Author: Erin Stephenson

Penschilderijn*, also known as “penpaintings“, involve drawing an image in black ink on top of a substrate prepared with a white lead oil ground.  The technique originated in seventeenth century Holland and was popularized by one of its most skilled practitioners, Willem van de Velde the Elder.  The artist’s painting, Dutch Ships Near the Coast, became the first penschilderij in an American public collection when it was gifted to the National Gallery of Art in 1994.  Treatment of the painting began in 2010, which gave conservators the rare opportunity to conduct an in-depth study of the materials and techniques utilized in its creation.  Kristin deGhetaldi headed the treatment of the painting and presented the current study findings and treatment results to the Paintings Specialty Group.

Willem van de Velde the Elder built his career on pen paintings but his beginnings were much more humble.  He was born in 1611 as the son of a skipper and spent most of his early years on ships, giving him a natural familiarity with navigation and the sea.  He was also an excellent draughtsman and became skilled at sketching maritime scenes.  As his skill improved he was sought for victory images and his clientele of wealthy patrons increased.  He gained significant notoriety with his penpaintings as early as the end of the 1630s and it was said that his penschilderij were considered more popular and valuable than his other works.

Penpaintings were often done on panel or vellum primed with oil.  Working atop these surfaces with pen and ink made it difficult to make corrections to the composition.  Van de Velde was a perfectionist who was easily dissatisfied with the quality of his work.  If he did not like a sketch he would go over the basic outline in wet ink and quickly press the image to another substrate and begin again.  He also utilized both fine line drawing and washes to create his images, with washes becoming more prominent in his work by the 1650s.  Washes provided the advantage of covering large areas quickly without the need for intricate underdrawings.  This allowed the van de Velde workshop to generate larger penpaintings at a faster rate in order to meet teh demands of the market.

A visual analysis of Dutch Ships Near the Coast was conducted in comparison to van de Veldes other known works and some characteristics stood out.  Although it is dated to the 1650s, the work is smaller than his other penpaintings and it lacks the expected fluid washes in favor of fine linear strokes.  The penpainting does have an underdrawing, though it remains unclear whether it was sketched in silverpoint or graphite.  In addition, the ground layer composed of calcium carbonate instead of the slightly darker ground that is common in his similar works.  Finally, unique raised lines are present where the ink is applied and in other areas of the white ground.

Scientific analysis was conducted in an attempt to clarify some of these discrepancies.  Cross-sectional microscopy revealed two layers of lead white, with the topmost layer containing particles that were more finely ground.  The presence of only carbon black in the ink design confirmed that iron gall ink or bone black was not used by the artist.  Analysis using GCMS was conducted in an attempt to identify the binding medium of the ink, though the tests were inconclusive.

Conservators decided to create a reconstruction of the penpainting to gain insight into the identity of the oil binder and find possible causes for the raised lines.  Linseed oil was used in the reconstruction but it yellowed quickly, leading conservators to believe a slower drying oil was used in order to avoid the discoloration.  Next conservators tested reed and quill pens dipped in gum based ink to determine how the ink was likely applied.  Reed seemed like a good candidate but they produced broader and less precise lines than the sharp, fine lines created with quills.  Goose quills were ideal but quills from raven and crow feathers were also acceptable.  It was hypothesized that the sharp quills may have scratched the ground before it was totally dry and created the fine lines.  However, that did not account for the raised nature of the lines or the fact that they existed in areas where ink was not applied.  At that point conservators wondered if the lines could be the result of engraving techniques.

The Dutch artist Experiens Sillemans was a contemporary of van de Velde and also created penschilderijn.  Sillemans was known to use printmaking practices such as engraving in the creation of his works.  The technique involved pressing a freshly inked engraving on to a primed support.  To create raised lines, however, van de Velde would have had to press his inked copper plate into the soft preparatory ground of the support.  Given the art historical evidence, it still seems unlikely that van de Velde used this practice as no two of his penschilderij are alike.  In addition, there are no prints in his oeuvre to suggest he was a practiced engraver.

Ultimately the technical study of Dutch Ships Near the Coast left more questions than answers and conservators are hopeful that future study will lead to greater understanding.  At that point it was time to address the treatment of the piece.

Examination of the painting revealed fills and overpaint, discoloration, flaking, and crumbling around the fills.  The abraded surface was almost ghost-like in some areas and the face of one man in the foreground was completely lost.  The painting was stabilized using sturgeon glue.  During removal of the varnish layers, Kristin did not have to worry about solubility issues.**  Once the painting was given an isolation layer of MS2A varnish, losses were filled using a mixture of Aquazol 200 and Modostuc.  To begin the inpainting process Kristin  isolated the painting with MS2A and then used pigmented micropens under magnification to conduct a painstaking recreation of the abraded areas.  The damaged background was reinforced with thin HB micro graphite sticks.

