Date, Name, and Location

Catalogues dating from the 1960s onwards were studied, since this is when sales of pre-Columbian antiquities at Sotheby’s began. The first occurrence of a Maya antiquity was in the 11 February 1963 London sale (a sole terracotta figure head). During the following decades, the quantity of Maya antiquities increased significantly but decreased following the millennium (see table 1). The 1980s saw the greatest quantity of Maya antiquities, matching Gilgan’s results, and can be considered a peak in sales (the volume has been steadily decreasing since). ^{Footnote 65}

Table 1.

Total number of lots with Maya antiquities and total number of Maya antiquities offered for sale at auction from 1963 to 2016

Prior to being sold in dedicated auctions marketed as pre-Columbian art, antiquities from Latin America have been sold in auctions alongside “tribal art”; “primitive art”; “decorative works of art”; and “American art.” Apart from an isolated occurrence in 1961, “pre-Columbian” appeared in sales titles from 1964 onwards. The first auction titled simply “Pre-Columbian Art” took place on 6 March 1971 in New York. Up until the establishment of the Pre-Columbian Art Department at Sotheby’s in 1980, there were 191 sales including antiquities from Latin America in both the London and New York offices (only 116 included “pre-Columbian” in the sale title [see fig. 2]).

Figure 2.

Number of auctions per year from 1960 to 1979 with “pre-Columbian” included and excluded from sale title.

Throughout the 1980s and 1990s, pre-Columbian art became a stand-alone sale, likely owing to the great quantities of antiquities from Latin America on the market at this time (if the numbers for Maya antiquities is a marker of other pre-Columbian antiquities [see table 1]). Yates suggests that “the creation of a sale for only Pre-Columbian objects signalled that the ancient art of the Americas has discarded the ‘primitive’ moniker and entered the mainstream of western collecting.” ^{Footnote 66} From the year 2000 onwards (when there was a decline in the quantity of Maya antiquities), pre-Columbian art was once again sold alongside other antiquities and is currently sold in conjunction with African and Oceanic art in New York. ^{Footnote 67}

It is interesting to note that when the quantity of Maya antiquities on the market declines, sales catalogues begin to change. Early catalogues are devoid of photographs and have minimal descriptions. When photographs are introduced, they are black and white and only accompany select lots. In the mid-1980s, catalogues began to include color photographs and lengthier descriptions. Lobay describes this change as one from mere “shopping lists,” which catered to a wholesale venue, to catalogues that catered primarily to retail-based operations geared for private buyers. ^{Footnote 68}

Of the 150 sales that included Maya antiquities, 110 were in New York, 39 were in London, and one was in Paris (see table 2). In New York, 3,158 relevant lots were sold with 3,534 artifacts; in London, 87 relevant lots were sold with 100 artifacts. The sole Paris sale included 25 lots consisting of 25 artifacts. This demonstrates that the majority of Maya antiquities are sold through the North American market, which is to be expected based on its proximity to the ancient Maya cultural area. The catalogues also reflect this dominant market, since it is the New York catalogues, rather than the London catalogues, that were first supplemented with photographs and lengthy descriptions of lots.

Table 2.

Number of sales, Maya lots, and Maya antiquities across the New York, London, and Paris auctions from 1963 to 2016

Catalogue Cover

As previously discussed, glossy color catalogues began to appear in the mid-1980s. They advertised an individual lot or group of lots on their covers, thereby drawing increased attention to one or several items in the sale. Of the 150 catalogues that included Maya antiquities, only 17 (11 percent) featured a Maya antiquity on the cover (although nine of the remaining 135 catalogue covers were unavailable to the author, so there are potentially more than 17). The first instance of a Maya antiquity on the cover of a Sotheby’s catalogue was on 6 March 1971, when a black and white photograph of a Jaina figurine advertised the pre-Columbian art sale at the Parke Bernet Galleries in New York. ^{Footnote 69} The lot successfully sold for $1,150 (today’s equivalent of $6,822)—the highest price reached for a Maya antiquity in the sale. Of the 16 remaining sales with a Maya antiquity on the cover, 13 lots sold, two failed to sell, and the sale data was unavailable for the final lot.

