the 16th century firing miniature muskets.
The architecture of the gate is fanciful because the gate is symbolic for Amsterdam as a whole. Rembrandt did not represent an actual episode in the city's history, but the function of the Amsterdam civic guard throughout history. This must have been very flattering to Banninck Cocq, his lieutenant, Willem van Ruytenburch and the 16 other guardsmen from District 2 portrayed by Rembrandt. It does not seem very likely to me that there was someone, Hendrick Uylenburgh or whomever else, who was given the task or took it upon himself to coordinate the individual civic guard portraits to make a whole of them. If this were the case, then that person failed miserably or actually wonderfully. Leaving everything aside except the architecture the viewer springs from actual locations in the city to purely imaginary ones and from imaginary ones to ones that are both imaginary and symbolic.
What a bumpy ride, but one that must have entertained visitors to the great hall of the Kloveniersdoelen immensely. Thank you very much.