On the Conservation of Japanese Paintings

My path to becoming a conservator of East Asian painting began fifteen years ago when I left Manila for Tokyo to study framing at Kanda’s Gallery, the largest distributor of contemporary Japanese prints in Japan. It was here that my apprenticeship in meeting the exacting requirements of Japanese craftsmanship for simplicity, elegance, and fineness really began. It was here, too, that I began to appreciate the importance of art presentation.

In my free time, I visited museums and became fascinated with the beauty of Japanese hanging scrolls and folding screens. I was amazed at their excellent condition. The silk brocade borders were so rich in colour and, although they were centuries old and had the patina of age, the paintings were well-preserved. Art books often show only the paintings themselves, overlooking the way they are presented, I had never really realized that there could be so many variations in scroll mounting. I started reading about conservation in general and also began experimenting with prints given to me by Mr. Kanda, the owner of the gallery. I had always meant to return to Manila to start a framing section in my parents’ gallery. I did, but I was not content with being just a framer, I wanted to become a conservator as well.

With the help of my father’s friend in the Japanese Foreign Ministry, the doors were opened to me. It would have been extremely difficult, even impossible, for a foreigner to be accepted as an apprentice in one of the best studios for art restoration and conservation in Japan. Even for a Japanese, gaining entry into a renowned studio such as Handa Kyuseido would not have been an easy task.

Life as an apprentice was tough. I received a low-stipend from my sensei (master), Tatsuji Handa, for which I was very grateful. By accepting me into his studio he had already done me and honour. I worked twelve hours, sometimes longer, a day, and from six to seven days a week. I sat on my haunches, in the Japanese manner, and often did the same tedious work for several days, like pasting washi (hand-made paper) together and rolling them up. Many times I wanted to give up. But when the day was over, I always looked forward to the next lesson. ‘Under-the-table work’ is how my sensei describes the livelihood of a conservator and mounter of paintings, because there is little recognition for the hard work, which requires decades to learn and master. He himself began when he was a teenager, because each painting is different. I learned patience and discipline, two virtues vital to my craft. After several years my sensei and an official at the Tokyo National Institute for the Conservation of Cultural Properties recommended that I attend a programme on restoring Japanese cultural properties. The course lasted three years, and in November 1989 I was awarded a certificate by the Japanese Ministry for Cultural Affairs, making me one of the first non-Japanese to be certified to restore cultural treasures. 

As a rule, all paintings must be treated with the same care. But one thing for sure, a good painting will always be a good painting no matter how damaged it is, and a bad painting will always be just that, no matter how beautiful the silk around it.

In restoring paintings, the number one priority is safety. All the procedures must not cause future damage to the painting. Since most of the paintings I receive are badly damaged, brittle, and falling apart, I take extra precaution in handling and storing art works. The second priority is preventing future damage, by using the best materials and applying the skills I have learned. (I visit my sensei two to three times a year, and every two years I stay for a month and a half to retrain and polish my skills. For me, Handa Kyuseido is the best source of information on art conservation, from materials to the latest technology.)

I begin by cleaning the art work, being very careful not to overdo the process. About eighty per cent of the work in restoring a scroll is devoted to preservation. I see to it that even the minor damages are corrected. I believe. However, that the antiquity of the painting must be maintained, and that damage wrought by time can only be toned down so that it will not be too noticeable. This is where problems with a client can arise. Sometimes a client will ask me to redo the painting as if the damage had never existed, or to paint in something that seems to be missing. I advise against overpainting, but if the client insists, I will do it. Since the client is paying for my services, it is only proper that I do what is asked. After all, it is the client’s own appreciation of the art work that matters above all else.

Being a conservator is not just about restoration. I work in the shadow of the artist, following his every brush stroke. Every detail is closely studied — the calligraphy, the artist’s seal, the silk borders, the end knobs — and recorded on paper and in photographs. I research the painting’s history, and try to find who the artist was, who were its owners, whether it formed part of a collection, and how many times it has been restored.

It has always been a practice in Japan for collectors to have their paintings routinely restored. A painting that has been restored properly and is carefully stored can last two to three hundred years. It was also customary practice for scrolls to be displayed seasonally or only on special occasions. Since scrolls were kept rolled most of the time and stored in the wooden, air-tight boxes, they lasted a very long time. It has only been recently that people have begun displaying their scrolls for long periods of time or as permanent fixtures. Exposed to pollution, varying temperatures and levels of humidity, and other elements, a scroll has little chance of maintaining its liveliness for more than a few decades. As for screens, keeping them folded for six months out of a year will prolong its life. If the folding screen is displayed flat against a wall, covering it with a cotton blanket will slow down the process of deterioration. 

