The "Imiseba method" - Documenting the late Twentieth Dynasty wall paintings in Theban Tomb 65
TT 65, the tomb of Imiseba, Southern Front Wall (inked, color-coded detail).
As to clarify some of the choices we made when utilizing digital technology in our recent epigraphic work, it is necessary to turn back to the documentation material created in the past. Perhaps the best way to appreciate some of the attributes digitalEPIGRAPHY represents at its core is by introducing a one of a kind epigraphic project that took place between 1998-2009 and focused on the faithful representation of the wall paintings in Theban Tomb 65. Before explaining the specific documentation method developed for TT65, we should recall our thoughts about the purpose of epigraphic documentation, as it was established for this very project 20 years ago.
To completely understand the essence that the tomb’s high-quality artworks exhibit required a unique set of skills and knowledge from the epigrapher. This special knowledge became even more important when searching for the most suitable representation of these artworks, especially for creating a system where fine details could be appreciated just as much as the entire composition. Additionally, it had to be acknowledged that any method designed for interpretation would deduct from the original, regardless of the technique applied. The drawing must often simplify the original in order to emphasize the important details of a complex, multilayered decorated surface. However, in certain cases it is undesirable to make this kind of selection, leading to a preference for using photography in order to achieve objectivity. With all options considered, our guiding principles for creating the facsimile copies in TT65 were established as follows:
Preservation - presenting the wall paintings in their original context as realistically and precisely as possible, with the highest level of stylistic accuracy.
Interpretation - synthetizing and visually indicating all the information that the expert can learn about the ancient artwork, implied in an understandable form.
Representation - creating artistic renditions of each scene that are comparable to the original in quality and which themselves can be appreciated as stand-alone works of art.
Integration - showing the relationship between fine details and the whole by treating individual scenes as part of a larger context.
Nina de Garis Davies’ inked tracings of the Southern Front Wall from the 1920’s (left) and tracing the same scenes on transparent paper in 1999 (right – Photo by Tamás A. Bács).
The tomb and its history
The tomb was constructed on the northeastern face of the Sheikh Abd el Qurna hill on the Theban west bank, and was originally intended for the high official Nebamun during the reign of Hatshepsut. The artists decorated the tomb with typical Eighteenth Dynasty scenes. Based on our most recent research, it seems certain that the walls in the transverse hall were all prepared for decoration although left unfinished in different stages. While most of the hall's scenes show only traces of preliminary sketches (e.g. grids, kheker-friezes, hieroglyphs and some better-preserved figures), the east walls introduce some raised reliefs of exceptional quality. During the second half of the Twentieth Dynasty, the tomb was selected for reuse by Imiseba, an important member of the Theban clergy, chief of the altar-chamber and temple archives of the estate of Amun. While extending the tomb structure with a longitudinal hall, he restored the damaged parts of the walls, then covered the early reliefs with a layer of gypsum plaster and redecorated the whole tomb with wall paintings. In Late Antiquity TT 65 was merged into the so called "Monastery of Cyriacus", which also led to some Coptic activity on the wall surfaces, as proven by the numerous crosses and graffiti and the defacing of most of the painted figures. As can be seen from this short introduction, today's visitor encounters a relatively complicated scenery on the walls, with a multiplicity of layers of different activities.
Metal vases of the Southern Front Wall decorated with Bes, ducks, griffin, foreigners, horse heads etc. as copied by E. Prisse d’ Avennes in the XIX. Century.
Similarly, the epigraphic activity in the tomb has also been rather complex: it covers almost two hundred years, serving the interest of many scholars and artists beginning with A. Dupuy in 1832 (as an artist of the Hay expedition), and later continuing with Prisse d' Avennes, P. Newberry, H. Winlock, and Nina de Garis Davies. Two of the above-mentioned scholars/artists had a special role in the general understanding of the painted decoration. Prisse d' Avennes made several watercolors in the tomb, the best known are the representations of the "golden vases" from the foreign tribute scene (Southern Front Wall) and the depiction of the sacred statues of Amenhotep I and lahmes Nofertari (Northern Rear Wall). He often "completed" the damaged parts of the wall scenes, presenting his material from an aesthetical point of view rather than aiming for the accurate depiction of the original wall paintings. In comparison, the large format paintings made by Nina de Garis Davies are not only extremely faithful representations of all the damages and erosions of the two scenes she documented (most of the Southern Front Wall and the entire Southern End Wall), but are also essential studies on the tomb’s color scheme.