Although it is expected that the cover lot will sell because of increased attention and advertising, and the data demonstrate that this is usually the case, it is obviously not guaranteed that all cover lots will sell. Goodman commented in response to an auction on 23 November 1998 that “the market for Pre-Columbian works is still selective at the top, as is demonstrated in the failure of the cover lot to sell.” ^{Footnote 70} Thus, while the catalogue cover offers prime real estate for marketing specific items and advertising unique or important antiquities, it does not guarantee sales. Market trends, economic currents, and legislation likely all influence the success of particular sales. The choice of which lot to advertise on the catalogue cover is no doubt a strategic one, but its success or failure is dependent on the larger issues facing the art market at the time.

Number of Lots and Antiquities

A single sale can consist of several hundred lots, all from one collection or owner or each from different owners and collections. Lots can consist of a single antiquity, several antiquities of the same media, or antiquities of mixed media. Early sales often contained several antiquities in one lot, but recent sales usually contain lots consisting of a single antiquity. Of the 150 catalogues that included Maya antiquities, there were a total of 3,270 Maya lots and 3,659 Maya antiquities (see table 1). Some antiquities have been resold at auction over the years so the results undoubtedly include duplicates. ^{Footnote 71}

Tracking these duplicates is difficult, but the author has been able to identify 36 Maya ceramics that have been resold on at least two occasions (see table 3). The number of additional auctions that these ceramics were sold at total 41. Since these ceramics were in individual lots, subtracting 41 from the total number of Maya lots gives a more accurate number of 3,229 lots. Subtracting this same number from the number of antiquities identified above gives a more accurate total number of 3,618 Maya antiquities. Although this number does not take into consideration duplicates of other types of antiquities, it is clear that over a period of 53 years a considerable number of Maya antiquities have entered the market. Interestingly, Yates recorded 3,677 lots of South American antiquities at Sotheby’s over a 39-year period. ^{Footnote 72} This demonstrates that greater quantities of South American antiquities have been sold at Sotheby’s compared to Maya antiquities, which speaks not only to collecting practices but also to targeted looting of specific antiquities. It may also suggest that there are greater quantities of South American fakes and forgeries.

Table 3.

Maya ceramics resold at Sotheby’s

Note: The auctions and lot numbers are organized chronologically.

As previously discussed, the 1980s saw the greatest quantity of Maya antiquities offered for sale at auction, as verified by the total number of lots and antiquities (table 1). Lobay has noticed that the decades following important legislation show signs of heightened auction sales, which he suggests was due to collectors off-loading items at auction to avoid legal issues. ^{Footnote 73} The data for Maya antiquities appears to match the pattern of increased sales following important legislation (including the 1970 UNESCO Convention, the Mexico-US Treaty of 1970, the US and Mexican laws of 1972, and the 1977 case United States v. McClain). ^{Footnote 74} There was a slight decrease in the quantity of Maya antiquities from 1990 onwards, but it is not until the millennial decade that there is any significant reduction. Levine and Luna have also demonstrated a significant reduction in the quantity of Mesoamerican antiquities offered for sale at auction from 2000 onwards, as has Yates in regard to South American antiquities. ^{Footnote 75} As Yates later suggests, ^{Footnote 76} such reductions are likely a result of fewer antiquities being offered at public auction (rather than a result of a decrease in public demand).

Of the 3,229 Maya lots identified, it is unclear how many sold because sales data for every auction was not available to the author. Of the 150 auctions with Maya antiquities, sale data was only available for 78 auctions dating from 1966 to 2016. During this 50-year period, there were 2,240 lots, and, of these, 1,649 sold, demonstrating that 74 percent of the lots successfully sold (see fig. 3). Lots not listed in the final sale data may have been omitted, passed, or unsold—the latter is known as a “burned” lot. ^{Footnote 77} Lobay observes that if lots did not reach their reserve price at auction, the auctioneer informed the final bidder that there was a possibility to make a private transaction with the vendor outside of the auction. ^{Footnote 78} Consequently, although there were 591 lots not listed in the final sale data, some of these may have later sold in a private sale. Therefore, there may have been more than 1,649 lots that sold during the 50-year period. In comparison to the Maya lots, 2,798 South American lots were sold during a 39-year period. ^{Footnote 79} This suggests that South American antiquities are more popular among collectors purchasing at Sotheby’s public auctions.

