 |
Norway, ca. 1200 – 1250
University Collection of Antiquities, Oslo
The Madonna is represented in a symmetrical,
frontal aspect. She is seated on a four-posted throne,
wearing a long tunic, cloak and crown. Her child is missing.
The golden crown and garments are set with imitation jewels.
The drapes of the garments are in »dampfold«
style 1. She has pointed shoes that are resting
on a semi- circular base. The Madonnas face is oval, with
large eyes and highly curved eyebrows. The nose is straight,
and the mouth small. Her fluted hair is parted in the
middle and end on her shoulders with a braid-like ornament
on each side of her face. Height 90 cm, depth 24 cm, width
31.5 cm.
The Dyste Madonna belongs to a larger group of Madonnas
with many similarities; in construction, three-dimensionality
and polychrome attitude. In Norway, only one remains,
while in Sweden there are many, which more or less conform
with Dyste. There are a few elsewhere in Scandinavia,
one in Iceland 2 and at least one in Germany 3.
This shows the connection with a wider European tradition
4.
The production of this type of Madonna must have been
going on for some decades or more, from the Romanesque
well into the Gothic period. The drapes vary in style,
from concentric, almost abstract, to more naturalistic,
Gothic; while the main shape of the sculpture is kept
the same. Many of these sculptures are of high quality.
When observing the Madonnas, one could be tempted to think
that there might have existed a written instruction for
this type of sculpture. This could have been used by both
the woodcarvers and the painters, in a special religious
context, perhaps within a monasteric order.
The sizes of the Madonnas coincide to a great extent.
So do the main features of the sculptures, although they
also have many individual characteristics: variations
in drapery, types of wood, facial expressions and individual
proportions. It is likely that they were executed by different
craftsmen at different workshops.
It has been claimed that the Norwegian Madonna could be
a copy of one of the Swedish ones 5. I cannot agree
with that as she has quite individual characteristics
that remove her from her Swedish »sisters«.
There may have existed more of her type in Norway 800
years ago 6.
It would be interesting in the future to investigate further
into this group of Madonnas, on a combined technical and
art-historical basis. I am sure it would be possible to
reach a deeper understanding of this intriguing group
of sculptures.
 Construction
and Carving Technique
The main part of the Madonna is carved in limewood (Nat.
Tillia), the posts are turned in beechwood (Fagus sp.),
and the wooden pegs are of oakwood (Quercus sp.). The
sculpture consisted originally of more than 10 parts,
including the posts, nails, lost hands, lost child and
some parts which were added because of mishaps in the
carving, or large cracks in the wood during seasoning
7.
The Madonna is cut out of the tree trunk with the back
towards the core. A study of the annual growth rings below
the base suggests that the diameter of the tree was much
larger than the base of the sculpture. The growth rings
in the outer sections of the base, are tending out compared
to the curvature of the sculpture.
Cross-sections of polychrome layers from Dyste Madonna
A) hair:
l – chalk-glue ground
2 – yellow-orange mordant layer, containing red lead and
ochre
3 – gold leaf
B) cheek area:
1 – chalk-glue ground
2 – insulation layer (drying oil with resin?)
3 – light red – lead white and vermilion
The Dyste Madonna was hollowed out from the
back, after at least the preliminary carving of the front
was executed. The hollow is shaped like two rectangular
boxes; the smaller being the upper part of the body, the
larger the lower part. The base, which constitutes the
widest part of the sculpture, is hollowed out even more.
The posts have been secured to the main part with oak
pegs, (which are squarely shaped at both ends), through
drilled, tapering, round holes. Most of these penetrate
through to the hollow of the back. It can be seen that
the woodcarving of the front was done with the intention
of leaving the three- dimensional finish to be carried
out in the chalk/glue ground. In many places sharp toolmarks
are left as they are on the wooden surface. On the lower
part of the sculpture, a kind of drawing knife was used
to smoothen the surface. (Rows of parallell marks showing
on the surface.) Small gouges must have been used to carve
the drapery of the garments, and larger ones on both sides
of the sculpture, where the posts were to be joined to
the main part.
Madonna from Dyste – general view after restoration
It seems that flat-cutting tools have mostly
been used elsewhere (chisels, adze?). The fluted hair
and necklace edges, must have been cut with a pointed
knife. There is a hole in the top of the head with remains
of a wooden peg. This might have been made in order to
fasten the wooden block to the work-bench, or to prevent
cracking of the head during seasoing. It could also have
been used to insert a wooden handle, which would be a practical
device during the process of painting the sculpture 8.