Documenting the wall paintings
Interestingly, although these early draftspersons kept coming back to copy details of TT65’s extensive wall paintings, none of the tomb’s walls were published in their entirety. One understandable reason must be found in the amount of exquisite details these paintings provide, exhibiting the finest examples of late-Ramesside artisanship. Furthermore, Imiseba's decorative program is not only distinctive in its theme covering mostly temple and royal scenes, a divergence from the typical Ramesside private tomb decoration. Its uniqueness largely originates from the way every single wall was decorated with a single monumental scene depicting a particular event, namely Theban festivals such as the Opet and the New Year's Festival.
Accordingly, as one of the main tasks of our documentation process, each and every wall - the largest of them being 10x4 meters – had to be represented as a standalone drawing. (Present article concentrates on the interpretation of the painted decoration, since it is the most intact and complex of all the visible layers.) The following epigraphic technique, essentially called “Imiseba method”, was initially developed for TT65 and has been used extensively in many additional projects since. When invented in 1998, the method combined cutting edge scanning technology with the advantages of traditional facsimile drawings and the visual effects of freehand color-coding.
Facsimile drawing of the ceiling in the transverse hall was a physically demanding effort (penciling a flock of ducks in the transverse hall – photo by Júlia Schmied 2006).
In accordance with our established epigraphic principles, the most important cornerstone of the “Imiseba method” was to create a facsimile copy of the wall paintings indicating as much information as possible. To achieve this goal, some aspects of other Theban tomb publications (mostly the Archäologische Veröffentlichungen [DAIK] and the Theben series) were considered and compared with the specific requirements demanded by the complexity of TT65. After much consideration, the main criteria of the documentation process were set as follows:
By considering each wall's scenery as an inseparable unit within the composition, every single wall had to be represented as an individual painting. Thus requiring so far unparalleled volumes of documantation in private tomb publications.
Although the publishing size of these volumes were to be reduced considerably, even the smallest details had to remain recognizable as these provide the paintings their unique characteristics.
The representation of color had to be given an emphasized role, as it was the most important ingredient of the original. Therefore, as opposed to the most common epigraphic practice of indicating only painted outlines, a method had to be developed to represent the wall paintings as colored surfaces.
Finally, all the secondary visual information (including damaged areas, preliminary sketches, later and contemporary graffiti etc.) to be registered during the drawing procedure had to be presented in unison with the wall paintings.
Flock of ducks on the ceiling of the transverse hall (detail).
In situ 1:1 facsimile pencil drawing of the same unit, labeled with damage and color information ready to be inked in the studio.
Before drawing in the field started, a reduced (1:50), proportionally accurate wall map was created to mark each individual transparent sheet’s exact spot on the wall. The main organizational units of this map were based on a grid, laid over a rough sketch, providing the necessary labels to identify the sheets in the studio. Taking such sophisticated preparatory steps may seem a bit of an overkill for some, however, we had to prepare for storing hundreds of transparent paper sheets in the forthcoming years and maintain easy access to the tiniest bits of visual details, whenever necessary. It is also worth mentioning, that a typical 4x10 meter wall was copied onto around 150 individual sheets of transparent paper – usually moderate in size for scanning and storage reasons and easier transportability.
Once all preparations were done, an accurate, 1:1 pencil copy was created onto the sheets (transparent paper was replaced in later seasons by more durable matte acetate film) applying a certain amount of overlap along the edges of each sheet. Soft (usually 2-3B) pencil was used over the surface to minimize the artist’s pressure on the wall. Beyond offering the most complete, stroke by stroke representation of the current state of the wall paintings, these pencil facsimiles had a few peculiarities worth mentioning. In accordance with the desire to capture the ancient painters’ style as precisely as possible, each individual brushstroke was outlined on the pencil drawing with a clear indication of the stroke’s taper, orientation etc. Obviously, this extra quality measure put a lot of pressure on the artist, elongating the drawing process by a considerable amount, however creating a firm base for indicating the color of each brush strokes in the future. With that in mind, each pencil drawing received numerous labels regarding the hues used by the ancient artist – a safety measure to help with decoding a colorful wall painting in the studio.