Figure 3.

Comparison of the total number of lots offered for sale in 78 auctions and the total number of lots that sold.

Sale Prices

Sales results are sometimes inserted into catalogues after the auction has taken place, and they were therefore available to study. Sotheby’s required a fee for a printed list of all of the prices in the early catalogues, and, therefore, they were not attached to every catalogue studied by the author. More recent sale data is proudly exhibited on Sotheby’s website and, thus, more accessible. During the 50-year period for which sale data was available, the highest and lowest sale prices were recorded. For a fair comparison of prices, I adjusted all of the prices to the 2016 US dollar inflation rate using the US Department of Labor’s consumer price index (CPI) inflation calculator and rounded up to the nearest dollar. ^{Footnote 80} For example, $350 in 1966 is the equivalent of $2,595 in 2016. To convert Euros (€) and British pounds (£) to the US dollar, historical exchange rate data was used, ^{Footnote 81} and the US dollar value was then converted using the CPI calculator. The highest sale price achieved for a Maya antiquity during the 50-year period was the equivalent of $500,353 (in 2007) and the lowest was the equivalent of $117 (in 1980).

As Figure 4 demonstrates, the highest sale price has fluctuated through time. Sale prices in the 1960s through the early 1980s remained at $100,000 or under, but higher sale prices of $300,000 have become increasingly common from the 1990s onwards. Interestingly, a Jaina figurine from an auction on 17 May 2007 that sold for the equivalent of $500,353 is similar in style to a Jaina figurine that sold at auction on 9 November 1968 for today’s equivalent of $25,887 (both auctions were held in New York). This demonstrates that the relative value of Maya antiquities has increased over time. ^{Footnote 82} One of the higher sale prices reached in recent years was in Paris in 2013 (€325,500, equivalent to $435,150), attesting to the growing popularity of Maya antiquities outside of North America.

Figure 4.

Highest sale prices of Sotheby’s auctions over a 50-year period.

These findings parallel those by Levine and Luna regarding Teotihuacan antiquities. ^{Footnote 83} The authors propose that sale prices of these antiquities have continued to increase, despite quantities sold at auction decreasing, because Sotheby’s may have shifted their focus to a smaller number of higher value items. Additionally, they suggest that the establishment of various MOUs between the United States and Latin American countries may have discouraged the trade in Mesoamerican antiquities for which provenance information was weak. Sotheby’s vice-president of North American compliance, Elyse Dreye, communicated with the authors and agreed that the decrease was likely partly due to the company’s “decision to accept on consignment only those objects for which we can substantiate provenance going back far enough to satisfy our standards.” ^{Footnote 84}

These results have not been compared to sale estimates because, as Yates has explained, price estimates do not relate to how much the piece is valued at but, instead, are marketing tools designed to drive up the price or attract consignees. ^{Footnote 85} The official Sotheby’s 1980 price guide for Maya antiquities suggested a $700 price point for a Jaina warrior, for example, even though a Jaina warrior had sold for $3,750 in 1968. ^{Footnote 86} Furthermore, Yates suggests that estimates are subjective, and “no two experts will come up with the same price tag for the same object.” ^{Footnote 87} Lobay has also stated that determining the value of ancient objects is extremely difficult because they are usually rare and in many cases one-of-a-kind; thus, it is only the final sale price that becomes the true value of the object and provides the benchmark for subsequent sales. ^{Footnote 88} Therefore, it is the sale results, rather than the estimates, that speak to the market value of Maya antiquities.

