The 1930s cleaning of the Parthenon Sculptures in the British Museum

Ian Jenkins

The surface condition of the Parthenon Sculptures

The unreliability of documentary evidence leaves the sculptures of the Parthenon as their own best witness for the consequences of the cleaning of the 1930s. As was, however, demonstrated at the conference, even this material evidence is subject to differing interpretations.

i. Method and Terminology

Use of photographs

There was dissent at the conference over the extent to which the condition of the sculpture before-and-after cleaning can be inferred from old black-and-white photographs. There are two principal photographic sources: those published by A. H. Smith in his monumental book, The Sculptures of the Parthenon (1910). The other is the photographs by Frédéric Boissonnas and W. A. Mansell for Maxime Collignon's publication, Le Parthénon (1912). It is right to be cautious about the use of such images, but it is also my firm conviction that if the right questions are asked of them, they can be a valuable and indispensable resource.(121)

Use of Casts

These are most effective when documenting substantial losses, such as the well known instance of damage occurring through continued weathering of the west frieze and other sculptures left on the Parthenon after Lord Elgin's time.(122) Again, there is the loss of a fragment of drapery from South Metope III mentioned at the conference by A. Mantis(123) and the flake from the rump of one of the horses of west frieze block II observed by St Clair.(124) More problematic is the attempt to use casts as a record of the condition of previous surfaces. While casts can record a given surface remarkably accurately, they can also replicate the plastic form of sculpture while at the same time presenting an altered or entirely new surface.

Even in the former case, where the surface is faithfully reproduced, reading a cast is problematic. St Clair claims to see evidence of the removal of tool marks in a comparison of a nineteenth century cast with the original of the shoulders of the Helios of the east pediment. When the two were brought together in the Duveen Gallery, however, his claim was shown to be unfounded. The marks of the claw-chisel indicated on the cast are preserved in the sculpture.(125)

Original Surface

There was also discussion at the conference of the terminology used to describe the surface condition of the sculptures of the Parthenon, both those in London and in Athens. One difficulty has to do with the notion of an original surface. The term was used at the conference with no strict definition of what this actually meant. As we see them displayed in the Duveen Gallery of the British Museum, the sculptures are the product of their history in the Museum and in other previous lives. The variety of surfaces that result are in a sense all original, but what seems mostly to be meant by the term is 'the nearest we come to the ancient finished surface as left by the carver's final touches'.


This term has been used to mean more than one thing. I shall not here enter into a lengthy discussion of its origin and use.(126) Suffice it to say that I have avoided the word altogether in preference for coating, by which I have in mind the orange-brown coatings which were discussed at length in an article with A. Middleton.(127)

ii. Types of surface encountered in the sculptures - an overview.

Natural weathering

It is important at the outset to set the whole issue into proper perspective by observing first that the sculptures as they now are and in pre-cleaning photographs owe the greater part of their surface to natural weathering. For example, in many parts of the north frieze blocks showing the cavalcade, large areas of pitting framed within discernible contours are the result of attack by biological growth on the permanently shaded north side of the temple.(Fig.4) This natural damage is repeated over much of the north frieze, both in London and in Athens. It accounts for substantial loss of the original surface.(128

Another sort of weathering is the visible impact of wind and the rain, ice and dust which the wind drives against the stone. Parallel lines in the marble represent the contours of the geological bed of the stone. They are differentially weathered so that softer elements are stripped out leaving the harder standing proud. Where this weathering occurs, even the higher surface is always below that of that which was originally carved. A good example of it is to be found in the distinctive markings of the head of Selene's horse from the east pediment.(Fig. 5)

Finishing processes

Where neither natural weathering nor cleaning has removed it, the original surface as defined above survives. Although never extensive, and now less than it was, nevertheless enough survives of the original surface to show that the carvers did not work the stone to one uniform finish. Instead, in one relief or, in the case of the pediment sculptures, in a single figure, a variety of finishing processes is found.

