Examining Art Fraud

  • Kenneth Polk
  • Duncan Chappell


This chapter reviews the processes that establish whether a given art work is authentic, including scientific tests, consultation with experts and examination of the provenance. Of these, the role of provenance is underscored as a major process. Unfortunately, in numbers of cases, the true authorship of a work proves difficult to determine, leading to contentious situations where the authorship of a work remains in dispute. There are few examples of true fraudulent intent among the instances of non-authentic art, although these few cases are likely to be widely known and reported. While the actual number of such frauds is difficult to state definitively, it seems safe to conclude that most often the assumption of criminal intent in non-authentic art is unwarranted. Therefore, while terms such as “fake” or “forgery” are widely used in discussions of non-authentic art, the hidden assumptions of such words regarding intentionality will prove to be incorrect.


Art fraud case studies Policing Prosecutions 


In this chapter, we address the topic of art that is deemed to be non-authentic art that in other discussions might be referred to as “fakes” or “forgeries” (although we hope to make clear as we proceed why for our purposes as criminologists we will avoid using these terms). As the words are used here, “non-authentic art” will require first a clarification of what is meant by “authentic art”, with “non-authentic art” becoming thereby art that is not “authentic”. Our understanding, and definition, of authentic art is that it is, simply, what it is claimed to be, with non-authentic art becoming art that is not what it is claimed, or appears, to be. For example, an art object that is asserted to be the creation of a famous artist, when it has been created by another, lesser known (if known at all) artist, would within this usage become “non-authentic art”.

The Issue of Authenticity in Art

How can we be sure that any given art object is “authentic”? Conventionally, there are three approaches suggested as aids for examining claims about the authenticity of a given art object. First, one can submit the work to some form of scientific investigation to assess its origins. Second, the art object can be examined by known experts who have experience with the art in question, who, as connoisseurs, might be able to assess the claims made for the art. Third, the provenance, or ownership history, of the art can be examined to determine if a clear trail can be drawn from the time it was created by the original artist to the present time.

The Role of Forensic Science

One suggestion for help in determining whether an object is authentic or not is to turn to a variety of scientific tests that explore possibilities in this regard. Charney (2015, pp. 26–32), for example, provides a list of 16 different tests that can be employed in such an examination. There are illustrations of the important role of such testing when these have been used in actual criminal trials of suspected frauds involving art, as in the early case of Otto Wacker who was charged with fraud for selling bogus van Goghs (these events took place in 1932). In the criminal trial, what Charney (2015, p. 27) called “definitive evidence” was produced by a chemist who found that the artworks in question utilized a mixture of resin and lead to make the oil dry faster, a technique that van Gogh never employed, that evidence being central in the conviction of Wacker. In a more recent case, the noted German fraudster Wolfgang Beltracchi was found out in part because of his use of titanium white, a paint scientific testing revealed that had not been developed when the artworks in question were supposedly created (Charney 2015, p. 157). Also in a recent case involving the Knoedler Gallery, a painting in 2011 supposedly by Jackson Pollock that had been sold for US$17 million when sent for forensic testing was found to have contained a yellow pigment not available until well after Pollock’s death (Charney 2015, p. 162).

The results in these cases were definitive and helped bring down the fraudsters who employed techniques that testing demonstrated could not have been used by the artists claimed as the authors of what were determined to be non-authentic works. These cases, and other similar examples, demonstrate how scientific testing can play a defining role in cases where the authenticity question reaches a criminal court in the form of a trial for art fraud.

Unfortunately, such testing can be much more effective when it provides clear evidence that the work in question is not what is claimed by the defendant. In actual use, it has proven less helpful in demonstrating that the work is authentic as described. Testing, of course, can be of assistance in building a circumstantial case. If the artist in question is left-handed, it is helpful (but far from definitive) to find from a microscopic examination of the brush strokes that the work appeared to be produced by a left-handed person. A general observation is that while scientific procedures may at times produce evidence that a work definitely could not have been done by the claimed author, such testing is acknowledged to be much less decisive in making the case that the work was done by the claimed author. In the terms of the present analysis, while scientific testing has proven to be decisive in establishing in some cases that the art in question is not authentic, it has time and again been shown to be far from conclusive in determining that the work is authentic.

The Role of Connoisseurship

A second way of assessing the authenticity of a work is to consult those who have expertise in assessing the work of a given artist. At times, those who are experts in dealing with a given artist have what is to an outsider an amazing sense of the work of an artist and are able to judge the authenticity with a clear sense of command regarding the task. Often, such experts are family members of the artist, and their association over the years of being close to the artist gives them a unique sense, or “feel”, about a work’s authenticity. Unfortunately, however skilful such experts can be, there may be other experts who may not share the conclusions of such connoisseurs. Even when contrary opinions are not available, the language used by such experts may be far from convincing, especially if the matter reaches a courtroom.

