Undoing the Past

Changing Attitudes Towards the Restoration of Greek Pots

By: Kyle M. Phillips and Ann H. Ashmead

Originally Published in 1988

View PDF

When we first started work­ing on the publication of The University Museum’s Attic red-figure pottery for the Cor­pus Vasorum Antiquorum, Dr. Roger Edwards, then Curator of the Mediterranean Section, showed us a large, heavily restored vessel of a type called a stamnos by Classical archaeologists. After our initial ex­amination of the pot, it was sent to the conservation department for cleaning and restoration using modern techniques and materials. When next we saw the stamnos it was in a transition stage, stripped of its overpaint, and exposed as an un­dressed pastiche made up of pieces from eight different pots, including an “alien” neck and foot.

Relegated to the storage shelves in The University Museum for half a century, this pot is an excellent ex­ample of certain appalling 19th cen­tury techniques and attitudes towards restoration. In contrast to the present-day desire for ar­chaeological “purity,” the previous restorer labored to create a com­plete pot, with all its figures fleshed out, to he treated as a prized decora­tive object. This article documents a shift in attitude, and hence in res­toration methods, as seen in this one pot—a pot now cleaned of over-painting, relieved of its foreign parts, and rebuilt so that it is worthy of gallery display (Fig. 1).

The First Restoration

To study what the initial restorer had created, we turned to old photographs of the overpainted pot, which showed it with its 19th-century accretions (Figs. 2, 4-5). On Side A (the best-preserved side of the painted vessel), an armed warrior extends a phiale mesomphalos or shallow bowl to receive ritual wine, which he will later pour out in liba­tion. The wine is being poured by a woman at the left who holds a pitcher. At the right is a partially preserved figure, restored as a woman. On Side B, the center is dominated by a standing bearded male who holds a staff; he is flanked by two women.

The restorer had fleshed out the scene, making it appear complete from a normal viewing distance. We suspect that cracks and minor blemishes visible on the early photographs were also painted over at one time, but we have no photographic evidence to prove this. Even in its overpainted state, some specific aspects of the restoration were visible:

Side A: The woman at the left (fig­ure I) is complete. On the central warrior (figure 2) the lower Iegs and feet, the lower portion of his cloak, and part of the shield were added. The restorer even cut in the (miss­ing) compass incised lines on the shield’s border. The figure at the right (figure 3) is the most extensivey restored. Here were added the head, right shoulder, and upper body; on the lower body, the forward drapery and foot were painted in.

Side B: The woman on the left (figure 4) is restored from neck to knee. Her real fingers were ignored, and a new hand was painted farther down the vase. The top of the central man’s (figure 5) head and the rear portion of his body, from his left elbow (thrust outward to the rear) to a point just below his right knee, is entirely filled in. The woman at the right (figure 6) is preserved except for the top of her cloth headdress.

The handle zones (Fig. 5): These are nearly intact: only the leaf tips and tendrils of the upper right pal­mette of the A/B handle needed retouching. The tongue pattern on the shoulder and the base meander were heavily overpainted.

First Publication and a Misleading Drawing

In 1932, Edith HaIl Dohan gave the vase its initial scholarly publication. Since the vase was a loan from a Miss Hegeman, Dohan was faced with a dilemma. Had the vase been owned by The University Museum, she would have removed all ex­traneous restoration before present­ing it, as she stated in a letter to John Beazley, a noted Classical ar­chaeologist and expert on Greek vase painting (12 Dec. 1930; Ar­chives of The University Museum). Her solution was to provide a Museum Journal article (1932) with two fine drawings by M. Louise Baker, illustrating what she con­sidered to be the original painted lines (Fig. 6a,b). Baker’s drawings are an essential complement to the old photographs, and succeed in conveying the style of the artist. Nevertheless, they must be used with caution since Dohan was deceived by the extensive over-painting.

In the light of our knowledge of the cleaned painting, inaccuracies in Baker’s drawings become clear. On Side A, there are errors in the depic­tion of all three figures. Although a segment of the warrior’s shield (lying between 1 and 3 o’clock) is actually restoration, Dohan was misled by the compass drawn lines of the restorer into believing that the entire shield and upper portion of the staff were original. A black palmette frond on an inserted fragment was inter­preted as a female hairdo on the figure (3) at the right; this probably led Baker to interpret the entire fig­ure as female. The folds of this figure’s cloak from knees to head were obscured by extensive over-painting.

On Side B, details in the original painting of the woman at the left (figure 4) were obscured by paint added in restoration, so in Baker’s drawing part of her neck, chin and upper cloak folds are missing; near her knee, a long vertical fold was eliminated but two (overpainted) arcs were added. The fingers of her outstretched hand, held behind the head of the central male (figure 5), had been painted out and of course are not in Baker’s drawing. As for the central male, the restorer covered up his eye, the left-most lines of his shoulder, small drapery folds at his waist, and the long line of internal drapery crossing the body from below the shoulder fold to the lower stomach. This unnecessary obscur­ing of original internal drapery folds by the restorer strikes us as peculiar.