At one extreme there are the high polishes of some parts of Helios from the east pediment and Ilissus from the west;(129) (Fig.6)

then, more common, there are the smoothed, but unpolished surfaces;

the parallel lines left by the rasp;(Fig.7)

the coarser result of the claw chisel;(Fig. 8)

and in some places, the coarsest of all, is the heavy pitting of the point.(Fig. 9)

Over this combination of worked surfaces are laid the effects of two and a quarter millennia of weathering, and on top of this again the effects of nearly two hundred years of Museum cleaning.

The nature of the 1930s cleaning

Within the 1930s cleaning, even in the case of a single piece, there is not just one treatment, but rather a variety of different treatments can be found all on the same stone. Interference with the surface of the marble in the 1930s, whatever we think of its absolute merits was by and large carefully and thoughtfully done. It was not the aggressive violation that is implied in the documents of the Board of Enquiry. In particular, Plenderleith's deposition to the Board suggesting the chipping of the marble and the rubbing of it down to a depth of one tenth of an inch does not represent the actual situation. His account was perhaps more an expression of his professional frustration than a reflection of what actually was done.

Nor does the colourful language of Cesare Brandi, writing in 1950 reflect the true situation. His talk of a 'ferocious and irreverent scouring' implying an extensive removal of original surface is unacceptable. Claims that iron tools were used to scratch the marble and that a point was used to outline the figures against their background are insupportable. The tools are today preserved in the Museum's Department of Conservation. There are no iron ones among them, nor do official documents speak of such.(Fig. 10)

iii. Types of surface - a detailed analysis

Surfaces unaffected by the cleaning of the 1930s.

Unaffected surfaces, both weathered and tooled in antiquity, can be found in the great majority of pediment sculptures and in the east frieze. There is also the tray-bearer of the north frieze. The cavalcade of the north frieze is affected only in very particular places. South Metope XXVII is also virtually untouched.

Weathered surfaces affected by the cleaning of the 1930s.

There are subcategories here:

a) There is the buffing of the biologically-attacked surfaces of the horses of one of the north frieze chariots.(Fig. 11) There is a slight blunting of the pitting and a blurring of its contour edges.

b) Different again is the rubbing of surfaces destroyed by wind and rain. In order to understand what was done here, first let us again identify a weathered surface that is not affected by rubbing. On the thighs of the Lapith of Metope XXVII can be seen a clear line demarcating two surfaces.(Fig. 12) On one side is the weathered and unrubbed surface; this is stratigraphically lower than the smooth surface on the other side. This smooth surface has coating and, although some of this may have been removed in the 1930s or earlier, its smoothness is not the result of over cleaning. It is very close to the surface as left by the ancient carver. The line, between these two surfaces - preserved and weathered - marks the edge of the rain-and wind-shadow and is entirely natural. There is very little, if any rubbing of the weathered surface on this stone. Not even the background is seriously affected, as in the case of other metopes.

On SM XXIX by contrast the weathered surface is rubbed, mostly on the background, but the rubbing is also carried over onto the figure of the girl. The weathering of the drapery of the girl appears smooth, milky and blunted.(Fig. 13)

The rubbing of weathered surface also occurs in the lower drapery, front and back of figure G of the east pediment. This is perhaps the best of all examples, for here a clear boundary can be seen in the difference between the unrubbed drapery above and the rubbed drapery below.(Fig. 14) There was very limited removal of coating here, since the original surface survived only in the sheltered folds of the drapery. The difference between rubbed and unrubbed is slight and there is no discernible drop in level such as is suggested by Plenderleith's report.