A question and answer session followed Kristin’s presentation and someone asked what additional theories she may have regarding the cause of the fine lines.  Kristin said she has a few weak theories.  One theory is that the penpainting was put in the sun to bleach and dry, during which time the black ink may have absorbed more heat and created the lines.  Her second theory was that a slower drying oil like walnut or poppy may have left the grounds soft and created uneven drying which could have led to the raised lines.  She stressed that more study is necessary.

I thought this was a very interesting presentation and look forward to hearing about future developments in the study of Willem van de Velde’s penschilderij.

* A past study by David Freedberg, Aviva Burnstock, and Alan Phenix refer to these works as penschilderijen.  Since I am not fluent in dutch, nor an expert on penpainting, I deferred to the spelling used by Kristin deGhetaldi.

** The question of solubility was raised during the question and answer session, to which Kristin replied she detected absolutely no solubility issues in the materials of the penpainting, especially since the painting had already been subjected to harsh overcleaning in the past.

American painter Clyfford Still (1904-1980) was a leading Abstract Expressionist artist.  In November of 2011 the new Clyfford Still Museum opened in Denver, Colorado.  For more than seven years prior to the opening conservators in the ARTEX Conservation Laboratory worked to prepare Still’s paintings for travel and exhibition.  One of those paintings was the oil on canvas,  1943 (PH-286).  Barbara Ramsay carried out the treatment of the painting and kept coming back to the question, “Should we hold the artist accountable for the materials and techniques that he has used, or should we attempt to reintroduce aspects of his original intent as we perceive them?”   As one might expect, the answer is often complicated and open to interpretation.

1943 (PH-286) is a key early painting in Still’s oeuvre that serves as an excellent documentation of his artistic process.  The painting represents his transition from abstracted figurative work to the complete abstraction for which he is known.  It is understood that Still considered the painting to be an important work and he exhibited it more often than most of his other works.  The painting also important because it has the inscription “White and Black” on the verso, which may be an early title because he did not completely abandon titling his work until some time in the 1940s.

The stretched painting was executed on unsized cotton duck canvas with a white ground.  The canvas showed signs of being unstretched and restretched more than once.  It was slightly stained but otherwise it was generally stable.  Examination of the paint layers revealed mingling of matte and glossy areas and a discolored surface coating.  The areas of highest gloss appeared bluish white under ultraviolet irradiation but whether there was a local or overall surface coating remained unclear.  There were also areas of marked drying cracks in the black paint layers but all of the cracks appeared to be stable.

After minor consolidation, the structure of the painting was stable so the focus of treatment was on the aesthetic qualities.  Barbara’s approach was to open up a dialogue between curators, the representatives of the Still estate, and conservators to discuss the interpretation of the condition and determine a course of action that prioritized minimal intervention.  Unfortunately, such conversations are often complicated and this case was no exception.  Some of the stakeholders felt the painting should be left as is.  Others wanted the drying cracks to be inpainted.  Still others wanted a complete aesthetic treatment to remove and replace the discolored varnish, saturate the matte areas, and inpaint the drying cracks.  Barbara wanted to do more research before determining her ideal course of action.  The critical questions:  could it be done safely and should it be done at all?

To answer those questions Barbara focused on Still’s working process.  GCI analysis suggested that he used titanium white ground, zinc white, carbon black, organic colorants, and some commercial tube colors on his paintings.  Many of Still’s paints are known to be particularly sensitive to water and organic solvents.  The variable gloss in the paint surface posed several other issues for consideration.  The matte areas could have indicated an original difference in leanness in the paint.  It was possible that Still applied an overall surface coating that soaked into areas of underbound paint over time and left an unexpected variable gloss.  There was also a chance that Still applied the coating to specific areas of the painting.  So, did Still apply a coating?  Was it local or overall?  Did he intend or like the uneven surface?  Research turned up examples where Still did both localized and overall surface coating applications.  It was documented that paint that appeared “too dry” was not his original intention.  However, what did he mean by “too dry”?  Was this an “overly matte” surface or did it describe a surface that had traces of efflorescence?

Additional study by the GCI identified a drying oil on the surface and it was determined that Still likely coated the painting with the oil in the 1970s.  Examination under ultraviolet irradiation supported the theory that the oil coating was absorbed by leaner areas of paint, resulting in localized matte patches.  Although the oil was applied by the artist there was no documentation to indicate what effect he was trying to achieve.  Since the discolored coating was very disfiguring and misleading, it was decided that it should be removed.  But could it be removed?

