Data Journeys in Art? Warranting and Witnessing the ‘Fake’ and the ‘Real’ in Art Authentication

This chapter approaches questions about data and data journeys by examining demonstrations of fakery and expertise in popular accounts by forgers and their pursuers. We examine how relations between tellers and audiences are configured – who can be trusted, and what can be relied on when it comes to knowing the real from the forged. The various ambivalences regarding the nature of art, of perception, and of expertise, as well as the ways in which moves and techniques (re)produce the expert-teller in fraught conditions, bring a shiftiness to the constitution of data and evidence in this domain. Taking our cue from STS scholarship on the fixation and circulation of visual evidence in scientific practice, we discuss moves and techniques that point to (particular) features of a work of art to resolve authenticity questions, as well as those that point away from and negate features. More though, at stake in the case of forgery is not just how individual objects get rendered discernible, but also whether there is anything to discern at all. This chapter examines how experts find a place to stand as they account for the potentially unfaithful objects under their care.


Introduction
The BBC's Antiques Roadshow is a programme that has, for decades, featured experts travelling from town to town to appraise artworks owned by the locals. It banks on the element of surprise: frequently, what owners believe or assume about their possessions is turned on its head in the process of expert appraisal. In the early days, Impressionist and Modern Art expert Philip Hook remembers being able to deliver delightful surprises regularly, by revealing bygone objects as prized treasures. Later, as "members of the public […] got more and more optimistic about their property" (Hook 2014, 263), the reversal went the other way, the surprise less pleasant: "'Value? Not very much, I'm afraid. But it's such an interesting thing. Take it home and enjoy it.'" (Hook 2014, 263).
The surprises and reversals on the Antiques Roadshow depend on experts recognizing objects for what they are, positioning them relative to other objects ('rare', 'common', 'exquisite', 'decorative', and so on), and assigning them prospective market value. The objects presented do not always make such identification and valuation easy. Akin to the way scientists review and process instrument traces or specimens (Lynch 1985;Amann and Knorr Cetina 1990; Halfmann this volume), a careful adjudication comes into play, a reckoning with the possibility that things may not be as they seem. Materials are read for their trustworthiness, and spoken of -as Martin and Lynch (2009, 262-263) have argued in relation to cell biology -"as agents of their own visibility and identity: as showing and hiding themselves; presenting deceptive appearances; obediently complying with procedures or remaining recalcitrant." One reason why art experts account for the possibility of deceptive appearances is the risk posed by forgeries -works intentionally designed to pass as those by a valued artist. Part of the job of appraisal is precisely to distinguish a genuine Constable from "a modern painting in the style of Constable which has been ovenbaked in order to produce an apparent early-nineteenth-century craquelure in the paint surface and then claimed as a genuine Constable" (Hook 2014, 212). The matter of art forgery is not always black and white: the lines between a fake and an unintentional misattribution, a fake and a heavily restored item, or a high-quality fake and a low-quality original, are in specific instances blurred (Jones 1990;Hook 2014). Yet because the determination of origins and authorship matters so greatly to the price an item can fetch on the market, the epistemic game of authentication pulls towards a binary: is it or isn't it… In this chapter, we complement this volume's analyses of data journeys in the sciences with an excursion into the efforts and complications of making features of artworks warrantable and witnessable in light of questions about authenticity. We thereby follow Steven Shapin's argument, in 'The sciences of subjectivity', that so-called subjective forms of knowledge-production merit study for how they are anchored and go beyond the idiosyncratic. Shapin observes, for instance, that much of the talk of wine connoisseurs, "is referential, that is, it points to characteristics in the wine that connoisseurs come to know about, and taste communities can and do coalesce around more or less stable way of designating these characteristics," (Shapin 2010, 178).
In art authentication, the stakes involved in distinguishing the 'fake' from the 'real' inform efforts to determine what's given and what can be relied upon in the face of possible ambiguity and deception. When concerns about authenticity emerge, claims about what is known and knowable, seen and seeable, for whom and when, cast artworks in the role of would-be 'data' to be mobilized as evidence for determining true origins. In this chapter we highlight some key dynamics and variations of such casting, and discuss the accountability relations between experts, audiences and works of art that are thereby enacted.