c) Another type of surface is that of the back of the head of Selene's horse. Old pre-clean photographs show very little patina, some biological attack and an otherwise relatively smooth surface. (Fig. 15) This was made smoother by the rubbing of the 1930s. Suggestions that the front of the head is spoiled by over cleaning are of course indicative of how little this problem has been understood. The back is, however, rubbed.(Fig. 16)

d) A fourth type of weathered surface is to be found in the Ionic frieze. Partially sheltered for so long by the coffering of the peristyle, this is less destroyed than the surface of the more exposed metopes, but nonetheless the original finish has been weathered away. How is it certain that the original surface was not scraped off in the 1930s? If it was scraped off then, signs of the coating would show up in the pre-clean photographs. They do, but only in isolated areas that are the exceptions proving the rule. Where sculpture, such as the east frieze, was unaffected by the 1930s cleaning we see a surface in the photographs that is remarkably similar to that which we see today. Where the surface was affected, as in the north, south and west friezes, the sculpture is unnaturally smooth, and the sugary texture of the marble has a milky finish. The effect is concentrated more on the backgrounds than on the figures. There is no very great loss of surface, but where it does occur, there is some blunting of detail. There is not, however, any loss of detail such as veins in flesh or folds of drapery.

It may be wrong to presuppose that such blunting, the creation of an artificial sfumato, must necessarily all be the result of the 1930s cleaning. One area where it occurs is the heads of the horses of one of the north frieze chariots, Block XXIV. It is certain that this was polished in the 1930s, but take out the dirt that shows in Boissonnas' photograph, and would the image be very different?(Fig. 17)

It is also important to take into account the fact that the polyethelene glycol wax put on to protect the sculptures after they were last cleaned is contributing to this sfumato by closing the pores of the marble and making it appear smoother than it actually is. This protective coat, commonly known as carbowax, has served the sculptures very well in keeping out the dirt and preventing the need for another clean since 1970. It may, however, be suggesting that the smoothing is more than it actually is.(Fig. 18 - Fig. 19)

That said, substantial parts of the frieze, the metopes and much less of the pediment sculptures did have their weathered areas rubbed smooth by the 1930s cleaning, and this is a change between the way they were before and how they appear now.

Coated surfaces affected by the cleaning of the 1930s.

In the 1930s the coating was more extensive over the south frieze than ever it was on the north. There never was, however, on either frieze anything like the survival of coating that St Clair has claimed was on the sculptures when they came to the Museum in 1816.(130) The simple proof is the lack of coating on frieze blocks in Athens. Not only is the north side of the frieze susceptible to biological attack, but also to erosion by wind. I am assured by my colleagues in Athens that the run of blocks from the north frieze displayed in the Acropolis Museum have never had their coatings artificially removed. The only explanation therefore for the absence of coating on the Acropolis Museum blocks is that, it was never there when the blocks were found. Only some joining fragments and Block X of the north frieze have coating. Block X has a different history from the rest having been removed from the building to make a window at the time of the conversion of the temple into a church.(131) Its different surface can be explained therefore by its different fortune. All other blocks are denuded of their coating, we must suppose by natural causes. If that is true of the Acropolis sculptures, then it follows that it must also be true of the British Museum's blocks from the north frieze. In the Boissonnas photograph, the few traces of the coating that remain can be readily picked out, notably that around the calf of the boy on North Frieze XLVII. This coating was largely removed in the 1930s.(Fig. 20)

There are two separate categories for the removal of coating in the 1930s:

a) coating removed and resulting surface unpolished

b) coating removed and resulting surface polished

An instance of category (a), coating removed and surface unpolished can be found in the now speckled cloak of the marshal who appears on West Frieze Block I.(Fig. 21) This retained its dark coating up until the 1930s, as early photographs show.(Fig. 22) In the 1930s this coating was largely removed, leaving traces on the background and on the figure. This speckled effect is very reminiscent of the recently publicised cleaning of the Hephaesteum temple frieze in the Athenian Agora, with steel chisels and wire brushes, in the 1950s.(Fig. 23-Fig. 24) The precise methods of this cleaning are well documented.(132) Homer Thompson, Field Director of Agora Excavations, explained how, using steel chisels and brass brushes, the Greek workmen stripped the coatings, leaving traces only where they were sheltered in the recesses of the ancient tooling.(Fig. 25) 'In these areas', writes Thompson, 'no effort was made to remove the last particles of deposit from the miniature depressions, particularly since it soon appeared that the thoroughly cleaned figures stood out more effectively against a slightly off-white ground.' The resulting cameo effect was curious even for the time.