Since color is one of the main characteristic elements of a painted wall, we wanted color representation to have a prominent role in our publication. Naturally, as stated at the beginning of this article, a drawing doesn’t necessarily have to aim for faithful representation of the actual work of art but rather to provide the essence of the epigrapher’s extensive study of the material. Therefore, in TT65, color photography was used as the primary source for representing the myriads of hues appearing on the walls, while the epigraphic documentation presented a simplified color scheme emphasizing the ancient artists’ design choices. We must also keep in mind that it would have been impossible to indicate each and every hue of a specific color that appeared in the tomb, as it varied based on the actual day’s mix in the ancient artist’s bucket and brush. Furthermore, in many cases there was a huge discrepancy between the original hue and its present-day appearance on the walls. Affects of discoloration, aging, various invasive coating etc., all had to be counted with in the documentation process. As will be seen, representing color was one of the most challenging and, at the same time, most rewarding elements of the publication process.
The problem with color in TT65
(1) Not all painting materials are equal, as shown on this detail: the dark blue layer that was supposed to cover up the red preliminary sketches and column sectioning was largely dissolved, exposing the layers underneath.
(2) In certain areas, the walls are covered by some kind of thick black residue, making it impossible to determine the actual color underneath. One peculiarity of this phenomenon is its connection to specific hues, such as dark blue, black and yellow.
(3) The later Coptic occupancy took its toll on the wall paintings as most of the faces got smudged (defaced) by wet fabric, occasionally making it impossible to deduce their original appearance.
(4) In TT65, most of the walls are covered with a thin layer of mud/dust diminishing the once glorious appearance of its colors. This obtrusive coating is relatively recent, as no such traces are shown on the fragments found in earlier excavation debris.
In order to indicate color, the “Imiseba method” developed a system that represents the wall paintings in a way that attributes each basic color (black, dark blue/green, light blue/green, dark red, red, pink, yellow, light yellow and white) with a dotted greyscale pattern. Applying a color code at a later stage of the documentation process allowed the artist to indicate the appropriate hue values even for the thinnest brush strokes, such as vertical text dividers or minuscule hieroglyphs, while having an immediate visual impact on distinguishing between light and dark painted features. Additionally, it is of great importance to remark that the color values determined during the initial documentation process referred to their original (aka. ancient) hues, instead of the tone to be seen on the walls today. For example, dark hues that tend to lose their original values were represented as black or dark blue, if traces of the original could be determined with certainty. Obviously, this decision deeply affected the final appearance of the inked drawings, providing a very different type of visual documentation than a photograph could ever deliver.
Indicating damage, preliminary sketches and later human impacts
A damaged area on a painted wall doesn’t just represent discontinuity of the decorative surface, but often has its own importance as an indispensable data source. The indication or omitting of damaged areas can only be justified on a case-by-case base, however there are a few various treatments worth mentioning here:
With utterly omitting damages on a drawing – which method can be dated back as far as the Davieses – fragmentary line segments become more “readable”, however, with the additional drawback of losing data about the missing portions.
With adding damages to those parts of the drawing where interrupting carved/painted elements (a good example being the Chicago House method), we can have a better understanding of the decorative surface.
- With the thorough inclusion of every bit of damage on the surface, the drawing can easily become overcrowded with visual data (like most of the tomb publications in the Archäologische Veröffentlichungen series).
With the above options in mind and despite stretching the amount of work to be done in situ even further, one of the main characteristic elements of fieldwork in TT65 became the faithful representation of all damage appearing on the walls. According to the method, blank areas outlined by randomized dashed outlines indicated the edges of damaged areas. In some cases, damage caused by pigment falling off the walls provided extra information about missing hieroglyphs, while other instances indicated clear signs of later activity in the tomb (good examples being the defacing of the figures or incised crosses throughout the walls).
In many cases, damage caused by fallen-off pigment provided information about missing text.