The following tour of some of the colourful characters, controversies and efforts at revelation in the artworld is based on published 'insider accounts', documentaries and news reports -materials aimed at inviting a broad audience into an appreciation of the ways artworks may be designed to deceive and how such deception may be detected. Adopting an agnostic perspective on the possibility of turning artworks into data that travel into public accessibility, we find in the metaphor of data journeys an impetus for exploring what gets made available to make sense of contested works of art, how such warrants for sense-making are circulated beyond the expert realm, and what sorts of complications arise. Sabina Leonelli's (2016, chapter 3) relational definition of data proposes we think of data as what can be circulated and exhibited to others in corroboration of claims; she also points to data becoming salient qua data in and through the material form or "packaging" of information. In the first part of the chapter, we discuss practices and complications of "packaging" visual difference so as to make fakery available for 'all to see'. In the second part, we discuss practices and complications of treating artworks as material traces of their own origins. As our focus is on the public face of authentication, we will pay close attention to what becomes witnessable and portable, and for whom, amidst attempts to recognize objects for 'what they really are'. We will also show how such attempts, in turn, cannot be divorced from those of producing and preserving art's market value.

Here, See
To begin then, how are artworks, or aspects of these, mobilized in relation to claims to knowledge? How does art get worked up such that what is passed off as one thing can be exposed -in ways warrantable and witnessable -as actually something else?
Let us consider one prominent attempt to both spot fakes and cement the status of the teller: Thomas Hoving's (1996) False Impressions: The Hunt for Big-Time Art Fakes. In it he asserts that as much as 40% of the works he examined as the director of the New York Metropolitan Museum of Art were phony or tantamount to being so. Yet dealers, collectors, curators and artists were said to be reluctant publicly to speak about this phenomenon and reveal the art world's seedy underbellysomething Hoving took upon himself to remedy.
False Impressions forwards itself as an authoritative guide to the techniques and machinations of faking and 'fakebusting' in a number of ways. It 'proves' itself by: First, detailing years of experience and seniority of Hoving as a well-placed insider. Second, the extent of examples that serve as data points for the argument. Forgery, after forgery, after forgery (after forgery…) is presented to the reader: a sixth century BC Greek bronze statue, a fourteenth century enamel plaque, a Renaissance cup, a twentieth century sketch, etc., etc., etc. in page after page after page. Third, the specificity of claims. Hoving names names: former colleagues' experiences and mishaps are detailed -friend, foe, and neutral person alike are identified. Such features shore up the book's said ability to spill the beans about "the real world of fakes." (Hoving 1996, 7).
Especially given the rampant and yet largely unremarked deception said to be afoot, much is at stake in how Hoving practically marshals visual materials in support of contentions about the status of specific objects. To understand how he does so, consider one example. In the main text of False Impressions, Hoving describes his immediate appreciation of two versions, one authentic and one a forger's copy, of a sketch of a boy called Henri Leroy by the nineteenth-century artist Jean-Baptiste-Camille Corot. Hoving claims he had no trouble spotting the fake: I, for one, instantly selected the drawing (figure 48) that combined an unmistakable heavyhanded and academician's touch […] with congeries of tiny mistakes, most of them in the rendering of the child's costume. [The] phony seemed all too obvious, far too plodding and deliberate for the nervous and carefree genius of Corot (Hoving 1996, 193).
These sketches, which Hoving found in a book by art forger Eric Hebborn, are reproduced in a black and white photographic insert in False Impressions. They are accompanied by a caption specifying which is which, followed by another pointer regarding the grounds for this determination: "The academic correctness gives the phony away." (Hoving 1996, insert, figures 15&16) Both the main text and the caption provide a firm upshot of what is on display. Despite Hoving's making a personal statement ("I, for one…"), readers are solicited plainly to 'see' for themselves certain observable features that make the imitation different from the original ("unmistakable heavy-handed and academician's touch", "mistakes"). They are instructed where to see some of these features (the child's costume) and given to understand that these are tell-tale anomalous giveaways ("the phony seemed all too obvious"). 1 Just as laboratory science proceeds on the assumption of the "in-principle distinguishability of 'natural' from 'constructed' objects" (Lynch 1985, 82), here the fake Corot is charged with exhibiting features of artificiality ("academic correctness") in a way the genuine 'specimen' does not. In Hoving's account, the distinction is delivered in the way his remarks work up the two sketches in relation to one another. This achieves a 'fixation of evidence' (Amann and Knorr Cetina 1990), whereby visual expectations regarding what fits the "nervous and carefree genius of Corot" are mobilized as the background against which tell-tale signs of fakery can be located.
Here we find the familiar dynamic of multiplying the witnessing experience (Shapin and Schaffer 1985, 25). Despite being clearly in a position of being presented to, in being shown the materials that gave rise to Hoving's determination, readers can warrant the outcome of the comparison to themselves. By getting his audience to engage with the perceivable difference that bears out his assessment, Hoving makes it at least notionally verifiable. The reader is invited to locate within the sketch designated as "the phony" the giveaway signs of its dubious status -and it is the comparison with the sketch designated as the Corot that makes for these signs' repeatable, widespread witnessability.