An instance of (b), coating removed and resulting surface polished, can be found on a Parthenon north frieze block in the British Museum where the remains of coating were removed in the 1930s in one localised area.(Fig. 26) The abrasion was, however, restricted to this one place and not carried over the body of the whole block. This is an exception that proves the rule that the cavalcade of the north frieze is largely unaffected by the cleaning.

The south frieze cavalcade had retained more of its coating and consequently was more affected than that of the north. On a block of the south frieze before and after (Fig. 27-Fig. 28) cleaning the coating lies extensively over the marble, streaked by the erosion of natural weathering. The surface today is speckled with coating and it was not polished. There is, however, extensive polishing of the background of many blocks of the south frieze, both previously coated and uncoated. The whole of the background of South Frieze block XIX showing horsemen is extensively rubbed, both where coating remained and where it did not.(Fig. 29)

The coating is removed with varying degrees of care. On the whole the standard of care is high. This was not the violent scrub it is often portrayed as. Little of the marble, if any, is removed in the process. The situation is not, however, always the same. In one inconspicuous place on a metope we find coating removed with an indiscernible interference with the marble stratum.(Fig. 30) In another, where less care has been applied, there is visible scratching or gouging of the surface.(Fig. 31) Happily, this is by far the exception rather than the rule.

It is, of course, now considered dangerous ever to remove a historic surface, no matter how disfiguring from the body of an ancient sculpture. One fragment of the Parthenon Frieze in the Acropolis Museum has had its coating removed and the detail is unnaturally blurred. This is the fragment of Block V of the East Frieze showing the head of Iris.(133) It used to be shown in the Acropolis Museum with a cast of the remainder of the block preserved in the British Museum. It was afterwards separated from the cast and is now shown as an isolated fragment.(134) It was probably then that the surface was stripped back to the raw marble.(Fig. 32-Fig. 33) The Hephaesteum frieze is a more serious case. The removal of coating from sculptures in a protected, museum environment has a limited effect. In the case of sculpture exposed to weathering, removal of the coating appears to speed up the continued erosion of surface. Already by 1974, when at the behest of J. Dörig, the east frieze of the Hephaesteum was cleaned and photographed again, all traces of the colours reported in the 1950s had disappeared.


I shall sum up by answering three major questions:

1. How much of the total area of the sculptures was affected by the cleaning of the 1930s?

It should now be evident why the answer to this question is not so straightforward as it might seem. Since percentages have been published already, an estimate at least should be given here. Taking together Helios, the backs of the heads of his horses, part of figure G and the back of the head of the Horse of Selene we arrive at a figure of some 10% of the total east pediment. Of the frieze, the east was not touched at all and I am going to exclude the cavalcade of the north, because it is so little affected and it would be misleading to include it. The total area then, covering the two blocks of the west frieze, the chariot sequence of the north frieze and most of the south amounts to about 40% . The metopes are more affected, but not all to the same degree and more on the backgrounds than on the figures and I would estimate their figure at around 60%.

2. Are the sculptures ruined?

Some people have said that, with their disfiguring coatings removed, the sculptures look better than ever.(135) Certainly, to judge from photographs, they look better now after their 1930s cleaning and the further cleaning of 1969-70, than they did for much of the nineteenth century and the early decades of the twentieth. Much of this nineteenth-century dirt still marred the sculptures in the 1930s. Indeed, even now some blocks of the frieze have a different colour because they retain carbon from pollution. These are the most weathered blocks of the north and south friezes that were not affected by the 1930s clean. There can be no doubt that some part of the outrage expressed at the 1930s cleaning had to do with the fact that people had become used to seeing the sculptures in their dirty state.(136)