Coptic crosses incised within the plastered wall surfaces were only one type of decorative elements not associated with the original paintings. There were numerous charcoal figures sketched onto the walls – another reminder of the tomb’s Coptic occupation, among the many graffiti left by later travelers and scholars working in the tomb (including Prisse d’Avennes’ tape used for fixing his drawings onto the walls). Yet another important layer of data was provided by the preliminary sketches partially hidden underneath the final decoration. By studying and meticulously copying these red (sometimes yellow) sketches, many structural changes could be determined before the scenes received their final layout (occasionally, figures were omitted, postures were changed, or hieroglyphic inscriptions received refinements). Naturally, all this information became part of the pencil drawings, while extensive signage was applied throughout the documentation process to differentiate between preliminary sketches, graffiti, damages and painted lines, providing multiple layers of information for working in the studio.
Studio work and the challenge of inking in 1:1
Creating a 1:1 outline drawing of an entire painted wall in the “studio”, using Rapidograph ink pens (Photo by Klára Horváth 2003).
The in-situ drawing process was designed to produce an entire 10x4 meters wall worth of facsimile sheets each year (in approximately 30 workdays). In a stark contrast with the often rushed and physically trying fieldwork, inking the TT65 material was a much more leisurely affair that was spread out to multiple months between excavation seasons. First, every sheet of transparent paper was taped to the floor of a room large enough to accommodate an ancient Egyptian painting that averaged to 10-12 meters in length with the typical height of 3.5 meters. Finding the exact alignment between sheets was still utterly difficult, despite a few centimeters of overlap drawn between the adjoining pieces. A typical wall painting had to be “built up” in long horizontal segments stretching along the entire length of the wall, but occupying only a certain 90 centimeters of its height at a time. This rule was set based on the maximum paper width scanners could take in 20 years ago. Taping the facsimile pencil drawings to the floor was followed by covering them with a long stripe of blank transparent paper, 90 centimeters in width, also enrolled along the entire length of the decorative surface. Retracing each and every pencil line by Rapidograph ink pens started at the upper left corner of the wall, slowly proceeding towards the bottom, attaching yet another stripe (with a little bit of overlap) for each segment. During this process, various line widths represented the many different layers of information, deliberately introducing dashed and dotted variations to distinguish between preliminary sketches, damaged areas, graffiti and final paintings.
Today, with our computer-oriented mind, it is hard to imagine how much effort was needed to manually create a Rapidograph outline drawing in such a scale, while having stayed consistent to the epigraphic method for more than 10 years. Once the inked line drawing was finished and ready to be taken to the next stage, each 10-meter long segment was scanned by using a roll scanner, and printed on paper reduced to 1:4. Unfortunately, there were no computers available to process/modify the amount of data while being scanned, as file sizes would have been about 1GB, which seemed unimaginably large at the time. However, scanning and immediate printing worked without a hitch, providing a significantly reduced version of the inked line drawings that were finally small enough to be taken over on a studio desk. After each segment was cleaned up (which involved using correcting fluid over any dust particles picked up by the scanner), the separate sheets were manually taped together along the overlapped parts, applying double-sided adhesive. At the end, each wall of TT65 were rebuilt as a fully inked outline drawing, conveniently reduced to a manageable size and ready for the last stage: color-coding.
The color-code system created by freehand for TT65 was essentially a set of dotted textures ranging from dark (black) towards light (white) shades.
Color-coding the outline drawings
Although inking the facsimile drawings in their original size was a huge effort, it wasn’t the most time-consuming part of the process. Before we get to the final stage of the drawing process, we must re-establish a few steps of the procedure, since we must keep in mind that there was no computer technology readily available when TT65 was documented. The main reason behind creating the inked outline drawing in 1:1 instead of reducing it to a more convenient size was to achieve maximum accuracy when reproducing the original brushwork. Working in original size allowed the epigrapher to faithfully mimic the ancient artist’s painting style while keeping a close attention even on the smallest of details. Thus, reducing the outline drawing after being inked created extremely fine results that were identical to the original in every aspect. Additionally, a lucky coincidence helped our work in TT65: when reduced to 25%, the reassembled drawings could be re-scanned in a single unit, conveniently fitting within the 90 centimeters limitations of the scanner. This feature preserved the quality of the drawings and no further assembling was needed to produce our publishing material.