Material Class

The most common material classes of Maya antiquities offered for sale at auction were ceramics (not including ceramic figurines), and the least common were wood and metal items (see table 4). These results correlate to the material classes deemed at risk in the International Council of Museum’s Red List of Central America and Mexico since ceramics are the first material class mentioned. ^{Footnote 89} Figurines and greenstone are the second and third most popular antiquities offered for sale, respectively, no doubt because they are portable items (thus, easier to export) and highly decorative. Only 4 percent (n = 135) of the total number of Maya antiquities were stone sculpture, whether complete carved panels, stelae, or parts of larger sculptural programs. It is likely that a much larger number of stone sculpture exists on the invisible market and is sold through private sales—the Placeres façade being such an example. An unusual amount of documentation is available about the looting and sale of the façade on the Trafficking Culture website, allowing what would have remained in the invisible market to become visible. ^{Footnote 90} The overall results of the material class study are similar to Gilgan’s findings. ^{Footnote 91}

Table 4.

Material classes of Maya antiquities offered for sale at auction from 1963 to 2016

Within the dominant material class, there were several characteristic types: Codex (identified by a white or cream background and black lines, often with red borders around the rim and base [see fig. 5 for an unprovenanced example]); ^{Footnote 92} incised or carved (such as the “Chocholá” style); ^{Footnote 93} other painted styles (such as “Chama,” “Holmul,” and “Ik”); ^{Footnote 94} stucco; ^{Footnote 95} appliqué; ^{Footnote 96} and plain (that is, undecorated). Table 5 demonstrates the different number of ceramic types, clearly demonstrating that painted ceramics (other than Codex) are the most prevalent and the least prevalent are plain ceramics. Over a period of 53 years, only eight plain Maya ceramics have been offered for sale at Sotheby’s. This low number is to be expected because undecorated ceramics are generally regarded as less valuable in comparison to highly decorative types. Thus, the material types sold on the auction market directly reflect market interest.

Figure 5.

Cover lot (lot 159) of a Codex ceramic from a sale on 23 November 1992 (K2096 © Justin Kerr).

Table 5.

Number of ceramic types offered for sale at auction from 1963 to 2016

Not only do the types of material offered for sale reflect the interest of the general market for Maya antiquities, but they also reflect the expected materials. For example, the poor preservation of organic materials in the Maya region hinders their archaeological recovery. It would be highly unusual for textiles, rubber, or bark paper books (known as codices) to appear on the auction market in any great quantity, and their absence from the recorded data reflect this reality. Although materials such as lithic and obsidian are prevalent at ancient Maya sites, their prosaic use for everyday tools likely make them undesirable on the art market. When they do occur, it is usually in the form of elaborate “eccentrics.” ^{Footnote 97} Of the three antiquities recorded as “miscellaneous,” two were iron pyrite mirrors and one was gold. Mirrors are found throughout the Maya region, ^{Footnote 98} but gold is rare in comparison and has only been recovered from a limited number of sites. ^{Footnote 99}

Owner and Provenance

As Yates explains, auction houses are under no obligation to reveal the identities of their consignees or buyers. ^{Footnote 100} Owners may not want to reveal their property for fear of publicizing it to criminals or because they do not want to reveal the sale of certain items to family members. ^{Footnote 101} Thus, the auction market has created strict rules to protect the identities of buyers, sellers, and the provenance—otherwise known as the history of ownership—of objects offered for sale. In fact, “there is no accepted international approach to standards of proof in provenance.” ^{Footnote 102} This makes tracking the provenance of antiquities extremely difficult and allows illegal antiquities to infiltrate the market. ^{Footnote 103} Provenance information in sales catalogues may consist of the name of a previous owner or the name of a dealer or museum from which an antiquity was purchased. In some instances, information regarding the length of time an antiquity has been part of a collection is also offered.