By evoking the relevant background for seeing the difference, the figurative gesturing Hoving does is more akin to an open palm that casts regard rather than an index figure that surgically singles out. This opening of palms is both a showing of what's at hand and a drawing of attention to who is providing this opening -that is, a gesturing back at the gesturer. The seeing enabled, then, warrants the fake at the same time as affirming that Hoving is in possession of the requisite knowledge and skills to know the significance of what is being displayed. It achieves a particular distribution of expertise, configuring the reader as one who does not know but can appreciate (cf Woolgar 1991). As an opening of palms, Hoving's consideration of the sketches is also notable for its non-exhaustiveness and generality. As a result, the gesturing sets up a situation in which readers' own efforts and abilities to see what is treated as plainly there for Hoving, are made accountable, as well.

See It, See It Not
Hoving's use of the sketches of Henri Leroy follows, as we have mentioned, an earlier such pairing by English forger Eric Hebborn in his autobiography Drawn to Trouble (1991). As part of his book, Hebborn too cites decades of professional experience, describes forgery, after forgery, after forgery (he crafted), and names names. And he too details a murky art world wherein the small-time scams of individual dealers and collectors complement the institutionalized dishonesty ingrained in the art trade.
As he claimed, anyway, Hebborn made it a point not to attribute his works. In his own dealings he simply offered the 'fortuitously found' works themselves (mainly old master drawings), letting those in the business of proffering attributions derive their own conclusions. Because of this practice, Hebborn argued that he did not delude; dealers, scholars, and collectors deluded themselves by seeing what they wanted to see. The frequency with which this happened he attributed to a number of factors, ignorance and greed among them, but also the mistaken belief that imitations or forged works can be straightforwardly identified for being of inferior aesthetic quality.
Let us consider then how such arguments inform Hebborn's presentation of the sketches of Henri Leroy. More directly than Hoving, and more playfully, too, Hebborn appeals to readers to get involved in appraisal by using their own eyes. The sketches are given side by side -the text reads: It might perhaps amuse you to test your own abilities as a connoisseur, and decide for yourself which of two photographs (Figs 48 and 49) represent a detail from the original. Even if you happen to be Joe Bloggs in person, you will still have a fifty-fifty chance of being right. Look carefully, take your time, and seek the hesitant line of the copyist as opposed to the strong sure line of Corot. The answer is given at the bottom of the page. (Hebborn 1991, 226) Having teased readers regarding their ability to spot the difference, but also given them information to check if they picked the right original -in tiny letters at the bottom of p.226: "Answer: Fig 48" -Hebborn then goes further: Now, having read the solution, look at the two drawings again and you will suddenly notice how poor my version is, how faulty the construction, how harsh the modeling, and all sorts of ghastly errors which escaped your notice before.
The guessing game from before here develops into distinct valuations. The tongue-in-cheek disparaging of the 'fake' that "suddenly" appears "poor" also prods readers to recognize the dependency of their seeing on their understanding of what they are looking at. When a work is branded a 'fake' the eye seeks features that confirm its inferior status.
In yet another twist though, Hebborn carries on from the previous text to state: But what if I should now tell you that the answer at the bottom of the page is wrong?
With this, the features just established to anchor the distinction threaten to become mirage-like. Overall, the side-by-side juxtaposition combined with the textual instruction solicits a comparison between the sketches that 'packages' perceivable difference into tell-tale signs, but ultimately in a way that renders these unreliable as evidence.
Didactic comparisons of the kind employed by Hoving and Hebborn are a familiar device to educate non-experts about fakes; they have, for example, been a staple feature of museum exhibitions about fake art dating back to the 1950s (Lenain 2011, 264;Casement 2015). Philosopher Nelson Goodman (1983) distils the utility of this technique by arguing that, for the novice, the side-by-side juxtaposition of original and forged works: (1) stands as evidence that there may be a difference between them that I can learn to receive, (2) assigns the present looking a role as training toward such a perceptual discrimination, and (3) makes consequent demands that modify and differentiate my present experience in looking at the two pictures.
Through the deliberate placement of works side-by-side, Goodman claims novices become aware of the possibilities for learning-to-distinguish; by implication, proficient viewers can confirm their skill.
As we have seen in the previous section, Hoving sets the stage for comparison much in the manner that Goodman describes. The sketch identified as the real Corot is established as the measure by which the flaws of the other become perceivable. The reader is thereby aided in locating Hoving's assessment about 'academic correctness' in the way the fake differs from the original. It is aid that is more akin to someone giving directions by waving their hand along a bearing rather than pointing to a dot on a map, but aid nonetheless. In Goodman's terminology, the reader is cued to (1) see for themselves that the two sketches are different; (2) make sense of that difference in terms of the expert's assessment of what this difference amounts to; and (3) appraise one as an original, whose features then are seeable as character-istic of Corot's style, and the other as a fake, whose features then are seeable as flawed imitations.