Furthermore, the record of connoisseurs indicates that such experts are not infallible. The history of art is riddled with experts who were shown to be wrong in their assessment of authenticity. One example of this is the false Vermeers produced by van Meegeren. In the 1930s, Abroham Bredius was considered to be one of the outstanding scholars regarding the works of Vermeer. However much he knew about Vermeer and his paintings, Bredius was wrong in his very public assessment of the false Vermeers produced by van Meegeren, since his praise of the “new Vermeers” ultimately was shown to be incorrect (see the story recounted by Charney 2015, pp. 94–100). The problem of the inconsistency of known experts has been recounted at length, so much so that observers such as Charney (2015, p. 26) are able to assert, with considerable evidence one should add, that the use of such experts has in the present day “been demoted from its perch as the definitive method for authenticating art”.


Provenance refers to the ownership history of the object in question. As we shall see, where the work has been produced in very recent years, the various players (artist, dealers, collectors) who have been directly involved with the work may still be available, or at least their actions regarding the object may be within the living memory of those close to them. In these circumstances, it may be possible to define a clear history of ownership. As time erodes these ties, the issues of provenance mount.

Confirmed Identity with Solid Provenance

One form of what appears to be solid provenance arises when the work is purchased by a major collection or collector shortly after it has been created, often in the lifetime of the artists. The National Gallery of Victoria (the NGV) purchased several works by the noted Aboriginal artist Emily Kam Kngwarray before her death in 1996, including such paintings as Ananty (Wild Potato) purchased in 1989, the very large (roughly 3 by 8 metres) Anwarlarr anganenty (Big yam Dreaming) purchased in 1995 or Awaly purchased in 1991. The Gallery description of each of these gives a sense of the origin of the work and how it came to be in the collection of the NGV. In this case, there can be little room to doubt that the provenance of these works is firmly established. There are a number of other examples of purchases where the work was obtained shortly after it was created, so that there is little question about the authenticity of the object, as in the case of the painting by the Australian artist John Brack (who died in 1999) The Car (painted in 1955), purchased by the NGV in 1956, or his Nude in an Armchair (painted in 1957), purchased by the NGV in 1957. Similar conclusions can be drawn about the work of other Australian artists purchased by the NGV, such as Fred Williams, since many of the 478 works by this artist now in the NGV were purchased before, or shortly after, his death in 1982. Where the museum has purchased the work so soon after its creation, often while the artist is still alive, few questions can be raised about authenticity since the trail of ownership history is clear for all to see.

A rather different example of what appears to be solid provenance has been provided by a recent auction conducted by Sotheby’s in Sydney (Cockington 2016). One of the items auctioned was the painting, Sydney Harbour by Sir Arthur Streeton. The work was painted in 1907 and first was acquired by a main patron of the artist, Sir Walter Baldwin Spencer, living in Melbourne, for £75 (at one stage, Spencer was said to have owned 41 works by Streeton). In 1919, the work was purchased by Sir Thomas Elder Barr Smith, a pastoralist in South Australia, for 500 guineas. It remained in the hands of that family until 1976, being sold at that time but staying in private hands. The work is still in its original frame. In this instance, the artist had established a solid reputation at the time the work was created, and the pathway of ownership from that time can be traced without interruption to the time of the current sale, and thus becomes yet another example of confirmed identity with solid provenance (although it is notable that, since its original purchase, the work has remained in private hands).

Examples of Acceptable Provenance: Reasonably Firm Identity with Acceptable but Perhaps Incomplete Provenance

As we shift the focus to earlier centuries, it is no longer possible to rely upon, or to expect, thorough and complete ownership histories for art objects. We have the example of the tracing by Pitman (2006) of the provenance of the painting by Raphael, St. George and the Dragon, created in the opening years of the sixteenth century. The work can be found today in the National Gallery of Art in the United States, where it has been since being donated to the museum in the 1930s. Before that, it was definitely in the hands of the Russian Government, previously it had been in the collection of Catherine the Great and before that it was known to have been owned by Henry VII, Henry VIII and Charles I in England. Documentation indicates that it was commissioned by the Duke of Urbino in Italy in 1506. While it was presumed to be in England in the 1500s (it apparently was commissioned by the Duke of Urbino in response to overtures made to confer upon the Duke the Order of the Garter of England), the first known and dated sighting in England that can be verified was in 1627 (Pitman 2006, p. 106). While the gap in the provenance here involves a long period, today it is generally accepted that the St. George and the Dragon in Washington is a work by Raphael.

Sometimes the identity of works of early artists is supported by work done by gallery scientists, as is the case of the Raphaels in the National Gallery in London that have been subjected to careful analysis, showing (among other things) that the procedures used by Raphael in the paintings were consistent with practices used at the time he was working (see, e.g. Ashok et al. 2004). What we see in these examples is a weaving together of observations based on what limited provenance information is available, expertise from those knowledgeable about the art period in question and scientific analysis which can be helpful in providing assurance that the techniques employed were consistent with what would be expected from the artist examined.