The missing hands of the Madonna had tenons which were
inserted into mortises in the vertical frontal sides of
the sleeves.
Remains of old glue can be observed on the inner walls
of the mortises, along with wooden chips. The chips were
used to fill out the gaps in the joints. This arrangement
has not lasted, as the hands have fallen off. (The same
being true for almost all the Madonnas of the same group.)
The missing child was originally secured in the middle
of the Madonna's lap with a large wooden peg. A smaller
hole 4 cm below the knee, indicates where the legs of
the cild were fastened to the main sculpture. The hollow
of the reverse-side is relatively roughly made, with only
flat cutting marks showing (adze, chisel?). It was worked
from the base and upwards. A flat board may have covered
the hollow, as there are rows of nails along the sides,
which have not been hammered in completely.
The overall visual impression of the Madonna is a block-like
almost Ancient Egyptian frontality. If one investigates
further, the carving of details executed quite asymmetrically
around the middle axis. This difference between the right
and the left side makes the main impression of symmetry,
not static, but dynamic and alive.
Surface
preparation
After completion of the woodwork itself, small bits of
finely woven canvas were glued over the cracks and weak
parts of the surface in 5 different places (21 x 21 pr.
cm2) 9. Subsequently animal glue was applied to
the surface (insulation layer) 10. These glue layers
can be studied beneath the remains of the original layers
of the chalk/glue ground.
The preparation of ground layers with chalk rather than
gypsum was a common practice in medieval art north of
the Alps 11. Before the surface of the sculpture
was coated with the ground, a kind of porous putty had
been stuck into the crevices between the posts and the
main form to bridge the gaps. The putty seems to consist
of chalk, wood filings and animal glue. On top of this
the normal ground was applied. In some areas, at least
7 layers can be observed on cross-sections. On the reverse
side of the Madonna, only the head was coated with the
ground.
All unwanted irregularities and sharp edges on the surface,
have been smoothened out in the ground layers. The three-
dimensional »jewels«, are formed in the ground.
The thickness of the ground varies between 1 cm (imit.
jewels) to less than one mm in other areas. The most common
thickness seems to be between 1/2 -1 mm.
Reconstruction
of the polychromy
The study of painting materials and techniques is based
on surface examination using raking light and a binocular
magnifier. It is also based on sampling and cross-sectioning.
Visual comparisons with the Swedish Madonnas of the same
group have been very important.
The main impression of the Dyste Madonna in her original
state must have been one of supreme goldenness. The painted
areas must have seemed like coloured apostrophs in this
context. Her crown and clothing are golden, set with red
and green imitation jewels. The beautiful necklace she
is wearing is also set with the same red and green »jewels«.
Along the edges of the clothing, as well as around the
edges of the crown, there are red lines accenting the
lining. These lines also become a decorative, rythmic
element on the surface of the sculpture. (The use of red
lines is a common feature in Romanesque polychrome sculpture.)
The inside of the crown and the wide trumpet-sleeves,
were painted in the same red colour.
The Madonna's hair is golden. Her compl exion is pale
with vigorous reddish-pink cheeks. Her mouth is red, the
eyes large with black contour drawing and black eyebrows
(the pupils are missing).
The throne is golden with square, red-brown areas between
the toruses. The posts are also golden, decorated with
black horizontal stripes; one thicker, surrounded by two
thin ones. They were applied three and three above and
below the toruses of the posts.
Stratigraphy
of Paint Layers
The cross-sections were observed in reflected visible
and ultraviolet light. The study of the pigments and dye-stuffs
are based on visual observation and comparison with samples
of known identity. The colouring material of this period
is quite well known and may, therefore, be estimated with
some certainty.
|
|
Reconstruction study of original
The insulation layer between the ground and paint layers
fluorizes with a green/yellowish hue in U. V. rays. This
may suggest an oil-resinous binding medium. As this layer
has not been absorbed by the ground, it may mean that
animal glue or glair had first been applied to the surface.
(As it is not visible on the cross-section (U. V.
rays), it could indicate that the concentration of this
medium was low and not possible to detect. The binding
medium of the ground is also glue).
The oil/resin? insulation layer has not been applied in
the areas where metal leaf was stuck directly on to the
ground. As it is hard to get silver and gold leaf to stick
to a chalk/glue ground, without applying extra binding
media, it is likely that a (low concentration) glue/glair
layer, has been brushed on first as gilding »water 12«.