However, even in instances where provenance is offered, it is sometimes too vague to be of use. Early sales catalogues tended to describe lots as “from the collection of a lady” or the “property of a private collector” rather than provide a name. This ambiguity has continued through time and appears to be an accepted condition of purchasing antiquities at auction. In fact, Michael Stoll claims that the “reluctance to inquire into a piece’s history is endemic to the field.” ^{Footnote 104} Neil Brodie suggests that auction houses have no real incentive to reform their policies concerning publishing provenance information, allowing anonymity to continue. ^{Footnote 105}

In addition to vague information, there have also been instances in which provenance printed in a sales catalogue is false. ^{Footnote 106} Frank Herrmann records the sale of two suits of armor in a July 1917 sale at Sotheby’s in which false provenance was provided in the sales catalogue, demonstrating that false provenance has a long history at the company. ^{Footnote 107} The armor was the property of the Pembroke family, who claimed it was fifteenth and sixteenth century in date. Experts inspected the property ahead of the sale and not only attributed a much later date to them but also claimed that the provenance attributed to them was false. Geoffrey Hobson, then a partner in the firm, admitted he had not thought to doubt their traditional attribution and considered removing them from the auction. Instead, he arranged for Lord Pembroke to issue a letter to the press to support the provenance, and the auction house continued with the sale. This strategy proved successful, and both suits of armor sold for over $10,000 each, far surpassing any price that had been paid for armor at auction previously.

Although provenance information may be hidden to buyers, Sotheby’s is required to perform due diligence into the provenance of an antiquity to ensure that it does not willingly sell illegally smuggled items. The difference between legal and illegal antiquities is by no means easy to distinguish. ^{Footnote 108} The company has outlined its precautions to guard against illicit trading, stating: “Sotheby’s will not sell property if it knows that it was illegally exported or imported unless irregularities can be legally rectified before the sale.” ^{Footnote 109} Although Sotheby’s requires consignees to sign a guarantee that they have no reason to believe that property was illegally imported, Goodman has admitted the company “do make mistakes … we could not stay in business if I interrogated our clients.” ^{Footnote 110}

Despite the apparent good intentions of the company, staff at Sotheby’s have publicly voiced disregard concerning due diligence. In the 1991 court case Regina v. James Hodges, the head of antiquities at Sotheby’s was asked: “If you … suspected that an item might have been smuggled … would you be prepared to sell it through Sotheby’s sales?”; to which she answered “Yes, I think we would.” ^{Footnote 111} Furthermore, in the same court case, the managing director of Sotheby’s claimed that the company was not breaking any laws in selling smuggled material and that it was not their responsibility to police the market. ^{Footnote 112}

Of the 3,270 lots with Maya antiquities recorded in this study (including resales), 605 lots (19 percent) had their current owner listed and 419 lots (13 percent) had a listed provenance (ownership prior to the current owner [see table 6]). Conversely, this means that 81 percent of lots contained no information about the current owner, and 87 percent of the lots were devoid of provenance information. This is startling, but such high numbers correspond to studies of other antiquities sold at Sotheby’s. ^{Footnote 113} The names of current owners of Maya lots appear as early as 1963 and continue to appear, albeit sporadically, until 2016. Information about the provenance of lots first appeared in 1966, again in 1970, and then not again until 1980.

Table 6.

Number of lots with current owner and/or provenance listed from 1963 to 2016

Provenance information appears more frequently and in greater quantity from 1984 onwards. In relation to the number of sales per decade (see fig. 1), it is clear that owner and provenance information appear most frequently in the 1990s. Lobay discovered that the frequency of provenance descriptions decreased, while the volume of antiquities increased, after new regulations concerning the trade in antiquities were implemented. ^{Footnote 114} While the quantity of Maya antiquities matches this pattern, since the 1980s saw the greatest quantity of Maya antiquities being bought and sold, owner and provenance information also increases. The growth in owner and provenance information for Maya lots may be related to an increase in the chances of a successful sale. Emily Fay has suggested that auction market epithets such as “from the collection of” or “previously sold at” become a proxy indicator for authenticity and value. ^{Footnote 115} No doubt this is because it demonstrates that certain antiquities have been coveted by other collectors and considered to be of monetary value. De-accessioned antiquities from museum collections, in particular, are regarded as being prestigious and valuable. ^{Footnote 116}