Dohan could see that most of the tongue pattern on the shoulder of the vase was not original. The res­torer had overpainted this area to cover plaster, which had been ap­plied in order to force the curve of the vessel between the shoulder and the neck (the latter supplied from another pot; see Fig. 7a). The under­lying, original tongue pattern was visible only in one or two small seg­ments. (Recently a ray pattern near the base of the world-famous Fran­cois Vase in Florence was dis­covered to be a similar overplaster­ing and painting.)

There are omissions too on Baker’s drawing. On Side A, for ex­ample, she left out the dotted decorative band on the lower skirt of the woman at the left. Perhaps be­cause of the considerable overprint, Baker chose to draw only the central panel with figures, omitting the bands of pattern that provide a frame for the central scenes, as well as a drawing of the pot’s contour. As a result, the drawings produce an impression of flat panel paintings, lacking any indication of how the figures relate to the vase’s curved shape.

The Vase Comes Apart

As stated above, the entire pot surface was recently cleaned of all overpaint, and of the fine plaster which was used to fill out the con­tours of the stamnos. Revealed to us was an intriguing conglomerate. The neck, which Dohan had believed modern, proved instead to be ancient and probably Etruscan rather than Greek, based on its shoulder decoration of applied red tongues (Fig. 7a). The foot was also ancient (Fig. 7b); it is perhaps from a large amphora, or possibly from the same vessel type as the body (a stamnos), or from another form, such as a krater. Just as the curvature at the juncture between shoulder and neck was forced, the profile of the lowerbody was also altered so that it could continue into the alien foot.

The most distressing restoration technique to 20th-century eyes was the use of scraped, sanded, and cut-down fragments from five black-and red-figure pots. These had been used as filler with no regard for their archaeological value, in­serted into gaps where a part of the primary (red-figure) stamnos was missing. The filler sherds include fragments of a small cup, a black-figure stamnos, a red-figure am­phora(?), and two other large pots. These pieces were placed ran­domly, at a diagonal or upside­down. Two decorated sherds show upside-down black palmettes. One, placed as the head of figure 3, con­fused Bohan who interpreted it as a woman’s hair (see above). The

other palmette was substituted for a male bead (figure 5; see Fig. 7a). The same two fragmentary figures had filler sherds in their lower body: a sherd with black-figure legs filled the space where the legs of figure 5 ought to have been, and black-figure drapery was placed where the drapery of figure 3 was missing (see Fig. 7b). One filler sherd has traces of a guilioche (braid) pattern.

The New Restoration

Once we had studied the cleaned pot minus all overpaint and over-plaster, we asked that the pot be taken totally apart in order to docu­ment its component vessels. This has now been accomplished under the skillful hands of conservator Stephen Koob, and the primary stamnos pieces rejoined (Figs. 3, 9, 10). What has emerged is a hand­some vase, the creation of a man known to Classical archaeologists as the Oreithyia Painter. Secondary or filler sherds come from at least seven vases (see box).

The Oreithyia Painter stamnos dates to the 2nd quarter of the 5th century B.C., and takes its place within the period that produced similar restrained monumental works in sculpture, for example, the masterpieces from the Zeus temple at Olympia. This vase is an excellent example of a painter who was con­tent to work with a restricted subject matter because it so perfectly ex­pressed his sentiments. (A stamnos in Edinburgh by the Oreithyia Painter, No. 81.44.24, depicts very similar scenes; see Dohan 1932:64­74, Figs. 6-9; Beazley 1963:496, no. 3). Understated elegance and pathos of the moment were impor­tant to the Oreithyia Painter rather than the lively experimentation of  the late archaic craftsmen (for ex­ample, the work of the Foundry Painter, Fig. 8).

Using the new restoration, we can now give an accurate description of the vase and an interpretation of the scenes portrayed. In the center of Side A, a middle-aged, heavily armed warrior (figure 2) is about to receive wine for a farewell libation from a finely dressed woman (figure 1), either his wife or mother. Behind the warrior stands a draped figure (3) who can be identified as a male (see below).

The two best preserved figures (1 and 2) are drawn neatly and in great detail. The warrior, identified as a mature man rather than a youth be­cause of his dark heard, stands calmly, burdened by elaborate and obviously expensive armor which tells us that he is a man of property. The archaic-looking black lion shield-blazon, small in proportion to the overall size of the shield, is espe­cially detailed, with full bushy Inane, open roaring mouth with tongue and teeth indicated, and high curled tail. This archaic device may indicate that the shield was a family piece, perhaps from the late 6th century or early 5th century B.C. The warrior has not yet finished arming, as his scaled corselet (which he wears over a closely pleated tunic, a chitoniskos) is open, its right flap projecting above his shoulder. His articulated scabbard should hang from a baldric, but none is visible across his chest; he holds his helmet and long spear in his concealed left hand. The helmet (a pilos type) is elaborate. The warrior wears greaves, which mold to the knees and terminate in projections. His knee and thigh muscles are very completely rendered in dilute glaze, best seen in a slanting light. His hair is bound in a fillet.