By contrast, if there is any training at all in the twists and turns of Hebborn's side-by-side game, it is to make readers aware of their susceptibility to priming. With his parody -the pointing to what "suddenly" becomes apparent when the answer has been revealed, and then the playful reversal -Hebborn drives home this point. When a work is made seeable as derivative of another, the perceivable difference between the two is cast in terms that makes the former seem inferior to the latter. Goodman's three features of the side-by-side technique are thereby questioned as enablers of learning. Readers are invited to recognize their own limits, and to become critical of the way a baseline for their seeing is provided. Hebborn doesn't contest that there are differences between the two sketches, but questions the way these are mobilized as tell-tale signs from which the matter of authenticity can straightforwardly be resolved. Difference may be witnessable, and may thereby be forwarded as 'data' for authentication, but its status as prospective evidence is shaky.
Hebborn's trickery seeks not only to confront his (lay) readers with their limitations, but also to undermine trust in expert determinations. A key refrain in his book is his assertion that experts had on several occasions mistaken his work for old master drawings, even as they would categorically deny that forged art can be of high artistic quality. His side-by-side game suggests that their confident claims and their efforts to anchor verdicts in specific features, may be built on quicksand. Yet, while on the one hand thus undermining expertise, Hebborn also relies on it to bolster his own status as a top-notch faker/artist: Just as there could be little satisfaction in scoring a goal in the absence of a goalkeeper, so it is that to sell a master drawing to someone lacking the necessary expertise to make a proper appraisal of it is at best a hollow victory. In other words, only the experts are worth fooling, and the greater the expert, the greater the satisfaction of deceiving him. (Hebborn 1991, 218) As a result, for wanting to uphold the notion of "proper appraisal" as well as cast scathing doubt on the claims and warrants put forth by authentication experts, Drawn to Trouble is pitched in an arguably tension ridden tone. Amidst the exposure (again and again) of misattributions, self-deceptions, and bias rampant in the art world, Hebborn reaffirms the tradecraft expertise of the fooled. A similar tension is evident in the fact that the quality of Hebborn's work is contested among art historians and connoisseurs, with some considering them to approximate 'perfect' fakes (Lenain 2011, 269), and others arguing that his fakes are not nearly as convincing as he claimed (Jones 1990, entry 257; Hoving above). The questions of which differences matter, how to demarcate what belongs and doesn't to an artist's visual signature, and who can reliably do so, thus add complexity to the assumption that there is something 'wrong' with the fake that, once located, becomes available for all to see.

In the Blink of an Eye
In the previous sections, consideration of the sketches of Henri Leroy provided an illustration of techniques through which fakery is made accountable and anchored in what is at hand, along with how this shores up the credibility of experts. The juxtaposition of the two sketches provided the basis for turning perceivable difference into 'data', for acknowledging how attributions direct the seeing of fakery, and for indicating how things can get, well, befuddling. Although Hoving and Hebborn differed in how they orientated to the sketches and what they demonstrated, for both the side-by-side placement opened the possibility, using the words of Cohen and Cohen (2012), for a kind of 'hot' authentication based on direct perception by viewers rather than the 'cool' authentication gained by reading expert pronunciations. In this sense, the visual 'data' warranting authenticity verdicts is made available to anyone.
In this section, we move from practices and complications of "packaging" visual difference to discussing how different approaches to authentication treat artworks as material traces of their own origins. We do so mainly on the basis of the well-known tale of Teri Horton, an American truck driver who tried to get a painting she had bought in a thrift shop for $5 authenticated as a Jackson Pollock. This story, particularly as told in the documentary film 'Who the #$&% Is Jackson Pollock?' by Harry Moses (2006), also allows us to begin to explore how attempts to warrant authenticity intersect with commercial stakes.
