|
OBJECTIVE METHODOLOGIES |
Veritus AG |
|
OBJECTIVE METHODOLOGIES |
| This section details the
use of technologies that can be used to determine attributions. You will note that with
only one exception, the Digital Brushstroke Analysis, none will provide a
conclusive result. Therefore, although the technology may provide a definitive result, the
answer can still prove inconclusive. Twenty-first century technology allows us to make a giant stride towards accuracy in painting attribution. The proprietary software used by Veritus provides a visible horizon for evaluation of attributions. The technology eliminates unwarranted bias, human errors and other problems that have plagued art attributions over the centuries. Computers make dispassionate digital decisions based upon facts not subjective opinions. Computers are not influenced by social pressures nor by any vested interests. Computers do not own paintings that they would like to see attributed to a certain artists. The person who is paying for the research can not place a computer under pressure to make different decisions. In this fashion, computers become the perfect arbitrator of facts and administrator of judgements. The decision was made to adopt the philosophy of conservatism. It was necessary for the painting to establish beyond doubt that it was the work of the designated artist. Nothing was attributed to any artist unless the data provided conclusively it was by the hand of that artist. |
|
Panel and Canvas Dating |
|
| "...very scientifically interesting but have no value..." | There has been a growing trend in examining panels to determine
their date and origins. Similar research into canvas fibre examination and thread count
was also popular. However, the master and his students generally purchased from the same
sources. Paintings sometimes appear on panels that are of strange or unexpected woods as
artists or craftsmen from other areas travelled widely. Famous forgers of the last century purchased old paintings of low value and then repainted over the surface. Paintings were sometimes transferred to canvas or onto a panel long after their original execution. Studies involving panels and canvas are very scientifically interesting and have exposed recent forgeries, but have no value in determining painting authorship. |
Paint Chemical Analysis |
|
| This technology has proven quite important in the determination of
later forgeries. It also allows for a more accurate reading of x-ray technologies. However, since most of the artists in an area purchased their paint from the same sources, the analysis does not provide any definitive information about authorship. |
|
Paint Build-up Analysis |
|
| The
Committees have used the digital analysis to evaluate paint build-up and other qualities.
The results of this analysis are far from conclusive. However, data gleaned may someday be
very useful to restaurateurs. |
|
Digital Colour Palette Analysis |
|
| "By the time experts gain the training and experience to make scholarly evaluation, they often no longer have the physical ability to do so with great accuracy." | This
technology has proven quite promising as a supportive technology in the determination of
authorship. It is a well established fact that each person perceives colours in a slightly
different fashion. That is why one person will say that a colour is greenish blue while
another will argue that the colour is bluish green. Colour palettes are often strong
indicators in authorship. As an example, El Greco during his Italian period painted
several very fine copies of art works that he studied. These art works are often
recognised by their palette colours. Colour usage has long been an important tool of the
expert in determining the attribution. An exceptionally gifted expert can identify several thousand different colours. Unfortunately, human perception and abilities diminish with age. This truism has historically created a difficult problem when it comes to attributing works of art. By the time experts gain the training and experience to make scholarly evaluation, they often no longer have the physical ability to do so with great accuracy. Alternatively, our computer analysis programme is capable of identifying several million colours with complete accuracy. Varnish colours and age-induced defects are digitally subtracted to see the original colour palette of the artist. This standard is more than two thousand times more accurate than the very best of human eyes. This colour analysis programme makes a careful evaluation of the exact colours used in various paintings, particularly those pigments that are not uniformly mixed. This analysis becomes extremely important when determining attributions, as the colour palette used by an artist normally evolves slowly as the artist uses particular paints. Artists also experiment with different colour combinations and these can be tracked through a series of paintings. Consequently, paintings from close to the same period of time often have nearly identical base colour palettes. The application of this technology then, also, becomes not only a valuable tool for determining who might have painted a work but, also, for determining when it was painted. The weakness of this technology is that it is always inconclusive when used alone. Therefore, it becomes a useful tool for the confirmation of other evidence. The limitations of this technology appear at another area within this site. |
| Digital Brushstroke Analysis | |
| "...makes it possible to detect, on occasion, when the same paintbrush was used on more than one painting." | This
technology became our mission critical programme. The technological ability to conduct
this research has existed for many years, however it was never cost effective until
recently. Breakthroughs in the area of higher math, specifically the use of Algorithms has
been able to make this form of research cost effective for the more important artists.