The mounting of paintings onto scrolls originated in China, but the Japanese took it to another elevation. The scrolls became a very important part of Japanese art, culture, and spiritual life. They also occupied an important place in Japanese architecture, from the most humble buildings to towering castles. Temples had their religious paintings, the nobles their portraits. Even peasants kept scrolls. The oldest style of mounting that exists in Japan originated from China. Called Yamato-hyogu, this style carried the influence of Song (960-1279) and Yuan (1279-1368) dynasty mounting.

Silk brocade reproduction in Japan runs into thousands of styles and colours, some dating back to the eleventh century. In fact, many styles were copied or derived from Chinese silk that can no longer be found in China today. This makes the style of Japanese mounting more complicated than Chinese mounting. Presently, the Qing dynasty (1644-1911) style is used in Chinese mounting no matter how old the painting; the scrolls are composed of one to three colours of silk, are simple in design and light in colour, and are selected from a few dozen designs. In Japanese style mounting, it was also customary to use silk from India and Persia.

History is the best source of information as to what types of silks were available, and the patterns and colours in vogue during a certain era. In the many years that I was studying in Japan, I listened very carefully to my sensei when he explained the reasons for his selecting certain colours and patterns. There is no written rule about preference and taste. Learning the basics of completing a combination that is both beautiful and true to the time takes at least a decade; yet, as Handa sensei often would say, taste is a gift — you either have it or you do not. 

There are many colour variations in Japanese mounting, from one to four different combinations. The silk brocade should correspond to the style of the painting and the colour should complement the art work. I often buy low-quality scrolls for their good silk borders or end knobs, and re-use them on a better painting. I also cut up old woodblock prints to patch up the holes in the better ones. 

Working on scrolls is like opening a Pandora’s box; surprises await. Often layers of problems are uncovered between the surface of the painting and the final backing. Recently, I undertook a project to conserve a group of paintings from the Honolulu Academy of Arts. One of the paintings, ‘Three Vinegar Tasters’ by Kanō Isen’in (Naganobu; 1755-1828) (Figs 1a and 1b) had the typical signs of improper mounting. Its hard texture was due to the heavy paste used to back the scroll, and the surface was cracked in too many places. After dismantling the scroll, the colour pigments were consolidated with nikawa (animal skin glue) and immersed in a liquid solution to wash away the dirt residue and raise the pH balance to a safer, less acidic range. The painting thus became cleaner, the colours alive and the circular line in the background more visible.

(Fig. 1a) ‘Three Vinegar Tasters’ after conservation

By Kanō Isen’in

(Naganobu; 1775-1828)

Hanging scroll, ink and light colours on paper

Height 44.5 cm, Width 56.5 cm

Gift of George H. Kerr

Honolulu Academy of Arts, 6156.1

The painting had six layers of washi, too many for a painting of this size. The cracks on the surface were due to the thick backing of washi and the thick paste solution. This is a tragic combination. After I removed all of the washi backing, I found that the painting was still too thick. The only way to prevent cracking was to thin out the back of the painting. This was done on a light table with the front of the painting flat against the table’s surface. Using a pair of tweezers, it took three days to remove the thin layers of washi. Then the cracks and tears were reinforced with narrow strips of fine usumino (a type of washi where the fibres are uniformly entwined and the surface is even). Before mounting, I presented several sets of silk to Stephen Little, curator of Asian art at the Honolulu Academy of Arts. Since there are hundreds of silks to make the selection from, it is difficult to come up with the perfect match. We discussed which would look well with the painting. In restoration, I feel that there should always be an exchange of ideas between the scholar and the technician. 

(Fig. 1b) Figure 1a before conservation

On a fan painting of plum blossoms by Ogata Kōrin (1658-1716) in the Honolulu Academy of Arts (Figs 2a and 2b), the in-painting was inadequate and the mineral pigment that was used was imitation gold, which is copper-based and therefore eventually oxidized into a green colour. Since East Asian paintings use a lot of mineral pigments, it is essential to have a background in chemistry. Understanding the chemical structure of a substance and how it will react to a solution is important. As much as possible I exhaust all the safe means of treatment before using a corrosive chemical. To remove stubborn stains such as foxing and mildew, I sometimes use a mixture of weak hydrogen peroxide and ether. The solution is carefully applied several times on the stain with a fine brush. Since the mixture is weak, it often takes several hours to remove a stain. 