Examples of Provenance That Is Less Firm, Where There Are Unanswered Questions About the Origins of the Works Including Its Authorship, but Where the Present Owner Seems to Have Responded Responsibly to Questions of Authenticity

An example where the provenance information is far from firm can be found in the painting of Henry VIII currently in the holdings of the Art Gallery of New South Wales (AGNSW). According to information provided by the AGNSW in their current website, the earliest information about the work is from a 1945 Christie’s catalogue, which listed the previous owners as “Messers Vokins” of London, England, dates unknown; the Earl of Wemyss, Scotland, dates unknown; and the Earl of Gosford, Northern Ireland (current owner). The work is reported as being connected with P&D Colnaghi and Co. in 1948, and then it was purchased by the Art Gallery of New South Wales at an auction at Sotheby’s in London in 1961. In a 1945 sale, there was a note that the work was “as by Holbein”, but in the current AGNSW listing, it is described as a product of an “Anglo-Flemish Workshop” with a date estimated to be 1530–1550. The vague and open dating is a common experience in the museum world. The work is clearly quite old, but unfortunately little is known about most of the years before the recent decades. There is, simply, no provenance data available for the work for much of the first 500 years of its existence. It seems reasonable to assume that it is not a work by Holbein, and the alternative description now provided by the museum is the vague workshop designation. While an accurate and correct description, and quite a reasonable one for the museum to give to the work, it is hardly a firm statement of the author of the object. It is necessary here to acknowledge that the absence of information about most of the early and middle years of this work, including the authorship, precludes any definitive statement regarding who was responsible for producing the painting, or even who has owned the work over the many years of its history.

It needs to be recognized that this absence of historical information is quite common for works that are many years old, and in these circumstances, it must be assumed that even the artist who created the work may not be known. The correct designation for such a work, as the museum in this example indicates, is uncertain. In these circumstances, which must be acknowledged as common, the descriptive terms must give credence to the absence of definitive information regarding authorship. While comments such as the painting being a work from an “Anglo-Flemish Workshop” may appear vague, such terms appear to convey accurately what is known, and not known, about the object.

The Problem of Non-authenticity

From the foregoing, we have seen that the question of authenticity can be complicated. In some cases, notably more common with works that have been recently created, it is possible to base the conclusions that the work is what it is claimed to be on the basis of the firm and known provenance of the work. In some cases, the authenticity of the work is generally accepted, despite the fact that there are gaps in its known provenance. In other cases, all that can be asserted is that the work belongs to a certain “school”, or is perhaps the product of a known “workshop”, the very words conveying the uncertainty about the work. In this last case, the authenticity of the work results from a recognition of the limits imposed by the absence of information. In all of these different situations, assertions can be made that the work is as it is claimed, that is, it is “authentic”. Taken together, these various distinctions provide a collective definition of authenticity. The procedures used to make these determinations also create the conditions that make possible the determination that a work is non-authentic. This problem—that of non-authenticity—arises when claims are made about the identity of a work that can be shown to be false.

How often do we encounter situations where a work can be deemed non-authentic? Early on it is necessary to address the question of “how much” non-authentic art there is. The answer to that has to be, at least from us, that we are unable to provide a definitive answer, however important this question may be. Thomas Hoving, who served for a time as head of the Metropolitan Museum of Art in New York, asserted in his memoir book that in his time at that museum, he had examined at least 50,000 works of art, and he estimated that a full 40% were “phonies” (1996, p. 17). Sloggett is more modest, both in her use of language and in her estimate of the level of the problem, asserting that roughly 10% of the works appearing on the international art market are examples of “mistaken identity” (Porter 2007). There is ultimately no source that can be consulted for a precise number or percentage that can be used as a benchmark for the important, but unanswerable, question of “how much” non-authentic art exists. All we can attempt at present is to indicate the rather different forms that non-authenticity can take, with most being cases of what can be termed “mistaken” or perhaps “disputed” identity, while, as we shall see, relatively few cases can be found where work is the result of the dishonesty that must be established in cases of art fraud.

The Problem of “Honest Mistakes”

A first group of non-authentic works consists of those cases that appear to mesh most neatly with Sloggett’s suggested term of “mistaken identity”, that is, cases where it is possible to determine that the original attribution is not correct; yet, there is no hint that the current mis-attribution is in any way dishonest or the result of a criminal intent to deceive.

As an example of such possible “honest mistakes”, The Head of a Man painting was sold a number of times between the 1920s and 1930s, finally being purchased by the NGV in 1940. At the time, it was considered to be the only van Gogh in Australia and displayed proudly as such by the NGV. In the early 2000s, it was sent to an exhibition of the work of van Gogh in Scotland, where concern was expressed about the correctness of the attribution. Close and intensive inspection by authorities in Amsterdam decided that it in fact was not by van Gogh. It was, in short, a case of “mistaken identity”. At no point has there been any concern that the mistakes that occurred were a consequence of known dishonesty. The gallery in question in fact attempted to assure the public that, while the original attribution was not correct, this work should not be viewed as a “forgery” (NGV 2007).

This tale becomes even more involved since the purchase in 1928 was by a German industrialist by the name of Semmel. As a Jew, he became concerned by the election in early 1933 of Adolf Hitler to power in Germany and he moved to Amsterdam where he sold the van Gogh at an auction in the summer of 1933. Semmel later moved to Chile and from there went to New York, where he died in the 1950s. Once they found out that the Head of a Man was in the possession of the NGV, the Semmel heirs asked that the painting be returned to them, arguing that the sale was a consequence of the coercion imposed by the political conditions in Germany at that time. Drawing upon the so-called Washington Principles (US State Department 1998) that were developed in the l990s to deal with precisely such claims, the NGV decided to return the painting to the Semmel heirs.