Moses' and other accounts of the Horton case (Cole 2004(Cole , 2006Hoving 2008;Grann 2010) feature two seemingly diametrically opposed forms of expertise that were brought to bear on the determination. The first is connoisseurship, focusing on the general stylistic impression of the painting and its resemblance (or lack thereof) to the general stylistic impression given off by Jackson Pollock's work. In the documentary 'Who the #$&% Is Jackson Pollock?' by Harry Moses, Thomas Hoving appears as the poster boy for this type of expertise, proffering a negative verdict for the painting: My instant impression, which I always write down, you know, the blink, the one-hundredth of a second impression, was: Neat. Dash. Compacted. Which is not good. He wasn't neat. He wasn't compacted. It's pretty. It's superficial and frivolous. And I don't believe it's a Jackson Pollock. (Moses 2006) His nemesis in the film is art forensics expert Peter Paul Biro, who represents the second type of expertise. In a scene shot within his lab, Biro explains how he was able, with microscopes and high-powered photographic equipment, to locate a fingerprint on the back of the painting, which he then successfully matched to another print lifted off a blue paint can in Jackson Pollock's studio in East Hampton, New York. Another expert named Andre Turcotte demonstrates the match by pointing to the bifurcation pattern on both prints, presenting viewers with an animation that purports to demonstrate the overlay point by point. As Biro argued elsewhere in relation to his technique: Connoisseurship relies on an expert's close comparisons of a given work with closely related examples in order to discern where it "belongs" in terms of place, date and maker. By applying forensic methods, the process of attribution takes a novel and remarkable turn: it can use the evidence of fingerprints to trace a work of art back literally to the artist's hand. In effect, when the paper trail is missing or broken, forensics can at times fill in the gap. (Biro 2010, 157) Biro's emphasis on material traceability represents an effort to make art into data that is markedly different from stylistic appraisal. In addition to his use of fingerprints, in the film Biro also explains how he matched gold particles found on Horton's painting with gold particles found on Pollock's studio floor. 2 In his efforts to construct material provenance trails, Biro's approach differs from the more customary way imaging technologies and materials science are drawn upon as a line of defense against forgery -namely by testing for 'deeper-layer' manipulation that does not show on the surface of the work, and for "glaring anachronisms of materials and technique" (Craddock 2009, 1). 3 Too, most notably in the work of Simon Cole (2004Cole ( , 2006, fingerprint expertise has been argued to be more similar to, than different from, connoisseurship in its reliance on comparative judgment. But Biro nevertheless is the poster boy for 'science' in 'Who the #$&% Is Jackson Pollock?', and the difference between his and Hoving's expertise appears large and unassailable. Connoisseurship appears highly inscrutable -certainly the caricatured way in which Hoving appears in the film makes him the epitome of an old-fashioned authority vested in the experience and trusted judgment of particular persons (privileged white males in particular), and offends modern sensibilities regarding the accountability of experts (Porter 1995). With Biro, on the other hand, the emphasis shifts from the expert to the evidence, in line with "the desire to democratize" art authentication by "scientificizing" it (Grann 2010). The making of new data from the work of art -fingerprints, paint sample readings -and the demonstration of technical methods designed to make extraction and comparison of such data systematic, replicable and verifiable (having set the terms for fallibility 4 ), correspond to cultural notions of objectivity in investigating material links between the art object and its maker. To Hoving's put-down that "scientists are very interesting but come after the true connoisseurs" Biro retorts that connoisseurs need to update their understanding of what is and isn't a Pollock based on the evidence he uncovered (Moses 2006). 'Who the #$&% Is Jackson Pollock?' ends with the matter unresolved but also leaves viewers shaking their heads at 'ivory-tower' connoisseurship that refuses to reckon with material findings.
Hoving's approach gets a more positive billing in Malcolm Gladwell's Blink: The Power of Thinking Without Thinking (2005). Far from anti-systematic, the "instant impression" is here presented as a valid, if not easily explicable, way to know. The 'blink' points to a special kind of learnt receptiveness, developed through deep study of an artist's oeuvre. The knowledge that something is off can manifest in impressions such Hoving's above, or bodily reactions such as feeling cold ("as though there was a glass between me and the work", said one connoisseur), repulsed or uncomfortable (Gladwell 2005, 5) -something in line with a "sixth sense" (Hoving 1996, 19). The appeal to this acquired sense makes credible certain felt intuitions that might otherwise be dismissed as idiosyncratic reasonings, personal hunches, etc. The inability to isolate and nominate particular features as the grounds for a 'fake' verdict then does not invalidate such a verdict. 5 To paraphrase Gladwell, it is possible to know without knowing why, to know without being able to articulate how.
In the way these different approaches to authentication position art, knowledge, and appraisal to the public, as Cole (2006) has commented, "the truth ultimately comes down to which expert you believe." That 'you' are not yourself an expert is thereby underlined. At the same time, from the perspective of the artwork under investigation, the question of data remains highly relevant. Hans-Jörg Rheinberger's (2011) distinction between 'materials', 'traces' and 'data' helps to outline this. The artwork is construed alternately as a material to be interacted with, from which traces can be generated -this is what Biro was doing in scanning Horton's painting and Pollock's lab for fingerprints and paint samples -and a complex trace in its own right, in its totality a manifestation of its own history of becoming -this is what Hoving was engaging with. For Rheinberger, writing about experimental practices in the life sciences, the conversion of 'trace' to 'data' is about storing the informationcontent of precarious organic traces for future retrieval and pattern recognition (similar to Leonelli's definition of data as that which has been organized for witnessing, circulation, retrieval). In our case, treating art as trace is central and endemic to the work of authentication, but converting traces into data is not. 'Blink' thinking proceeds on a different basis than the ability to pinpoint data that can be appreciated or reactivated as key ingredients for authenticity adjudication; the subjectivity of the appraiser and the materials on which the verdict is made are here much harder to separate (cf. Shapin 2010).