As expected, solvent sensitivities in the paint layers complicated the removal of the surface coating.  As a result, it was determined that selective and partial cleaning was the most viable option.  Using solvent compresses the coating was reduced in the whites, grays and colored areas, but was left untouched over the black paint.  A varnish was not applied either to the cleaned areas or to the matte black paint passages.

The last step in the treatment was inpainting of the craquelure.  Barbara preferred to leave the craquelure as evidence of the artist’s materials and working methods but finally agreed to inpaint the most distracting craquelure with a reversible medium.  At that point the treated painting was ready for exhibition.

This was a complicated project full of the difficult questions that seem inherent in the treatment of modern and contemporary art.  As a result, the conclusion of Barbara’s presentation was followed by a very interesting question and answer session.  I was unable to keep a written record of the various dialogues that occurred but I can recall one question and answer that I thought was interesting and could lead to additional discussion:  One conservator asked Barbara why she decided to inpaint the craquelure when Still had chosen to exhibit it for so many years in that condition.  Barbara replied that she had considered this factor and believed it was a valid reason not to inpaint.  She did not want to inpaint the craquelure but she feared the other alternative was that the painting would not go on exhibit.  She felt it was necessary to compromise on inpainting the cracks to ensure that such an important painting would be included in the inaugural installation in the new museum.  She noted also that Still had not intended that the drying craquelure would form.  She felt the museum’s desire to show Still’s work to the public for the first time looking its very best was also a valid concern. What do you think?

Judy Dion’s study of 15th-century Spanish retables adds to the limited information available on the impressive altarpieces that have undergone less technical and historical research than their Northern European and Italian Renaissance cousins.  Dion’s research is the result of five years of collaborative efforts during her Mellon Fellowships at the Balboa Art Conservation Center and the Philadelphia Museum of Art.  The study involved extensive non-invasive examination of panel paintings from the collections at the San Diego Museum of Art and Philadelphia Museum of Art, as well as paintings in Spain, the United States, the United Kingdom, and mainland Europe.  Dion focused her investigation by concentrating on panel paintings made in the Spanish regions of València, Catalonia, and Aragón. Her findings address construction methods specific to Spanish retables, how construction features on individual or smaller sections of panels can provide clues to their placement within a complete retable, and the history and changing attitudes regarding alterations of disassembled panels.

To summarize the discussion of construction, retables consisted of a central panel surrounded by additional panels usually in the form of an inverted T-shape that measured anywhere from two to seven meters high.  The panels were in a fixed position and surrounded by framing components that served as pictorial and transitional elements between the scenes.  Review of the fabrication methods showed the panels, most often constructed from poplar or pine, were assembled using butt-joined boards with battens.  Transverse battens were used in combination with either vertical or diagonal battens but the latter were never used together.  All of the retables from the Valencia region examined in Dion’s study had diagonal battens and metal dowels were common additions.  Any flaw in the panels were filled with wooden shims and gesso before painting commenced.  Nonwoven plant fibers and fabric were also used in preparation of the panels.

In my opinion the most interesting part of the construction was during the assembly step that included the framing elements.  The frame components were attached to one of any two adjacent panels and then overlapped the face of the second panel. Raised framing elements were installed during the final assembly.  This construction caused a bare edge and gesso burr on the side of the panel with the attached framing element, while the paint layer extended to the edge of the side that was overalapped.  As a result, individual panels often appeared asymmetrical when retables were disassembled.

The disassembly of retables was often a response to changing aesthetics, demand for profit, and damages.  Individual or smaller panel sections could hold a higher value than an intact retable.  This led to frequent alterations of the panels to improve their appearance as stand-alone works.  Such alterations included but were not limited to repainting, the addition of false elements, and cutting down of the panel edges.  Fortunately, shifts in the methodology of collecting institutions encourage the recognition and appreciation of the historical context of fragmented works.  Panels that were altered for the aforementioned reasons are increasingly the subjects of reversal treatments.  In some cases the alterations are left visible, while in others missing or altered elements are restored to recover evidence of their original appearance.

This is a very abridged overview of Judy Dion’s thorough research and presentation due to my mere mortal status as a note taker.  I encourage anyone interested in this subject to seek out the conference postprints and/or approach Ms. Dion with any questions regarding her study.  She applied her findings to a hypothetical reconstruction of Jaume mateu’s The Birth of the Virgin, from the John G. Johnson Collection at the Philadelphia Museum of Art.  It is an excellent case study and definitely worth a review to gain a greater understanding of this research.