With the advent of modern computers and the use of pattern recognition algorithms, similar
to those used in the human genome project, what used to take months of computer time can
now be done in hours. This method of examination has proven to be often as identifying as a fingerprint. The technology allows for comparisons of key features and nuances consistently appearing in individual brushstrokes. The core of this technology is pattern recognition algorithms. When very large numbers of brushstrokes are compared, definitive patterns emerge. The computer effects these brushstroke comparisons at a rate of millions of comparisons per second. This proprietary technology is roughly similar to an OCR programme, but infinitely more sophisticated. It learns as its data base increases. The sophistication of this technology makes it possible to detect, on occasion, when the same paintbrush was used on more than one painting. It is similar to the forensic examination of a bullet. Some bristles leave distinctive patterns that are reproduced on more than one painting. Although this technology is used primarily to determine who painted the picture, it can, additionally, be employed to provide an indication of relative timing. For example, if the same brush is used to paint two different paintings but there is a marker bristle missing in one of the paintings, that painting would have been painted later. Bristles can be lost from brushes but cannot be replaced. The same technology allows for the examination of palette knife marks; unfortunately, palette knife analysis has proven inconclusive in all cases to date. Thumb smudges have been examined, although this analysis has proven inconclusive with one exception. A later period work was discovered to have a fragment of a thumbprint that almost exactly matched a fragment on an earlier very well documented work. A number of paintings were found to be by the hand of more than two artists. The same hand painted the heads, the clothing was painted by a different artist, the hands were painted by a third artist. Some paintings included the distinctive brush strokes of the master. However they also included the distinctive brushstrokes of other artists and were thus attributed to the "Studio of." The limitations of this technology appear at another area within this site. |
| Results and Implications | |
| "...positively identified a number of paintings that previously were resting under incorrect attributions." | Thus, we were able to use colour palette analysis and digital
brushstroke analysis to work in both directions historically, to provide a stronger base
for attribution. These two technologies have formed a foundation for much of the work
accomplished by the working Committees. These two technologies have positively identified
a large number of paintings that previously were resting under incorrect attributions. To
say that some of these results are surprising would be a gross understatement. Even with state of the art equipment, the process still takes a great deal of computational time to make the comparisons and build the data base. We use information learned from each run to increase the database of experience. This has sometimes allowed paintings with inconclusive results to be later attributed with confidence. Problems occurred with paintings that were damaged and/or heavily over-painted. After many computer runs with our expanding database, we have developed a level of confidence in a larger number of paintings. We readily acknowledge there could be a genuine original paintings that we failed to attribute conclusively due to the lack of enough original surface to provide an authoritative attribution. This becomes a philosophical issue. At what point is there insufficient original remaining to declare the painting an original? Veritus made no solo attributions when there was insufficient original surface to determine that the foreground and background were painted by the same person. Veritus chose to take a different direction from those who would rather see the errors on the more generous attribution side. For this, Veritus makes no apology. As we re-examine inconclusive works and other technologies improve, there could be some later additions to the catalogue. |
The Future |
|
| "The Internet, as we know it today, did not yet exist." | The
difficulty for many is to understand both the limitations and
possibilities of twenty-first century technology. Certainly, this technology
provides much accurate information, but this technology is not an end in itself. It is
still, impossible for a computer to evaluate artistic genius. It is possible, even
probable, that the day will come when a computer will be developed to act as a human and
evaluate the realm of the aesthetic. However, this day has not yet come to pass nor will
it be realised in the near future. Therefore, Veritus asserts that the role of art
historians, experts and critics, both of yesterday and today, remains vital in determining
which Old Master paintings are the most inspired and the greatest of masterpieces. The
research and conclusions of these individuals are essential to the continuing growth of
knowledge about artists of the sixteenth, seventeenth and eighteenth centuries. With the conception of this project, there was no intent on the part of Veritus to produce its committee findings on the Internet. The Internet, as we know it today, did not yet exist. Yet, as the Internet has developed, Veritus and its committees, have grown and changed with the technology of the day. With this growth and change have come the awareness that this Catalogue Raisonné must reflect the unlimited ability to record information that twenty-first century technology provides. Therefore, inclusions in these Catalogue Raisonnés will be regularly updated as each Committees’ research and findings are broadened and enhanced. Veritus looks forward to the changes and the contributions that the new technologies of the forthcoming century will make to historical art scholarship. The various committees are committed to providing an ongoing analysis of the paintings attributed to the masters they study, supported by technology rather than human opinion. Veritus is challenged to be on the cutting-edge of this technology and is committed to providing information that is both accurate and current. |
 |
© 1995-2008, All rights reserved |
 |
If you have any technical problems with this site, please contact the This site is being constantly updated and expanded, |
