Shakespeare portrait #4
Tom Reedy
Careful analysis finds nothing to dispute authenticity
ANNE McILROY
SCIENCE REPORTER
Saturday, May 12, 2001
They are the forensic scientists of the art world, using X-ray and infrared
technology to peer into a painting's past, analyzing works of art at the
molecular level to determine when and how they were made.
In the case of the painting that may be the only portrait of William
Shakespeare executed when he was alive, scientists at the Canadian
Conservation Institute were able to find no evidence it wasn't done in 1603.
According to the rag paper label affixed to the back of the painting, that's
the year "Shakspere," as the playwright spelled his name, was painted. He
was 39.
"Now it enters into the realm of the historians and so on. We can go so far,
but short of having a videotape showing Shakespeare being painted, I don't
know how we would be able to contribute any more," Ian Wainwright, manager
of the conservation institute's research laboratory, says. His lab was hired
by the painting's owner, who does not want to be identified.
It is easier for the chemists, physicists and other researchers called on to
authenticate old paintings to prove a work of art is a forgery or a
reproduction than to establish it is genuine. If they find a single pigment
that wasn't available in the artist's lifetime, they can make a case for
forgery. The weave of the canvas can show whether it was made by hand or by
industrial looms, which began operating after 1830. Radiocarbon dating can
establish the age of the wood frame.
While scientists can determine the rough age of a painting or if it has been
forged, they usually leave it to connoisseurs or historians to make the
final judgment. Connoisseurs are intimately familiar with the brush strokes
and other techniques of a famous painter, as well as subject matter, and can
make judgments on whether a newly discovered painting fits in with a body of
established work. Historians tend to follow the paper trail, looking at
deeds of sales, wills, letters and other documents in an attempt to verify
the provenance of a painting. In this case, because it is the subject, not
the artist, who is famous, it will be up to the historians to make a final
call.
The scientists provide the raw data for authentication. A work of art has to
pass their sniff test, which in the case of the Shakespeare painting
involved analysis with a half-dozen high-tech tools by three different labs.
Dr. R. P. Beukens at IsoTrace Radiocarbon Laboratory at the University of
Toronto determined that the rag paper label dates from 1475 to 1640.
Dr. Peter Klein of the University of Hamburg is a leader in
dendrochronology, the science of determining the age of wooden objects by
measuring the size of the annual rings from the wood and comparing it to
historical data. He dated the wood at 1597 at the earliest.
In Ottawa, scientists X-rayed the Shakespeare painting. Radiography has been
used to analyze paintings since its discovery in 1897. In the case of a
fake, the X-rays will show a double painting, where the forger gets an old
canvas and puts a new painting on top. X-rays also show the general aging,
cracking and warping of a painting. In this case, there is no evidence of
forgery, or that the painting is more modern than the label claims it is.
Researchers also used a wide range of scientific photography to look at
various features of the painting. These included using fluorescence to see
if the painting had been extensively restored. (It was relatively
untouched.)
An infrared scan was instrumental in showing that a painting known as the
Ashbourne Portrait of Shakespeare was an overpainting of a portrait of the
17th Earl of Oxford.
The Ottawa lab also studied the paint in the Shakespeare portrait. Oil paint
is made up of pigment, which provides colour and a medium, such as linseed
oil.
It also used X-ray spectrometry to detect lead, mercury, cadmium or other
metals in the pigment, looking for historical anomalies.
"If we saw titanium we would be on the alert for a fake, because titanium
white pigments weren't developed until the 20th century. If we found cadmium
in the red areas we would be suspicious because the cadmium family of
pigments were not invented until the 19th century," Mr. Wainwright says.
The researchers took pepper-grain sized samples of pigment to do a more
in-depth analysis of their chemical composition and structure, again
identifying pigments and looking for historical anomalies. They found none.
Family heirloom is put to the test
The owner of the purported Shakespeare portrait has spent most of his
savings on an expensive scientific authentication process. The Canadian
Conservation Institute applied a battery of tests to the painting, analyzing
inorganic material (such as the pigments), organic material (the binding
media used to make the paint), and the botanical elements (the wood it is
painted on and the paper label) of the painting.
General Artwork
X-radiography (like a medical X-ray) was used to look at the structure of
the painting, a process that points out double-paintings and changes not
visible to the naked eye. None were detected. Florescence (which uses
ultraviolet radiation instead of normal light) was used to look for areas of
overpainting. In faked or copied pictures, this shows elements added later
or changed. Nothing was found. Infrared reflectography was used to examine
the underdrawing, usually done in a carbon-based material to see whether it
differed too greatly from the finished piece. There were no signs of later
alteration.
Date
The stratigography of the paint was analyzed using microscopic
cross-sections of the material. If a layer of dust and varnish had appeared
between the paint used for the date and the rest of the painting, it would
be proof that the date had been added later, but results showed it was
painted at the same time.
Paint
X-ray and infrared spectrometry were used to identify the chemical elements
in the paint to ensure they were consistent with the era in which the
portrait was purportedly created. X-ray diffraction was used on tiny paint
particles to observe their crystal structure and compare it to the
structures of other known materials. No temporally anomalous materials were
found.
Label
The paper label attached to the back of the portrait went through
radiocarbon analysis (carbon dating) to reveal it dates between 1475 and
1640.
Wood
The two pieces of oak glued together to form the board on which the portrait
was painted were put through a dendrochronological analysis, where they were
dated using the patterns of annual growth rings in the timber. The test
concluded that the earliest possible date of this painting was 1597.