At the end of 'Who the #$&% Is Jackson Pollock?, Teri Horton has not succeeded in having her work included in Pollock's oeuvre, but she has received an offer for it, for $9 m. As one of her friends remarks in the film: "Horton brought this painting to life": an "ugly" thrift shop painting nobody much cared about has been upgraded to a possible Pollock, disputed and in limbo. The expert appraisals have helped propel, if not exactly a data journey, a journey of valuation composed of a great many moves, including: • A local art teacher pointing out to Teri that her painting 'could be' Pollock's work (prompting from her the question with the curse word that gave the documentary its name); • Teri and her son Bill's persistent efforts to find experts and brokers in the art world who would not dismiss the painting out of hand; • The technical investigation of Peter Paul Biro, which provided a turning point in giving the painting the 'weight' it needed to qualify as a possible Jackson Pollock; • The favourable testimonies of Nick Carone, a friend and contemporary of Pollock (who said the painting is technically consistent with how Pollock painted) and art forger John Myatt (who said he could not have forged this work).
In and through these moves, a once-unremarkable object is shifted into a realm of ambiguity. Such a shift is also evident in relation to the work's undocumented provenance, highlighted by experts in the film as a big problem. Initial dismissal out of hand -"there are no Jackson Pollocks in thrift stores" -gives way to ambiguity as we learn that Pollock apparently did throw away work he wasn't happy with. Allan Stone, an art dealer, did get a genuine Pollock out of the dumpster in East Hampton. Lee Krasner, Pollock's widow in charge of the inventory, may not have kept proper track of all paintings that left the studio, etc.
A particularly interesting additional character introduced in the film is Tod Volpe, an art dealer contacted by Horton after she read his book Framed: America's Art Dealer to the Stars Tells All. Volpe gets involved to try and "put money behind the painting", specifically by interesting a collective of Hollywood actors and Wall Street finance professionals to buy it from Horton as a way to improve its provenance. Backed by such august owners, it can then be sold again, with a muchenhanced exchange value. 6 In an analysis of the case, Tay Yong Chiang (2016) called this effort to make the painting attract money a form of "commercial proof". The term is significant for how it puts Volpe's work on par with the authentication work of connoisseurs and scientists as three possible modes of proving that shape the object's journey of valuation. New datapoints are added: dollars and owners' names. The creation of commercial proof reminds of the way auction houses like Sotheby's and Christie's perform themselves as "temple[s] of civilised style and judgment" (Lacey 1998, 3). 7 Through glossy catalogues, private viewings, staff in smart evening outfits, the presence and commentary of experts, the style and demeanour of the auctioneer, and the way items are prepared for their moment in the spotlight, objects attract hefty sums.
So a once-ignored thrift shop painting is brought to life through a combination of moves that include extracting material traces from it (Biro), constructing the possibility of genuineness on the basis of testimony (Carone, Myatt), and offering it as an investment (Volpe) so it can begin to circulate in the artworld proper, as artwork proper. The way these moves together warrant the possibility of genuineness shows the entanglement between efforts to know Horton's painting for 'what it really is' and efforts to build commercial success for it. Significant, too, is the way the film exhibits the mechanisms of which it tells. The idea for the documentary came from none other than Tod Volpe, who thought introducing the story to a broad audience would help build the painting's value, and who brought together the parties to make it happen. Clearly, the entanglement that shows up here between commercial stakes and authentication efforts -both put before the public's eye -complicates the common-sense notion that authentication precedes, and is a prerequisite for, sales. The casting of artworks in evidential roles intersects with their circulation in the market in a more complex fashion, as we will continue to unpack in the next section.

Is There Anything to See? Is There Anyone to See it?
Public demonstrations of the approaches by which works of art are made to speak to their own genuineness have their shadow side in the various ways that routes to knowing are presented as barred. The problem of revealing fakes is not only one of sizing up troublesome objects that trick our perceptions. It is also one of contending with a destabilizing doubt about the universe of objects deemed 'art' as well as the experts that speak for them. The knowing and knowability of art forgery entails a complex mix in which the truth is variously treated as available and elusive, publicly demonstrable and beyond simple verification, given up and held back.
Let's return to the writings of Hoving and Hebborn for examples of such oscillations.