Over the many years to the present, there has been the development of an appreciation of artistic work and with that the development of an art market and that in turn has created an increasing concern about the presence of false works attributed to famous artists. Indeed, of Corot it was famously quipped that: “Corot painted three thousand canvases, ten thousand of which have been sold in America” (this often-quoted observation has been attributed to Rene Huyghe, e.g. see Keats 2013, p. 99).

This presence of false works attributed to famous artists has a number of negative effects, including, of course, the loss of value of authentic works, even objects with what appears to be solid provenance. One result is that those interested in the protection of the authentic body of works of an artist will take steps to preserve the identity of the artist. An example was begun in the late 1900s, when the Rembrandt Research Project (RRP) was created in The Netherlands to address the problem of the many works which were attributed, incorrectly, to Rembrandt. After many years of work, the RRP in its first three volumes of published research findings found that 52% of works examined by the RRP had been correctly attributed to Rembrandt, while 43% of the objects were not correctly attributed to this artist, and in 4% of cases, the correct attribution could not be determined. As a result of the work of these researchers, in other words, close to half of the art objects previously attributed to Rembrandt could be considered as the work of someone other than Rembrandt. They were, in short, in the opinion of the RRP, cases of “mistaken identity” (The Rembrandt Research Project 1983).

“Disputed Identity”

Another group of works might be designated cases of “disputed identity” where a debate exists about the correct attribution of the work. In the BBC television programme “Fake or Fortune” (aired in Australia on 23 February 2016 by the Australian Broadcasting Corporation) the topic for the programme was a painting where the identity of the artist was in dispute. The painting had been included in the catalogue of works by Pierre Auguste Renoir published by the distinguished French firm Bernheim-Jeune (who also in earlier years had been the agents for the works of Renoir), but it had been rejected as not an authentic work of Renoir by the Wildenstein Institute, also located in Paris. Without the endorsement of the Wildenstein Institute, the painting will not be accepted for sale by the major auction houses. As such, the painting has little value. In the course of the television programme, considerable circumstantial evidence was reviewed which supported a claim of authenticity by the current owners, including a review of provenance information. The painting could be traced to the Monet family, which had held the painting since Renoir had gifted the work to his close colleague Monet, and it emerged from the Monet family as a consequence of the settlement of the estate of Blanche Monet in 1937. In addition, a variety of scientific findings supported the claim that the work had been done by Renoir (including a comparison of the pigments used by Renoir set out by the artist himself at the time the painting was completed, these pigments shifting to quite different pigments as Renoir’s career evolved). Further, there were supportive opinions of several Renoir scholars who were interviewed in the programme. The Wildenstein Institute reconsidered the earlier conclusion as a result of the submission to it of the new findings, but that organization concluded that the work still should not be considered an authentic painting by Renoir because it had not been signed by the artist, the provenance information was not compelling and the painting was weak and not up to the standard of a genuine Renoir.

Another illustration of the problem of dispute about the identity of the author of a work can be found in the book produced by the Metropolitan Museum of New York regarding the early Chinese artwork Riverbank. The opening chapter of the book is an “Indictment” listing 14 counts against the authenticity of Riverbank by Prof. James Cahill, arguing that the object is probably a modern bogus work created by the well-known Chinese faker, Zhang Daqian. In a later chapter, M K Hearn, then the curator of Asian Art at the Metropolitan, argued on the basis of an analysis of the physical evidence that: “While such a physical examination cannot settle the question of Riverbank’s authorship or period, it nonetheless makes it impossible to argue convincingly that it is of modern manufacture” (p. 112). This provides an important example of how scientific analysis can help in ruling out some hypotheses, but it leaves in place important questions about the “ruling in” of a hypothesis of authorship. Also, it provides an important example of the limits of connoisseurship. In this case, accordingly, if we accept the argument (and the data) assembled by Hearn, we are forced to reject the argument advanced by Cahill (accepted as a competent expert on Chinese art) that the object had been created in modern times by Zhang Daqian, but as an alternative all we can assert is that it has been created at a much earlier time. Thus, even if we accept Hearn’s argument, we remain with the observation he makes that we still do not know the identity of the original author of the work. We have “ruled out” the possibility that the work was created by a modern artist, but we are unable to “rule in” any conclusion about the identity of the creator of the work.

Another example of how authenticity can be disputed concerns the Rembrandt Research Project (RRP) and its evaluation of the work, The Polish Rider, attributed to Rembrandt, which is in the Frick Collection in New York. In the first round of results of the RRP, Josua Bruyn suggested that characteristics of the work of Willem Drost, a student of Rembrandt, could be found in the painting (although the technique in the painting of the face seemed to suggest to many the hand of Rembrandt). The Frick at no point changed the attribution of the work, and recent opinion has tended to support this attribution (Bailey 1994). This notable rejection of the conclusion of the original Rembrandt Research Project is not unique. In fact, slightly over one-third (44, or 37%) of the original works originally attributed to Rembrandt and considered by the RRP not to be by Rembrandt in its first three volumes, apparently, by the time the last volume of the RRP was published in 2014, had been “reinstated” (White 2015). In our terms, the conclusions of the RRP about these 44 works had been disputed, and these were now successfully “reinstated” as legitimate works of Rembrandt.