Setting these considerations on one side, the fact that the cleaning was unauthorised was a scandal; the way the Museum tried and failed to cover it up, was a scandal, but - was what actually was done a scandal? One can understand why Duveen, and perhaps Forsdyke tacitly, wanted to clean the sculpture the way they did. Their motives were no worse than those that inspired the cleaning of the Hephaesteum. Alison Frantz wrote innocently of her joy at seeing the sculptures after cleaning 'to be as fresh as the day they were made.' In the 1950s, as in the 1930s, the importance of preserving original surface was not the priority it is today. 'The Greek archaeological authorities', wrote Thompson, 'inspected the work both during its progress and after; they expressed themselves as well pleased - thank goodness.'(137) Now, however, stripped of its historic surfaces, coated in grime and discoloured by pollution. the frieze has lost its charm.

3. Should they have done it?

Once again, laying aside the scandal, the short answer to that question is, no. It cannot, however, be an absolute judgement. It is an archaeological one. Archaeologists are greedy for information. We want it all there, with nothing taken away, no matter how disfiguring. Since, more often than not, we cannot have it all, we have invariably to resort to reconstructing it. What we do not want to be told is that evidence was there once, but was taken off because it didn't look very nice. But this is only one point of view and since the beginning of archaeology, others have seen it differently. No shame or scandal should attach to such a view when it is expressed openly. The scandal of the British Museum's cleaning of the Elgin Marbles 60 years ago is not what they did, but the way they did it.

St Clair has a different view. He suggests the historical responsibility of the British Museum is to preserve the Elgin Marbles in the condition they were in when they first came to the Museum, and that anything short of that is a failure. If that were the case, then everybody throughout the world charged with responsibility for maintaining the material culture of the past, has failed. All museum curators know that objects in their care change as part of their museum life. Some museums have better archives than others, and the British Museum's history is better documented than most. In Archaeologists and Aesthetes I charted the history of the sculpture collections of the British Museum and wrote extensively of the many changes that the sculptures of the Parthenon underwent, including those occasioned by the foulness of the London atmosphere. No reviewer attached scandal to that account. No one attempted to condemn the present by the mistakes of the past. Events in the past, over which the present can have no control, were seen properly as what they are, history.

The controversial cleaning of the Elgin Marbles happened 60 years ago. All those involved are dead. The British Museum does not defend their mistakes, nor claim a right to a record of impeccable curatorship. No museum could. The British Museum is not infallible; its history is pretty much a formula for the human condition itself, a series of good intentions marred by the occasional mistake. The 1930s cleaning was such a mistake. That event colours, but does not change, the overall responsibility of the Museum's commitment to safeguarding the Marbles. The historical justification of the Museum's claim to safeguard them has not been the 'cynical sham' of infallibility that Clair attributes to it,(138) but the simple fact that if the sculptures had not come to the Museum when they did, they would not survive as they do. This is not an opinion. - it is a fact. In an ideal world Elgin's men would not have damaged the Parthenon in removing the sculptures, but in an ideal world the Athenians of the fifth century AD would not have attacked the metopes with chisels; Morosini in the 17th century would not have aimed his cannons at the temple, and in this century Nicholas Balanos would never have been put in charge of its restoration, nor would modern pollution have destroyed its surface.

The cleaning of the Elgin Marbles in the 1930s was an unfortunate incident of another generation and another age. The tragedy of the present generation has been to witness the progressive deterioration of the sculptures that have been left until recently on buildings in Athens, while some are still exposed. The continued deterioration of the west frieze still on the building until 1993, and the spoiling of all the Acropolis sculptures exposed to acid rain until the recent removal of some, but not all, to the shelter of the Acropolis Museum, is the tragedy of our time.

Writing in The Times newspaper,(139) the editor Peter Stothard called for an end to the 'Chisellers' War'. He sees the setting out of a full report of what happened at the British Museum in the 1930s as a precondition of such peace and demands transparency in the process. I would agree and only add that if transparency is desirable for one, it is so for all. And, as we look out from our glass-houses into those of our neighbours, who will dare to cast the first stonek