Brodie explains that various scholars have theorized a process of auto-regulation on the antiquities market, whereby collectors discriminate against poorly provenanced or unprovenanced antiquities by paying higher prices for well-provenanced pieces. ^{Footnote 117} He tested whether provenance affected sale prices at auction to determine if auto-regulation was in place and discovered no strong evidence to suggest this is the case. In a similar vein, Brian Daniels, Sasha Renninger, and Richard Leventhal undertook a study of pre-Columbian antiquities in Sotheby’s and Christie’s sales catalogues from 2000 to 2010 to determine whether antiquities sold with a provenance had a price premium. ^{Footnote 118} They calculated an 80 percent premium on antiquities with provenance that pre-dated 1970 and an 18 percent premium on antiquities without a provenance. However, as Brodie explains, it is often difficult to separate the influence of aesthetics and object quality from provenance, and, therefore, it can be difficult to demonstrate evidence for auto-regulation from sale data alone. ^{Footnote 119}

Lengthy or detailed provenance information is by no means necessary to achieve a high sale price at auction. Lot 126 from a New York auction on 7 May 2016 sold for $125,000, despite the provenance information listed simply as “acquired in the 1960s.” Julian Dawes, Sotheby’s co-head of day sales, has explained that “the longer that something is off the market, the better.” ^{Footnote 120} Thus, although the aforementioned lot had no specific provenance listed, the information provided suggests the antiquity may have been part of a collection since the 1960s and therefore off the market for roughly 50 years. It is likely the antiquity reached a high sale price not only because it had increased in value over time but also because it was re-entering the market after a long period of unattainability.

In some instances, prior Sotheby’s sales are offered as provenance information. This is likely because it can demonstrate that antiquities were in collections outside of Central America prior to 1970, while allowing previous owners to remain anonymous. However, in some instances, information about prior sales is omitted from catalogues, likely to disguise the failure to sell. For example, Lot 79 from a sale on 26 November 1985 was previously offered for sale in a sale on 27–28 November 1984 as Lot 328 (both New York auctions), yet this information was not provided in the 1985 catalogue. The lot failed to sell in 1984, having been given an estimate of $8,000–$10,000. In 1985, unsurprisingly, the lot was given a lower estimate of $5,000–$7,000, and it sold for $3,300.

As well as being relegated to a lower estimate after failing to sell, it appears that the way in which antiquities are marketed to buyers also changes after failing to sell. Specifically, owner and provenance information are introduced to lot descriptions. For example, Lot 193, an incised vase, in a catalog dated 12–13 May 1983 failed to sell and was offered for sale again as Lot 99 in a sale on 26 November 1985 (both New York auctions). No owner or provenance was listed in the 1983 catalogue, but, in 1985, the information “property of an American institution” and the provenance “Cedric Marks collection” was provided. It failed to sell in 1985 and entered the market for a third time as Lot 152 in a New York sale on 19 May 1992. The 1992 lot description stated that the vase was the “property of the Manoogian Collection” and that the provenance was “Mr. and Mrs. Peter Wray Collection.” Unfortunately, sales data for the 1992 auction was unavailable so it cannot be determined whether offering the owner and provenance information secured a sale for the vase. Regardless, the sales catalogues demonstrate differing information about owners and provenance in 1985 and 1992. Perhaps a private sale moved the vase into the hands of the Wrays after the 1985 sale, and then into the Manoogian Collection, before coming back to market in 1992. This may be evidence of three different owners of the vase in the space of eight years, with ownership averaging just 2.6 years, or it may be evidence of false provenance.

Another example of owner and provenance information being introduced to sales catalogues can be seen with Lot 108, a carved lidded vessel, from a New York auction on 20 May 1986. Although the author did not have access to the sales data for this auction, the vessel is unlikely to have sold because it appears again as Lot 85 in a New York auction on 19 November 1990, despite it being the cover lot of the 1986 catalogue. No owner or provenance information for the vessel was provided in 1986, but, in 1990, the provenance was listed as “Mr. and Mrs. Peter G. Wray collection.” The addition of provenance information appears to have secured a sale since the vessel sold for $18,700. Although the year of sale is likely influenced by economics or market trends, this example may demonstrate that provenance information is considered to be valuable and influential (perhaps more so than an antiquity advertised on the catalogue cover).