At times, the courts are called up to assess claims of authenticity where a dispute emerges. This situation arose recently in North America, where a well-known artist (although originally from Canada, he currently is based in both London and Trinidad) was involved in a US Federal court proceeding alleging that the artist had produced the work some 40 years ago (the suit according to press accounts was filed by the person who owned the disputed painting). In this instance, the lawsuit arose because the artist in question denied that he was the author of the work. Since the works of the artist have been sold for up to US$20 million, there was a considerable financial stake in the outcome. The judge in the case, however, found that the painting was emphatically not done by the famous artist, stating that he “could not have been the author of the work” (Perkel 2016, p. 1).

An unusual feature of this case was that the lawsuit was not filed by the artist, but by the current owner of the painting in question. In sharp contrast, in a court case that took place in Australia, two artists entered into a civil action against a Melbourne art dealer, accusing him of selling non-authentic paintings allegedly done by them (Blackman and Ors v Gant 2010). This is probably a more common form of dispute, since often artists (or their legal representatives) are forced to seek legal means to protect their reputations from bogus artworks supposedly created by them. In this case, the court found that the artworks were not created by the two artists and ordered that the non-authentic works be given to the two artists, who then were able to destroy the false paintings. It is interesting that this was a civil, not a criminal, case and that actually the court did not address the culpability of the dealer in question, focusing solely on the issue of the lack of authenticity of the works of the two artists.

Lest this be seen as unusual, an example was seen recently in Korea, where the representatives of the family of an artist are suing a museum they allege has been showing for a number of years a painting by the artist Chun Kyung-ja, which the family argues was not created by her (the artist herself, while alive, made the same claims). Interestingly for present purposes, a French firm has now been engaged to carry out scientific analyses to see if it is possible to determine the facts of this complex situation (Shin 2016).

Fraudulent Identity

Finally, we arrive at the problem of non-authentic art, where there is clear dishonesty by those currently placing the object onto the market. In these situations, the deception must be viewed as more than a mistake, since, alongside the deception, dishonesty is involved. While diverse terms are used by those in the art market to describe this situation, here, we note that at the core of them there is the issue of criminal fraud as these terms are used in the criminal law. As we examine the narratives of mistaken or disputed identity above, where the art object can ultimately be found not to be what it is claimed, we can see that the focus properly is upon the object itself and the accuracy of the description provided. When we turn to the question of fraud, we find that, in addition to this issue of accuracy of the attribution, there is the dimension of the motivations behind the introduction of that object into the art market.

Commonly in the criminal law, there are four “elements” that must be present for a successful prosecution for fraud to occur: (1) some form of deception on the part of the defendant must be proven and, in the case of an artwork, it must be shown that the object is not what it is claimed to be, that is, it is deceptive, or, in our terms, not authentic; (2) that deception must have arisen out of dishonesty or the criminal intention on the part of the defendant; (3) that dishonest deception must have produced harm felt by the victim; and (4) that victim was actually deceived by the defendant (Lanham et al. 1987).

In the circumstance of fraud involving an art object, a first matter for the prosecution is to prove that the art is not authentic, that it is in fact deceptive. Proving this in most situations will involve the use of testimony of expert witnesses who will provide the necessary evidence that the art is not what it is claimed to be. At times, this will be a simple matter of presentation of evidence of materials of some sort that are inconsistent with the period of the artist, or a similar situation where the materials used provide clear evidence that the object could not be authentic. In other cases, proof regarding this important element may be more problematic and rely on evidence of stylistic differences between the artist claimed as the author and the style employed by the person creating the deception.

The second element, dishonesty or criminal intent on the part of the defendant, will require proof that is most often circumstantial. For example, the working methods of the person creating the deception will be examined, since these often make clear that these procedures would only be considered if dishonesty was contemplated. Thus, there may be evidence, such as false labels among the working gear of those involved in the deception, or perhaps even the characteristics of the framing materials used in the creation of deceptive works.

In cases of art fraud, once these two elements have been established, the other two elements, namely of deception that must have produced harm felt by the victim and that the victim was actually deceived by the defendant, while relevant, are actually manifest. Thus, if a work is claimed to be that of a Golden Age Dutch painter, when it is a modern creation and it has been sold as an old work to the victim at great cost, one hardly needs to spend time demonstrating the harm suffered by the victim. Similarly, in such a case, it will be obvious that the victim has actually been deceived by the false work; otherwise, the large payment would not have been made.