Solubility tests using 10 % KOH were carried out in order
to classify the mediums of the glazes and transparent
layers on top of the gold leaf on the crown and garments.
The red glaze and transparent layer on top did not dissolve,
only swelled slightly. On the other hand, the yellow glaze
between the gold leaf and the red glaze, dissolved completely.
This indicates that the glaze is oil/resin based.
The two layers on top may be egg-based. (Egg yolk in the
red glaze, glair in the »varnish 13«.)
The yellow glaze on top of the gold/silver leaf (except
on the cloak), fluoresces with a yellow/green hue in U.
V. rays. On the silver leaf on the posts, the yellow glaze
is drawn together into droplets. These observations suggest
that the binding media is oil/resin based 14.
The transparent layer on top of the yellow glaze (which
seems to be applied directly to the gold of the cloak),
must be considered to be a protective varnish. It is not
soluble in 10 % KOH, and fluoresces with a bluish hue
in U. V. rays, which may indicate that it is glair 15.
Gilding
Methods
Three different gilding methods have been employed, which,
with the addition of yellow glazes and a varnish, results
in different golden effects:
1) Imitation gold (water gilding).
Areas of application: Sides of throne, posts, outside
of cloak, inside of sleeves.
Silverfoil is placed directly on the wet (with glue?)
ground. The silver is probably polished to enhance its
reflective properties. A thick yellow glaze is applied
over the silver (probably based on heat modified drying
oil/natural resin ). This glaze gives the visual impression
of gold 16. This was a much used technique in the
whole of Scandinavia in medieval times and is considered
to be a »saving technique«, often applied in
less prominent places 17. It may also have been
used to achieve a special golden effect in some cases.
2) Water-gilding with real gold leaf:
Areas of application: Cloak, tunic, crown.
The gold is placed directly on wet (with glue?) ground.
It was probably polished (as with the silver), before
a yellow glaze was painted in. On top of this a transparent
varnish was applied. The cloak seems to have only the
transparent varnish layer. The yellow oil/resin? glaze
modifies the luster of the gold 18.
3) Mordant/oil gilding:
Areas of application: The hair.
As in the painted areas, an insulation layer was first
applied to the ground. Areas to be gilded were subsequently
covered with an orange/yellow pigmented oil mordant. The
gold leaf was conceivably applied when the mordant was
almost dry. No glaze or varnish has been applied on top
of the gold leaf. The colour of the mordant helps to hide
damages in the gold.
Reconstruction
Attempts (missing areas and chronology of the works)
The polychromy of the Swedish Madonnas of the same group
as Dyste seemed to agree almost totally with her colour
pattern 19. It was possible on this ground to make
an hypothesis about the missing polychromy of the Dyste
Madonna.
The base was golden; genuine gold leaf with a yellow glaze
on top. There may have been imitation jewels along the
bent fore of the base, in the same manner as elsewhere
on the sculpture. A floriated design may have been painted
on, or punched, into the golden surface along the vertical
lower part of the base. Her shoes may have been golden,
decorated with a geometrical cross pattern. Her eyes may
have been blue 20.
It is likely that the child wore a golden tunic, with
the same red lines along the edges as the mother has.
He presumably held a Bible in his left hand and wore a
crown on his head. These may both have been golden and
set with »jewels«. The inside of the crown had
a red lining and along the edges of the outside red lines
were running. One may assume that also his hair was golden.
Before the painting proper could start, the gilding and
silvering were carried out: First silver leaf was applied
to the side of the throne, inner side of sleeves and outer
side of cloak.
Then the gold leaf was applied to the cloak, tunic and
crown. (Shoes, base, child.) The silver and gold areas
may have been polished at this stage to enhance the brilliance
of the metal surfaces. When this was done, a yellow glaze
was applied on top of the metal surfaces (not the cloak).
After the glaze had dried, the imitation jewels were painted
in with green and red glaze. At last, when all the golden
areas were dry, a transparent varnish was painted in on
top. This varnish was only applied in the golden areas
(except for the hair).
The areas to be mordant gilded or painted first received
an oil/resinous insulation layer. The yellowish mordant
layer of the hair was painted, and when almost dry, the
gold leaf was applied. No varnish or glaze was painted
on top of the gold in this area.
The flesh areas are painted »wet in wet« (drying
oil); the rose-coloured areas of the cheeks are blended
into the white areas. After the flesh paint had dried,
the eyes, eyebrows and mouth were painted in. Finally,
all the other red areas were painted; throne sides, inside
of crown and sleeves and the decorative lines along the
edges of the crown and sleeves.