(Fig. 2a) ‘Plum Blossoms’ during conservation, reversed on the light table with old backing removed

By Ogata Kōrin (1658-1716)

Fan painting on a hanging scroll, ink and colours on gold leaf

Height 23.5 cm, Width 23.8 cm

Honolulu Academy of Arts, 3009.1

I never throw away old pieces of silk, washi and silver and gold leaf paper no matter how small. This habit serves as a good purpose. In this case, I used gold leaf that I had saved from an old folding screen to patch up the holes in this painting. I was fortunate that the colour of the gold matched. The look of old gold leaf is extremely difficult to duplicate. No mineral pigment will stick to gold for a long period of time, and this is why the red colour of the plum blossoms was flaking off, a process that cannot be stopped permanently. The best thing to do was to steam nikawa to the red pigment. Then a small amount of gold powder was painted on the eges of the fan. It is best for the pigment to cut a little over the edges, making it possible to show more of the painting that is usually covered by the silk brocade border.

I used nishiki for the inner silk border. Nishiki silk originated in China and found its way to Japan during the Nara period (645-784). Hōryū-ji in Nara has a large collection of nishiki. The weave of nishiki has various colour patterns, but what makes it even more special is the gold thread that is added. It is one of the most exquisite and elaborate silks that can be found, and is therefore expensive. The cost of the a piece measuring seventy-two centimetres by thirty centimetres can run to US$500, but nevertheless, nishiki can only be purchased by the bolt. The end knobs of the scroll were the only things I could salvage from the original mounting of Kōrin’s painting. Made of carved lacquer, they were unusually fine and rare.

(Fig. 2b) Figure 2a after conservation

The river landscape attributed to Tenshō Shūbun (act. 1414-65)(Figs 3a-c) in the Academy had numerous holes. The artist’s seal was scratched out for reasons one can only speculate. There were a lot of insect droppings on the surface of the painting, yet it was difficult to distinguish between these droppings and paint from the artist’s brush. So when in doubt, let it alone. I dyed a sheet of Chinese paper of equal thickness to that of the painting to patch up the holes. The painting was placed on top of a light table after which a piece of patching paper was placed over the hole, and the outline of the hole traced on it. The shapes were cut out with a graphic knife. It is important the patch does not go beyond the edges of the hole nor should it be smaller than the hole; it must be a perfect fit. The patch was then pasted onto the front of the painting. 

(Fig. 3a) ‘ River Landscape’ on a light table before conservation, showing old reinforcements

Attributed to Tensō Shūbun (act. 1414-65)

Mid-15th century

Hanging scroll, ink on paper

Height 55.6 cm, Width 32.3 cm

Purchased with funds donated by the Robert F. Lange Foundation

Honolulu Academy of Arts, 7021.1

(Fig.3b) Detail of Figure 3a during conservation, with the old backing removed, before patching

The upper and bottom layer of the silk mounting dates to the Muromachi period and comes only in white. The threads of this silk are fine and loosely woven, allowing light to filter through. I dyed the silk a yellow-brown colour. Yellow because that was the colour used during that era, and brown to give it a time-worn look. I used dried cones of an Adler tree to get this effect. To give the blue silk a deeper colour and a softer feel, it was rolled over a metal pipe and pounded evenly for several days.

(Fig. 3c) Figure 3a after remounting, showing the new silk brocade mounting

Another painting conserved was a portrait of Shoki (Zhong-kui) by Ōishi Yoshio (1659-1703)(Figs 4a and 4b). It was very interesting that pages of a poetry book were used for the border instead of the usual silk. The whole mounting style could have been changed to give the painting more life, but what was most important was the historic value of the scroll. The verses are excerpts from several poems of various famous poets.

(Fig. 4a) Detail of the top of ‘Shōki’ (Zhongkiu) before conservation, sowing creases, old insect damage and old repairs

By ōishi Yoshio (1659-1703)

Edo period, late 18th century

Hanging scroll, ink on paper

Height 28.5 cm, Width 23.3 cm

Gift of George H Kerr in honour of Sir George Sansom

Honolulu Academy of Arts, 6153.1

(Fig. 4b) Figure 4a after conservation

The technology used for the restoration of this painting is fairly new. Being a good conservator also means using the most recent technology if it has been proven safe. Since the damaged area was large, the paper patch had to be weak. This prevents the borders of the damaged areas from flaking off. All the holes were patched using a leaf caster machine, which uses the same principle as paper making. After studying the make-up of the washi, I concluded that a combination of kozo (mulberry) and gampi (thymelaeaceae) paper fibre would be the best choice to use in patching up the holes. The art work was then placed in a chamber, secured with a rigid screen with holes large enough for the fibre to go through, and subsequently lowered into the water. Old kozo and gampi were cut into very small pieces and mixed in a blender for a few seconds. The solution was then poured into the leaf casting chamber. When the chamber was raised above the water level, the fibre quickly filled up the holes. The painting and the pages were then placed on a suction table for a few seconds to give body to the newly formed fibre. Next, the painting and the pages were transferred to a presser where they were compressed for several hours until they completely dried. When everything was completed, the silk, the pages and the painting were backed with washi and made into a scroll again.