Authenticity

In 1984, the following notice appeared in sale catalogues marketing pre-Columbian antiquities:

This statement remained in catalogues until 2000, but, from 2001 onwards, the first sentence was removed and replaced by a statement asserting that “all lots are sold on an ‘AS IS’ basis” (that is, caveat emptor). This “as is” basis appeared under the conditions of sale in early catalogues and explains that neither Sotheby’s nor the consignor provides any warranty about the accuracy of descriptions regarding, among others, condition, importance, provenance, or historical relevance of property. Thus, while the age of antiquities is not explicitly mentioned, it is likely that even without including the word “style” or excluding a date, Sotheby’s deem themselves free of responsibility for any mistakes regarding the age of an antiquity.

The decision to include such a notice in catalogues demonstrates that Sotheby’s is aware that forgeries, which are defined as copies of works of art made for fraudulent purposes, supplement the antiquities market, and buyers are at risk of unintentionally purchasing them. ^{Footnote 121} This is a problem that exists for art in general if suggestions that 50 percent of art on the market is not authentic is to be believed. ^{Footnote 122} The issue of forgeries in pre-Columbian art has concerned scholars for some time, ^{Footnote 123} and the expertise that Sotheby’s appears to offer here, in being able to distinguish between ancient and modern items, may have acted as an assurance for buyers.

However, Peter Watson demonstrates that Sotheby’s has failed to identify forgeries in the past. ^{Footnote 124} He recalls that the company went to great lengths to arrange the illegal export of an Egyptian lion statue out of Rome to New York, costing thousands of dollars, only to find out that it was not authentic, even though it had been assessed in person by Sotheby’s head of antiquities. The value of the statue dropped from a high estimate of $200,000–$300,000 to its final selling price of $8,000. Prominent antiquity collectors Sir Robert and Lisa Sainsbury are reported to have been well aware that inauthentic antiquities muddy the art market; they claimed to prefer to buy from dealers rather than auction houses because of the expertise that their dealers have in identifying forgeries. ^{Footnote 125} This speaks to the lack of confidence that some buyers may have in auction houses, perhaps because forgeries are being sold with success. The burgeoning online marketplace for antiquities is undoubtedly increasing the presence and sale of forgeries on the art market even more, as has been discussed by several scholars. ^{Footnote 126}

Yates suggests that the identification of antiquities as forgeries may be a method used by some governments and institutions to deter buyers from pre-Columbian antiquities. ^{Footnote 127} An example of this, she explains, is the 2013 Barbier-Mueller auction in Paris, which prompted Mexico to formally declare that 79 objects in the auction, supposedly of Mexican origin, were “handicrafts” (that is, modern forgeries). Unfortunately, despite Mexico’s protest, the auction achieved sales totaling more than €10 million ($12,948,000) and “established a new world record for the sale of Pre-Columbian Art.” ^{Footnote 128} Of the 3,270 lots that included Maya antiquities (including resales), only 12 (0.4 percent) included either the word “style” in the lot title or lacked an accompanying date. Interestingly, of the 12 recorded lots, five successfully sold. In comparison, Yates recorded three South American antiquities with the word “style” in her study of Sotheby’s catalogues, all of which successfully sold. ^{Footnote 129} Lobay explains that “determining the value of an object through its sale helps authenticate it as a genuine object and not a modern forgery.” ^{Footnote 130} Items that may not be genuine are given a sense of authenticity through their sale at auction and can become problematic for the marketplace, particularly if they are not marketed as inauthentic objects in future auctions. ^{Footnote 131}

In addition to the 12 lots with either the word “style” or the exclusion of a date, which appear to distinguish forgeries, an additional lot description highlighted a fake, which is defined as a genuine artwork that has been altered or added to for the purpose of enhancing the value. ^{Footnote 132} Lot 358 from a New York sale on 20 November 1995 was described as being “[a] Mayan Tecali Vessel, A.D. 550–950 and later, the straight walls slightly flaring … now incised overall with a modern frieze based on Mayan-inspired iconography.” Unfortunately, sale data is lacking for this auction so it cannot be determined whether fakes, like forgeries, have also successfully sold at Sotheby’s auctions.