1. Definition of »dampfold«:
»A
style of representing draperies as if they were wet and
clinging to the forms beneath«.
G. Zarnecki, J. Holt, T. Holland: English Romanesque Art,
1066–1200, London 1984, p.
414.
2. S. Jónsdóttir: Die Saga einer thronenden Madonna,
Reikjavik, Reikjavik, 1964.
3. K. Endemann: Das Marienbild von Werl, Westfalen
Hefte fur Geschichte, Kunst und Volkskunde, Westfalen,
1975, pp. 53–81.
4. Some of the literature which describes and discusses
this Madonna type:
-A. Andersson: Medieval Wooden Sculpture in Sweden, vol.
II. (Romanesque and Gothic Sculpture), Stockholm, 1966.
-A. Andersson: Mariabilden i skulptur, 1150 – 1450,
Stockholm, 1957.
– C. R. af Ugglas: Gotlands Medeltida Traeskulptur til
och med hóggotikens inbrott, Stockholm, 1915.
– S. Jónsdóttir: (see note nr. 2).
– K. Endemann: (see note nr. 3).
– H. Fett: Billedhuggerkunsten i Norge under Sverreaetten,
Kristiania, 1908.
– P. Tángeberg: Maleri pä romanske träskulpturer, Den
ljusa Medeltiden, Stockholm, 19..
5. C. R. af Ugglas, op. cit. (nr. 4), pp. 99–191.
6. ibid. p. 189.
7. The Swedish Madonna from Appuna, of the same
group, has still remains of her shrine intact.
8. Martin Blindhcim: Main Trends of East-Norwegian
Wooden Figure Sculpture in the Second Half of the Thirteenth
Cent., pp. 95, 96. P. Tängeberg: Mittelalterliche Holzskulpturen
und Altarschreine in Schweden: Studien zu Form, Material
und Technik, Stockholm 1986. (Holzskulptur und Altarschreine,
Miinchen 1989).
9. Since it has been proven that linen was utilized
in Norway already in the early centuries A. D. and wool
can be ruled out it is obvious that this must be linen
fibre. (Characteristic X-shaped joints visible under the
microscope.)
10. The stains are brownish and water-soluble.
Insulation of the wooden surface with animal glue was
a common procedure in this period: See: Alfraedi
Izlensk: transl. into English by C. Sanders: Arbók Hins
Islenzka Fornleifafélags, Reikjavik, 1974.
11. The chalk was analyzed (coccolith-analysis),
and seems to belong in a zone from Denmark to the
British Channel. (Upper Campania.) U. Plahter claims that
this is not a common type of chalk for other Norwegian
polychrome sculptures of this period. Description of how
to apply the chalk/glue ground: see note 10, Alfraedi
Izlensk.
12. A. Izlensk writes that new glue must be applied
before the metal leaf is put on (note 10). This is also
my own experience through practical experimentation.
13. U. Plahter has earlier found proteins which
may originate from egg in glazes on medieval Norwegian
works of art. L. Plahter, E. Skaug, U. Plahter: Gothic
Painted Altar Frontals from the Church of Tingelstad,
Oslo, 1979.
14. Fluorescent characteristics of linseed oil:
R. de la Rie: Fluorescence of Paint and Varnish Layers,
Part II, Studies in Conservation 27, 1982, pp. 65, 66.
U. Plahter states that droplet formations in the glaze
is a typical phenomenon of glazes containing natural
resin.
15. The binding media test was executed by U. Plahter.
About the fluorescence of glair: U. Plahter, op. cit.
(nr. 17) p. 22.
16. Alfraedi Izlensk, op. cit. (nr. 10). A description
of the process of »imitation«
gold is given.
17. S. A. Wiik, U. Plahter: St. Olav fra Fresvik
og St. Paulus fra Gausdal. To polykrome treskulpturer
fra middelalderen, Universitetets Oldsaksamlings ärbok,
1979, p. 222.
18. Concerning yellow glaze on gold: See U. Plahter:
The Crucifix from Hemse, Analysis of the Painting Technique,
Maltechnik/Restauro, 1984, p. 43.
19. See especially the Appuna and Hammar Madonnas
at the Historical Museum, Stockholm, Sweden. See alto
the literature list in note No. 5: A–G.
20. The Appuna Madonna has blue eyes.
author
Marianne Selsjord (Norway), AHVT B 040
|
|