Color-coding already partially applied over the outline drawings of the Southern Back Wall.
The color-coding process served an additional purpose as well: reinforcing the printed Rapidograph outline-drawing by meticulously inspecting and inking over every segment of the drawing. When establishing the color code system, the main organizing principle was to create a system where each basic color is instantly recognizable once the code is memorized. When creating the code, any pattern that contained features that could be easily mistaken for pharaonic decorative elements, such as stipes, circles, triangles etc. were deliberately avoided. At the end, the pattern-system, established for representing each basic color on the walls, contained only dots. These were spread out in various density ranging between white dots on black background to black dots on white background, with the addition of pure black and pure white on each end of the spectrum. Furthermore, the code was established in a way that darker patterns represented darker colors, imitating the greyscale appearance of the wall and aiming to mimick the “feel” of the original paintings despite in black and white. Adding the color-coded patterns to the outline drawing had to be done by freehand, one dot at a time, basically treating each wall as a gigantic coloring book. Applying the code for light blue, for example, in practice meant that the epigrapher had to draw circles over the area and fill in the gaps with the Rapidograph, while being extremely careful with spacing etc. so as not to change the density of the pattern. However, drawing freehand had a huge advantage regarding the “coding” of small details, such as red brush strokes, in giving the artist a certain flexibility tweaking the pattern to fit within the available space. Nonetheless, color-coding a 2.5-meter-long outline drawing (representing a single wall) took months and close to a thousand work hours to finish.
Color-coding challenges and problematic areas
(1) Rapidograph could provide only solid black ink, a huge disadvantage in comparison with our digital documentation toolsets, therefore less prominent features (damages, preliminary sketches, certain costume parts) had to be drawn by using dotted or dashed lines.
(2) In some instances, when it was impossible to determine whether a certain feature (such as a wig) had ever received its final treatment, multiple codes were applied simultaneously, reflecting the actual state of the painting.
(3) In cases where it would have been impossible to create a white pattern within a dark area, it was coded as a solid panel while the pattern was indicated afterwards using correcting fluid.
(4) In areas where pigment was mostly worn off, patterns were represented by outlines and/or negative spaces created by damaged areas, usually enhanced by just enough coding to support the color pattern.
(5) As most of the color-codes were meant to fill solid shapes, the coding system struggled to represent sporadically preserved dark pigment traces. At the end only the most significant areas were indicated.
(6) Representing preliminary sketches that show through dark areas often required the switching of colors, isolating these segments by adding a white background for better visuals.
Eventually, as each wall received its final color-code treatment, the TT65 facsimiles became ready to be digitized for publication. As soon as computer technology was ready for the task, the color-coded drawings went through yet another scanning process. The scanned material was saved as lossless TIFF files for further treatment and thorough clean-up in the studio. More recently, as digital epigraphy evolved exponentially with excellent solutions readily available to take over most steps of the documentation process, the TT65 facsimiles were revived to receive a proper digital treatment. The hundreds of penciled transparent sheet facsimiles, once created in the field, were scanned in high resolution in order to be archived, while the color-coded inked drawings were thoroughly cleaned using Adobe Photoshop. As a result, this unique epigraphic project, manually labored throughout years and drawn entirely freehand without any particular use of digital technology, is now ready to live on in the digital age.
Imiseba Southern Front Wall - detail (Click photo to discover in higher resolution)
Inked, color-coded representation of the Southern End Wall of TT65, the tomb of Imiseba (click to enlarge).
Creating facsimile drawings of the Southern End Wall in the tomb of Imiseba, using pencil over large transparent paper sheets (Photo by Júlia Schmied 2007).
 The Epigraphic Survey. Medinet Habu IX. The Eighteenth Dynasty Temple, Part I: The Inner Sanctuaries, with Translations of Texts, Commentary, and Glossary. United States, Oriental Institute of the University of Chicago 2009. 17, Fig. 3.
 The Epigraphic Survey. Medinet Habu IX. The Eighteenth Dynasty Temple, Part I: The Inner Sanctuaries, with Translations of Texts, Commentary, and Glossary. United States, Oriental Institute of the University of Chicago 2009. Plate 139.