Publication and Exhibit History

In the 4 May 1967 New York catalogue, four lots containing Maya antiquities included a reference to exhibitions that had showcased them. It was not until 1975 that lot descriptions included reference to exhibitions again and also included reference to publications in which the antiquities had appeared. Soon after, catalogues began to include this information on a regular basis. As Eva Pfeffer and David Gill and Christopher Chippendale explain, involvement in exhibitions increases the price of an artwork because it lends them authenticity and respectability. ^{Footnote 133} Likewise, publication of an artwork lends credence and undoubtedly increases its value. Of the 3,270 lots with Maya antiquities (including resales) recorded in this study, 287 lots (9 percent) were exhibited and 219 lots (7 percent) were published (see table 7). In relation to the number of sales per decade (see fig. 1), it is clear that, like owner and provenance information, the majority of publications and exhibits appear in lot descriptions in the 1990s.

Table 7.

Total number of publications and exhibitions listed in lot descriptions from 1963 to 2016

The decision to include exhibit history and publications is likely dependent on whether the owner of the sale offers this information to Sotheby’s. Some collectors have close relationships with museum curators or academics, offering to loan their collection for exhibit or share it for research purposes. The sale of the Barbier-Mueller pre-Columbian collection in Paris on 16 May 2013, for example, had 25 lots of Maya antiquities, and all were listed with accompanying publications. Only three of the 25 lots did not have accompanying exhibit history. This demonstrates that this particular collection had been widely published and exhibited. Many other Maya antiquities in Sotheby’s sales have been published in the Justin Kerr volumes, which are widely distributed among scholars (see discussion below).

The decision to omit publication and exhibit information from sales catalogues might also be taken, perhaps for anonymity or legal reasons. Watson, for example, claims that a Sotheby’s employee in Milan admitted to him that it is difficult to smuggle an antiquity if it has been published. ^{Footnote 134} He recalls an incidence of an Egyptian basalt statue that was sold through Sotheby’s in 1996 which demonstrates this situtation. ^{Footnote 135} At the time of the sale, the statue was in London, having been illegally smuggled out of Italy. The statue was published in three academic journals prior to the sale, proving that it had been in an Italian collection and later smuggled to London. The owner of the statue did not want these publications to be mentioned in the sales catalogue, and Sotheby’s apparently attempted to devise various ways in which to omit them. Legally, however, Sotheby’s could not ignore the publications, and the statue was removed from public auction and later sold through a private sale.

Justin Kerr Photographs

The final category for which information was recorded was the inclusion of Justin Kerr photographs. Kerr is a photographer from New York with a keen interest in the ancient Maya. ^{Footnote 136} He began his career as a commercial photographer and saw the benefits of applying the techniques of peripheral photography (also known as rollout photography) to the exterior of Maya vases. In the 1960s, Kerr began to photograph vases and other artifacts and, over the years, has compiled a portfolio of thousands of Maya antiquities. In conjunction with his late wife Barbara Kerr, he has published his photographs alongside scholarly essays and created a website so that his work would be freely available to scholars and the public alike. ^{Footnote 137} The gratis permission that Kerr has given for his photographs in scholarly and not-for-profit publications has led to an abundant and wide-ranging usage, particularly in art historical studies.

The antiquities that Kerr has photographed are largely in private collections, enabling otherwise concealed antiquities to be viewed and studied. The Kerrs photographed antiquities for various dealers and collectors of pre-Columbian antiquities, and Barbara has also restored antiquities for dealers. ^{Footnote 138} Therefore, it is no surprise that Kerr photographs have been included in Sotheby’s catalogues. Of the 3,270 lots with Maya antiquities (including resales), 59 lots (2 percent) were accompanied by a photograph taken by Justin Kerr, and the New York catalogue from 23 November 1992 even had a Kerr photograph on its cover (see fig. 5). The decision to include a Kerr photograph in a catalogue is likely a marketing tactic to add credence to both the monetary and scholarly value of particular lots—no doubt increasing attraction and perhaps even sale prices.