The warrior holds a shallow bowl (phiale mesomphalos) with central boss, a form frequently made of metal by the Greeks. He carries his cloak bunched over his concealed left arm and waits to make a farewell libation. The painter has captured a calm, monumental scene in this panel. In the warrior’s preparations for departure there is a moving poig­nancy, for this man is in such haste to be off that he makes his farewell libation with his corselet still un­secured. The warrior seems heed­less of his fate, which may be to fight and die.

The woman standing before him is, as one would expect from this period, timeless in her beauty. Her curly hair is rendered by close dots; the tie of her elaborate cloth head­dress or sakkos is also dotted. The broad folds of her cloak imply that it was woven from fine wool. These heavy folds contrast with the finer folds, perhaps linen, of her full-length chiton. The thin, tall handle of the oinochoe she carries suggests that this pitcher was also metal. The woman gestures with open palm.

Unfortunately, the third figure standing quietly behind the warrior cannot be precisely identified. The heavy cloak and fine, long undergar­ment could be worn either by a man or a woman; however, the staff held at a vertical is only appropriate for a male or a goddess. We consider the figure a male, quite possibly the warrior’s father.

Two areas of side A have inscrip­tions: the word kalos, or “beautiful” (reading from right to left), extends between the faces of the woman and the warrior. Two letters of another inscription are preserved above the shield, to the right of the staff. The word may be [kal]os, or a name, reading from right to left.

On Side B a bearded man holding a staff (figure 5) may be returning from or departing on a trip. He faces a woman (figure 6), who greets him with a standard greeting gesture of upraised hand, palm towards him. A second female figure (4) stands to his rear, at the far left. Two of her fingers project behind the man’s shoulder, showing that she is ges­turing to the right.

Many of the refined technical fea­tures of red-figure drawing are present on this pot despite its worn state. On Side A are visible the con­tour strips and delicate relief lines that originally outlined the figures; fortunately those of the faces are well preserved. The preliminary sketch (impressed guide lines laid down by the artist for his original design) is difficult to see because of wear, but the lines are present. Details painted with dilute and thinned black glaze, such as those on the helmet crest and the corselet, are especially fine.

The geometric patterns that sur­round the painted scenes were care­fully executed. Motifs within the handle zone were placed so as to accent the sturdy handles: a single large palmette between handle and base counterbalances the pair of smaller palmettes between handle and neck; each palmette is crisply outlined with relief lines. A tongue pattern accents the shoulder. Below the picture field is an elaborate double interlocking meander alter­nating with filled rectangles; this band is edged above and below by lines. The artist started painting the meander border under the A/B handle where the center rectangle is lacking.


We have documented the restora­tion techniques used in the 19th cen­tury on a single Greek pot. Some of these techniques are totally unac­ceptable to modern archaeologists and conservators. The question then raised is To what extent do these restoration techniques relate to 19th century attitudes?

When the 19th century restorer, whether working for a dealer, a scholar, or a museum, wanted to complete a pot, he frequently used spare sherds for which he had little regard; he had access to a varied supply of Etruscan and Greek frag­ments and used them to build up the pot’s shape or just fill in its gaps. He did not hesitate to mutilate these pieces by cutting them down to size, and by scraping and sanding away entire painted surfaces. Such sherds were common then, but today these same sherds would be respected items of value in collec­tions of ancient pottery. (One such group is in the Ella Riegel Memorial Museum at Bryn Mawr College, which we published in the Corpus Vasorum Antiquorum many years ago).

Subsequently the restorer applied fine plaster—generously, to mask the uneven surface and to smooth over and hide blemishes and frac­tures. Lastly, he wanted a complete scene. To achieve this the entire decoration was repainted; missing figures, limbs, and drapery were completed, and fractures and other defects were then touched up with paint. So thorough were these ef­forts that only the most well preserved sections of the pot showed the original work of the an­cient potter and painter. In contrast, the 20th-century restorer attempts to conserve what actually remains and to present only actual fragments of the pot, incomplete as the result­ing design and shape may be. Frac­tures, gaps, and even minor chips and abrasions are left as they are, documents of the final stage of the artifact’s existence.

Cite This Article

Phillips, Kyle M. and Ashmead, Ann H.. "Undoing the Past." Expedition Magazine 30, no. 2 (July, 1988): -. Accessed February 24, 2024. https://www.penn.museum/sites/expedition/undoing-the-past/

This digitized article is presented here as a historical reference and may not reflect the current views of the Penn Museum.

Report problems and issues to digitalmedia@pennmuseum.org.