Hoving moves from general arguments that fakes are easy to foil because there is always something that gives the fake away, to other general arguments that it may be difficult to know what to look for: [One] work of art can be proven a fake because the drapery is too nervous in style; another because the drapery is not nervous enough. A statue of the fourteenth century can be fake because it is too refined, too beautiful. Another statue of the same period can be condemned because it is not sweet and pretty enough. (Hoving 1996, 22) As well as appealing to 'blink' judgment as a basis for distinctions, Hoving also, at times anyway, appeals to the even less specifiable spiritual quality of art. As he writes, "I tend to look upon works of art as partly spiritual and mysterious and partly human and fragile. Their lofty nature helps me break free from the mundane" (Hoving 1996, 16). In contrast to this glimpse of the sublime, phony pieces are "nothing but mockeries, dead things" (Ibid., 333). In attributing such a "cult status" (Benjamin 1936) to art, the distinction between the fake and the real remains clear, but rather than located in visible material features it gets bound up in some way with that which transcends (and thereby also with the learned receptiveness and sensitivity of the connoisseur). Hebborn performs a similar slipperiness in denying the visibility of forgery in terms of lack, fault, or inferiority, while upholding a sense that there is 'good' and 'bad' art. The former is the domain in which he locates his own efforts, which he looks to genuine experts to confirm on the basis of "proper appraisal".
More broadly, the nature of False Impressions and Drawn to Trouble as exposés of the art market does not allow us to rest entirely assured that there is a way out from the havoc wreaked by ambiguities and close resemblances. The hidden nature of art forgery, the way in which it aims for close resemblance and for passing unnoticed, means that, as Sergio Sismondo has written for the practice of ghost management in medical publications, "we cannot tell how common it is from published exposés," (Sismondo 2007(Sismondo , 1429. Equally, discussions of the possibility of very convincing fakes that are hard to catch can be read as shoring up, but equally as destabilizing, the status of the teller by being self-serving without providing a stable reference point. Hoving attempts to uphold the notion that such reference points exist, while Hebborn embraces the slipperiness, acting as both exposer and trickster. These exposés then oscillate between upholding and destabilizing, for their audiences, the prospect of seeing and knowing, begging the question what there is to be witnessed in this space. Routes to knowing are also presented as obstructed in reports about the legal pressures that keep art experts from making public what they privately know. In its basic contours, the dilemma is not new (Easby and Colin 1968) -negative authenticity assessments destroy market value, making owners and dealers lose moneybut the large fortunes at stake in the art market now have made them more extreme. According to art lawyer Ronald Spencer, scholars are "nervous about taking a $500 fee and getting sued for $10m" (quoted in the Economist 2012). There are reports of experts refusing to make their doubts about new discoveries public, and of a scheduled debate about the authenticity of a set of Francis Bacon drawings being "cancelled a week before it was to have taken place […] due to 'the possibility of legal action'" (Economist 2012). The Andy Warhol Foundation dissolved its authentication board in 2012, citing the exorbitant costs of defending against legal challenges as the reason (Kinsella 2012). Authors of the catalogue raisonné, the authoritative list of works by a particular artist, report having received bribes and death threats (Cohen 2012;Economist 2012).
At the same time, getting proof of art forgery to hold up in court is difficult, as was shown in a recent case against two art dealers allegedly working for a crime syndicate that was flooding fake Russian modernist art into Germany: After five years of investigating the 1800-work collection in collaboration with more than 10 international experts […] authorities were ultimately unable to determine the authenticity of the bulk of the collection, after only four paintings were declared to be fakes. (Neuendorf 2018) In this case, warring art experts making opposite claims did not help, and the court seems to have put most stake in the scientific analysis of paint samples, to which, for reasons the account does not provide, only a fraction of the works in the collection were subjected.
The rules of what to do with art that is assessed as fake are also not clear cut: the works may be confiscated or -in rare cases -destroyed, they may be stamped or marked in some way, but also may simply be returned to a dealer or previous owner in exchange for restitution. So it is not unheard of to have artworks previously discredited as fakes resurface in the market after some time (Cohen 2012). Efforts to recognize a work for what it really is are hampered when it is difficult to mobilize evidence of fakery in a court of law, and comparatively easy to cut loose the ballast of unfavourable verdicts and re-enter an artwork into circulation.
Such pressures and troubles encountered by art experts in turn affect the market: in the absence of a catalogue raisonné or the possibility of expert certification, "savvy art-buyers" are reportedly "spending less than they otherwise would" (according to a source quoted in the Economist 2012). They also change the accountability relations between experts, audiences and works of art. Some institutions have decided to do away with the catalogue raisonné in favour of something less definitive. For certain artists' works, online repositories are being developed that allow collectors and others to make their own determinations. In the words of a representative of the foundation that maintains the estate of artist Alexander Calder: "You determine if your work is fake or not with the data we present" (Cohen 2012).