Cases of Art Fraud Are Rare

From all of the foregoing, we can make the following observations. First, the instances of art fraud, despite popular concern, are actually quite rare. We know of less than a dozen cases of art fraud that have reached the courts in Australia over the past four decades or so. That country is obviously not a major player in the world art market, but even in such countries as the United Kingdom or the United States, this same observation can be made. Unfortunately, official statistics compiled by policing or court authorities are silent on the question of how much fraud there is, but in all countries where art is sold, we find few cases of art fraud being prosecuted. In books on the various forms of non-authentic art, support for this assertion can be found in the scarcity of cases and the frequent use of the same art fraudsters by quite different writers as exemplars of this activity.

Five issues can be commented upon to account for this scarcity of cases of art fraud. First, we have to acknowledge that we can only know about cases that have been brought into the criminal justice system and where the media comment about the case brings it to our attention. Thus, there may be cases where either there is no comment by the media about the fraud or the frauds involved have not yet been uncovered by the criminal justice system. In Australia, Grishin (2015) has argued that he knows of at least 15 fakers who have been active in recent years and who have not yet been discovered. This number of “undiscovered” producers of what may be fraudulent art, while not a large number, is somewhat greater (to an unknown degree) than the small number of cases that have been brought forward into the courts of Australia. Second, the police are a necessary and important part of the criminal justice system, providing the entry point and the bulk of the investigation into the case. Yet, typically police are not eager to pursue cases of art fraud. This form of crime does not fit stereotypical patterns of what is considered to be “real” crime, few police have any training regarding art, the victims are often seen by police as not being “worthy victims” and the amounts involved most often hardly seem even to police concerned with fraud to be worthy of the effort involved. Third, their role in the fraud often places victims in a position where if they keep silent and are not identified as a victim in this instance, the non-authentic art that they have purchased, often for large sums, might later be passed on as authentic and sold on the market, thereby preserving the investment in the art that may otherwise become worthless. Fourth, and often overlooked, is that any of the large schemes to move non-authentic art onto a market as authentic works can only be successful if the works are produced by a person with some level of skill as an artist. Those who possess such skills are likely to seek recognition in their own name if they are artists. Fifth, we have to acknowledge that any prosecution will inevitably involve reasonably significant works of art, so that those working at the bottom of the market, for example, on EBay, for small amounts of money (e.g. US$500 or even less), will not appear in the records of prosecution for art fraud.

The Problem of Serial Offending

As one reads many of the accounts of art fraud, one is struck by the obvious fact that had the criminals involved stopped after one, or perhaps two, of the fraudulent transactions, in all probability the crime(s) would not have been discovered. Time and again, however, the ultimate undoing of the fraudsters occurs when they attempt repeated entry into the art market. This repetitive entry into the market poses perils for the offenders. In addition to unpredictable accidents that are inherent in the enterprise, the fraudsters must attend to such issues as avoiding the emerging awareness of the market that fraudulent works are originating from a particular source and thus must be ready with ever-new inventive stories that can be told to provide an acceptable, but false, provenance for the object.

Despite this, it would be exceptionally rare for even a skilful faker to be at work repeatedly producing works on the market without some within that market becoming suspicious. However unreliable they might be for legal purposes, connoisseurs are likely to be aware that a “new hand” is at play in the production of artworks. If there is continued entry into the market, rumours are likely to begin to spread, and even initial steps taken to address the source of the false works. At this point, it seems not uncommon that those involved develop what can only be termed “defective risk assessment” strategies, strategies that not only become “unstuck”, but then their very complexity leaves identifiable traces behind that ultimately provide critical evidence of the fraud that substantiates claims for the dishonesty of the defendants. (For an example, see the account of the activities of Drewe and Myatt in the work of Salisbury and Sujo 2009. These activities included the falsification of the archival records of major art institutions, like the Victoria and Albert Museum in London, which, when eventually detected, provided an invaluable evidentiary trail of deceit and deception leading back to the ringleader in the case, John Drewe.)

A Second Emerging Trend: The Issue of Two Kinds of Expertise as a Feature of Art Fraud

In addition to serial offending, we may also be seeing a trend whereby a person, who has ready access to a portal in the art market, uses that advantage to set up a fraudulent art scheme, rather than the artist creating the fraudulent object in the first place. This person makes use of a portal, or perhaps shifting portals, into the art market in order to insert the fraudulent objects. This form of art fraud involves two quite different forms of expertise. First, there must be an artist involved who can create objects that will pass review by knowledgeable individuals located in the art market. Second, there will be a person (who ultimately becomes quite central to the fraud) who arranges for the entry and sale of the object onto the art market. In one landmark case centred in the UK market, John Myatt served as the artist for the frauds, but it was John Drewe who arranged for the entry of the forgeries onto the art market (the details of the fraud, including rather elaborate schemes to obtain false support for claims of provenance, are described by Salisbury and Sujo 2009). These same dynamics are seen in the sad tale of the downfall of the Knoedler Gallery in New York, but in that case, the only portal selected by the fraudster was the Knoedler Gallery. The artist in this case fled to China, where he apparently remains free from extradition by the US authorities.