 About the ancient history of the tomb see: T. A. Bács, “First preliminary report on the work of the Hungarian Mission in Thebes in Theban Tomb No 65 (Nebamun/lmiseba)”, In Mitteilungen des Deutschen Archäologischen lnstituts, Abteilung Kairo 54 (1998), 49-64.
 K. Vértes, “Az epigráfia módszere es lehetõségei lmiszeba sírjában”, In Ókor VII/1-2 (2008), 55.
 About the architectural and technical survey of the tomb see: E. Harsányi, Zs. Kurovszky, E. Vadnai and L. Kriston, “Beszámoló egy thébai nemesi sír (TT 65) restaurátori felmérésérõl”, In Mûtárgyvédelem 27 (2000), 7-21.
 T. A. Bács, “Thébai kutatások: a 65. számú sziklasír es környéke”, In Ókor IV/3 (2005), 107-13.
 E. Prisse d' Avennes, Atlas of Egyptian Art (Cairo, 1991), PL.II.85-86; PL.II.88.
 The best representation of the two scenes is to be found in Egyptian Wall Paintings - The Metropolitan Museum of Art's Collection of Facsimiles(New York, 1979), 154-5.
 T. A. Bács, “Art as material for later art: the case ofTheban Tomb 65”, In W.V. Davies (ed.), Colour and Painting in Ancient Egypt (London, 2001), 95-6.
 Some of the published details of drawings made by the author using the “Imiseba method” appear in T. A. Bács “Theban Tomb 65: The Twentieth Dynasty Decoration.” In: Egyptian Archaeology 21 (2002), United States, Egypt Exploration Society, Page 24.; T. A. Bács “A Royal Litany in a Private Context.” In: Mitteilungen des Deutschen Archäologischen Instituts, Abteilung Kairo 60 (2004), Page 3.; K. Vértes “Aesthetics and objectivity – new ways and old tradition in object documentation.” In A. Gulyás and K. Endreffy (eds) Proceedings of the Fourth Central European Conference of Young Egyptologists. Studia Aegyptiaca XVIII, Budapest 2007, 396.; K. Vértes “Az epigráfia módszere és lehetõségei Imiszeba sírjában” In Ókor VII/1-2, Budapest, 57.; G. Schreiber The Mortuary monument of Djehutymes II – Finds from the New Kingdom to the Twenty-Sixth Dynasty. Budapest, Archaeolingua 2008, Plates LIX-LXI. and K. Vértes “Ten Year’s Epigraphy in Theban Tomb 65 – Documentation of the Late Twentieth Dynasty Wall Paintings in the Tomb of Imiseba.” In: Kousoulis, P. and Lazaridis, N. (eds) Proceedings of the Tenth International Congress of Egyptologists, University of the Aegean, Rhodes, 22-29 May 2008. Belgium (2015), Peeters Publishers.
 For the closest parallel of the “lmiseba method”, see K.-J. Seyfried, Das Grab des Paenkhemenu (TT 68) und die Anlage TT 227, Theben 6 (Mainz am Rhein, 1991 ), Beilagen I-IV.
 Using a grid for reduction during the copying process is a common practice since ancient times. See C. Traunecker, “Les techniques d'epigraphie de terrain: principes et pratique” In Assmann, Burkard and Davies (eds.) Problems and Priorities in Egyptian Archaeology (New York, 1987), 279-81.
 Somewhat similar outline treatment appears in N. De Garis Davies, Private Tombs at Thebes. Vol. IV: Scenes from some Theban Tombs (Oxford, 1963), Plates II-V.
 D. Piper, The Joy of Art (London, 1984), 108.
 D. Polz, “Excavation and recording of a Theban Tomb - some remarks on recording methods”, in: Assmann, Burkard and Davies (eds.), Problems and Priorities, 119-140.
 A few examples, among others, are to be found in The Festival Procession of Opet in the Colonnade Hall, Oriental Institute publications 112 (Chicago, 1994).
 E. Dziobek, “Reconstructing the wall scenes in Theban Tomb 63”, Bulletin of the Egyptological Seminar 9 (1987/88), 5-13.
 My sincerest gratitude goes to Tamás A. Bács for letting me develop and carry out the epigraphic program of this remarkable late-Ramesside painted tomb.