In Australia, these dynamics were allegedly found in the combination of Gant (the dealer) and Siddique (the person who created the artworks), both convicted in 2016 but subsequently acquitted on appeal in 2017 (see McKenzie-Murray 2016; Brown 2017), and also in the early case of Curvers (Hills 2002) (who apparently was the fraudster and mastermind) and Blundell (whose “inneundos” were, according to his telling, known by Curvers to be false works). Since the police were unable to find evidence contrary to this account, Blundell was not convicted (Curvers had died before the inquiry had begun). While this division of labour has the advantage of permitting the individuals to use their talents to an advantage, the downside is that the repeated entry involved not only begins to strain the ingenuity of the fraudster passing the works into the market, it also begins to leave indelible signs of intentionality and dishonesty necessary for a successful prosecution for fraud, as we have already noted, for example, in the Drewe and Myatt case.

The Issue of What Terms to Use

The perspectives found in the art arena are numerous, and it is absurd to think that there will be any consensus among them about how to use such terms as commonly used as “fake” or “forgery”. Some will elect to flop about, using one term here, another term there. Kurz, in the preface to the original edition of his classic book on fakes (his term), goes back and forth between the terms “fake” and “forgery” at least a dozen times on just the one page, implying equal meaning to these words (see Kurz 1967, p. vi). Others may attempt to impose a standard definition as they proceed. For example, Charney (2015) prefers the term “forgery” as the suitable generic term for intentional non-authentic art and defines this as “an object made in a wholesale, fraudulent imitation of something else”, which is seen as different from a “fake” defined as “an original object that has somehow been altered or ‘doctored’—a painting, for example, to which a spurious signature has been added” (Charney 2015, p. 16). A quite different take on how these words might be used can be found in the recent discussion by Sykes (2016), where we find that forgeries are: “what most of us are thinking of when we imagine art crime—bogus works made by accomplished modern-day artists deliberately to resemble the real thing”, whereas fakes “are works by other artists that are passed off as works by more important artists”, or this word can be applied as well to “copies of a work done at the same time or in the same era as the original” (Sykes 2016).

We are not attempting to argue that other writers, especially those coming from other perspectives, should not use such terms as “fake”, “forgery”, “confabulation”, “bogus” or similar terms in their writing. In such usage, the terms will presume a meaning close to their dictionary definition. A word such as “fake”, of course, can be used as either a noun, an adjective or as a verb. Typically, the meanings of “fake” would be rendered as, if a noun, “any person or thing not genuine”, or “an object that is made to look real or valuable in order to deceive people”; if an adjective, “not genuine, spurious”, or perhaps “not real, but made to look or seem real”, while, if used as a verb, “to make up and attempt to pass off as genuine”, or “to make an object look real or valuable in order to deceive people”. There is a hint in some of these definitions (and in the common usage as found in the media) of some level of deception, which in some usages implies even more: that the deception is deliberate, even dishonest.

Since it is recognized that terms such as “fake” or “forgery” are going to have wide currency in writing concerning art that is not authentic, it is important that we make clear why we have avoided such common terms. In our view, the word “fake” carries with it unfortunate connotations of intentional deception that may, or may not, be present when a non-authentic work is the focus of attention. This would be true in the situation of “an honest mistake”, or even “disputed identity”, as these terms have been discussed here. In the cases of, what we have termed mistaken or disputed identity, there is presently no concern with dishonesty. If dishonesty is not present, then the concern for fraud drops away.

While we have not employed the term “forgery”, our reasons in this context are a bit more complicated. Those with a background in criminal law, for example, will know that use of the word “forgery” may present problems. There was a case (R v. Closs 1858) in the middle of the nineteenth century in England where the presiding judge laid down a finding that forgery was an offence that involved documents, but since a work of art did not constitute a document, the criminal law regarding forgery could not be applied. While the logic may prove a bit confusing in the contemporary context, in some common law jurisdictions, such as Australia, this case still constitutes the relevant precedent; so technically speaking, the charge of forgery may not be pursued where the false object in question is a work of art (a charge of fraud, however, may be applied).

While words such as “fake” and “forgery” will have some reference to non-authentic art, only where the key elements of fraud are present is it going to be possible to speak of art fraud. It is interesting that many of the examples of non-authentic art discussed by Keats or Charney are actually examples of proven art fraud. Four of the six artists discussed by Keats (2013) as “modern masters” actually went to trial (although one, Keating, was let off by the prosecution when he became seriously ill). These same four are mentioned in Charney’s book, which includes the stories of several other individuals who were found guilty of fraud, including the recent cases (not mentioned by Keats) of such figures as Drewe, Beltracchi and those involved in the sad fate of the Knoedler Gallery in New York.


A first conclusion is that understanding non-authenticity in art requires first and foremost a grasp of what underlies authenticity. For works that are many years old, there may be only fragmentary evidence that supports a conclusion that the work is authentic. Thus, the combination of information about provenance, the combined evidence of experts and the scientific evidence will at best lead to a tentative conclusion that the given work is authentic.

Second, most of the situations where art is deemed to be not authentic will involve “mistaken identity”, either in the form of what have been termed here “honest mistakes” or “disputed identity”. Much of the time, dishonesty will not be a part of the current decision about the identity of the work, although at some time in the past, often in the distant past, the object may have been produced by individuals who were motivated by dishonesty.

Third, despite recent well-publicized cases of fraudulent artworks entering the upper end of the international art market, this type of crime is still quite rare, at least on the evidence currently available primarily from criminal justice sources. These sources clearly have deficiencies, not the least of which is the reluctance of the victims of art fraud to report their victimization to law enforcement authorities. But most of the instances where an object is deemed to be non-authentic will be a result of mistaken identity and will not involve the criminal intent required for a conviction for fraud.

Fourth, from the viewpoint of those working within a criminal justice framework, it is relatively easy to avoid the use of terms such as “fake” or “forgery”. In its common use, the word “fake” often presumes some level of dishonesty on the part of the person or persons placing the object onto the art market and such an interpretation would be inappropriate in cases of mistaken identity. While we would urge caution when describing the art object in a way that might cause confusion, we recognize that others may find terms such as “fake” or “forgery” more to their liking. We ourselves find it difficult to avoid using the term “faker” to refer to the artist that is responsible for the creation of an art object that is part of a fraud, especially when it is recognized that there may be a growing trend for the art fraud to be organized, on the one hand, by a fraudster who then, on the other hand, negotiates to find an artist willing to create the artworks (a “faker”) necessary for the fraud. In any case, those interested in the issue of non-authentic art will have to negotiate their way through language that at times may not correctly convey the exact legal meaning attributed to the art object in question.


  1. Ashok, R., Spring, M., & Plazzotta, C. (2004). Raphael’s early works in the national gallery: Paintings before Rome. London: National Gallery Technical Bulletin 25, Distributed by Yale University Press.Google Scholar
  2. Bailey, A. (1994). Responses to Rembrandt: Who painted the Polish Rider? A controversy considered. New York, NY: Timken Publishers.Google Scholar
  3. Blackman and Ors v Gant and Anor. (2010). VSC 109.Google Scholar
  4. Brown, G. (2017, April 28). Whiteley art fraud accused exonerated, p. 5, the news section of The Australian.Google Scholar
  5. Charney, N. (2015). The art of forgery. London: Phaidon Press.Google Scholar
  6. Cockington, J. (2016, August 24). Strong appreciation in Streeton works, pp. 2–3, the ‘Money’ section of The Age.Google Scholar
  7. Grishin, S. (2015, March 19). Is that a Whiteley? Why collectors buy Lousy Fakes as masterpieces. The Conversation. Retrieved October 10, 2016, from
  8. Hills, B. (2002, May 23). Judge gives go-ahead to sell 1000 fake paintings. The Sydney Morning Herald. Retrieved January 7, 2017, from
  9. Hoving, T. (1996). False impressions: The hunt for big-time art fakes. New York, NY: Touchstone.Google Scholar
  10. Keats, J. (2013). Why fakes are the great art of our age. Oxford: Oxford University Press.Google Scholar
  11. Kurz, O. (1967). Fakes (2nd ed.). New York, NY: Dover Publications.Google Scholar
  12. Lanham, D., Weinberg, M., Brown, K., & Ryan, G. (1987). Criminal fraud. Sydney: Law Book Company.Google Scholar
  13. McKenzie-Murray, M. (2016, May 21). Inside the Brett Whitely lavender bay fakes trial. The Age. Retrieved January 7, 2016, from
  14. National Gallery of Victoria (NGV). (2007, September/October). Not a forgery, but not a Van Gogh. Gallery Magazine, p. 10.Google Scholar
  15. Perkel, C. (2016, August 23). Painting owned by Canadian not by famous artist Peter Doig, judge rules. The Star. Retrieved August 26, 2016, from
  16. Pitman, J. (2006). The Raphael trail: The secret history of one of the world’s most precious works of art. London: Random House.Google Scholar
  17. Porter, L. (2007, December 2). Art imitating art. The Sunday Age, p. 18.Google Scholar
  18. R v Closs. (1858). 169 ER 1082.Google Scholar
  19. Salisbury, L., & Sujo, A. (2009) Provenance: How a Con Man and a Forger Rewrote the History of Modern Art. New York, NY: The Penguin Press.Google Scholar
  20. Shin, J. (2016, September 23). French firm to authenticate controversial Korean painting. Korea Bizwire. Retrieved September 24, 2016, from
  21. Sykes, T (2016, October 8). How Sotheby’s got hoodwinked: The art world’s $20m forgery scandal. The Daily Beast. Retrieved October 11, 2016, from
  22. The Rembrandt Research Project. (1983). The Burlington Magazine, 125(968), 661–663.Google Scholar
  23. US State Department (USD). (1998). Washington conference principles on Nazi-Confiscated art. USD. Retrieved January 7, 2016, from
  24. White, C. (2015). The Rembrandt project and its denouement. The Burlington Magazine, 157, 71–73.Google Scholar

Copyright information

© The Author(s) 2019

Authors and Affiliations

  • Kenneth Polk
    • 1
  • Duncan Chappell
    • 2
  1. 1.University of MelbourneMelbourneAustralia
  2. 2.University of SydneySydneyAustralia

Personalised recommendations