This last statement brings full circle this chapter's survey of efforts to make fake vs. real art witnessable and warrantable. It leaves the viewer to assemble the case, as experts and the organizations that employ them put materials on display but stay clear from making evidential arguments. Such an outcome reminds of dynamics of devolved judgment and tension-ridden witnessing when expert testimony is calibrated to lay juror assessment, prompted by the question of whether evidence can and should 'speak for itself' (Goodwin 1994;Jasanoff 1998).
Overall, the knowing and knowability of art forgery is subject to shifting orientations. Firm grounds for determination are both gestured towards and withdrawn, and those giving accounts of the world of fakes can, among other things, displace offering a definite depiction; display fact after fact to build a credible argument; defer to some individuals as authoritative experts; devolve meaning making to viewers; and indicate obstacles to being able to see what is shown. As part of these strategies, expertise is varyingly circulated around, shifted in a zero-sum fashion, mutated, or pushed on to elsewhere. As a result, readers or viewers are varyingly barred, invited, and demanded to partake in the process of sense making.

Conclusion: Varieties of Data and Journeys of Art
In Data-Centric Biology, Sabina Leonelli refers to Paul Edwards' (2010) discussion of "data wars" in climate science to distinguish two ways in which data are handled and valued in scientific research. In one model, associated with weather forecasting -where "original sensor data may or may not be stored; usually they are never used again" -"the idea of "raw data" is not highly valued and scientists tend to work with models of data built through statistical tools" (Leonelli 2016, 22). In the other model, associated with climate science, the collection and curation of diverse data supports work on "a variety of research questions", and the point is for the data to be 'there' and available to be accessed at different points in time.
What we have described may be a third variety of how data and data journeys feature in the production of knowledge, one that applies to instances where the stakes revolve around recognizing things for what they are, and assigning them their proper 'place'. Art authentication in this respect finds common ground with archaeology, which as Alison Wylie (this volume) asserts, "depends fundamentally on discerning the temporal structure of the material record of the cultural past." In both cases, inferences are made about the past from traces that endure into the present. Art authentication also finds common ground with forensics and medical diagnostics, which use the "clues" provided by what we can access here and now to make inferences about a there and then (Ginzburg 1989). The speculative point here is that data gain significance (or don't) insofar as they can be mobilized in relation to questions of origin, cause, perpetrator, instigating circumstances, etc.
Wylie shows such mobilization to be a "hard-won achievement", dependent on background assumptions, procedures of triangulation and the specific arguments archaeologists seek to make. In art authentication, too, efforts to determine and demonstrate what's given are informed by assumptions about art and authorship, about the craftiness of forgers (and the limits thereof), and, perhaps most fundamentally, about the artwork as a trace that, given the right approach, will speak to where it came from. The shiftiness we have documented in this chapter, especially in attempts to create public witnessability, resides in the varying ways these assumptions are embraced or contested. It also resides in how art is produced as covetable commodity and as investment, with works by popular artists fetching increasingly large sums as they change ownership. On the one hand, the determination and demarcation of what is and is not original has become more important as prices have risen; on the other hand the mere possibility of genuineness can spur financial speculation, and the pressure to avoid lawsuits has generated interesting readjustments of accountability relations between experts and art buyers ("You determine if your work is fake or not with the data we present.") As the epistemic challenge of authentication meets this commercial push-and-pull, what is extracted or pulled from the artwork, brought into view through comparison, gleaned through "blink" thinking, or added as information associated with the work, is attended to in various ways as secure or provisional.
More mobile than these data of different kinds, the examples in this chapter seem to suggest, is the artwork itself. What gets mobilized as evidence for inauthenticity at one point may not take the work out of circulation forever, and may be (temporarily) forgotten, marginalized, or erased. And yet, with the circulation of a work also circulates the possibility of revisiting it as evidence of its own origins. When, how and by whom that possibility is activated is circumstantial, but the importance of assigning artworks their proper "place" (in the double sense of origin as well as resale value) positions works in what we might call a permanent state of being proto-data.

Coda
It is worth noting that Hoving (at times anyway), Biro, and Hebborn all agree on the general availability of works of art to be open for inspection, despite differences in how they are made to speak and what they speak about. Whatever their varying moves, the underlying similarity those surveyed in this chapter share is the potential for discernment -if only we care to look properly.
And yet, despite the manner objects are positioned as sites available for "close looking, the making of fine distinctions" (Nagel 2004, section 13), some accounts of successful art forgery point in an opposite direction. British forger John Myatt, who was involved in the "the biggest art fraud of the 20th century", was known to have made poor-quality fakes for which he used "an easily detectable household emulsion paint developed in the mid-60s, decades after most of the paintings were supposed to have been executed. In some cases, he used K-Y Jelly as a medium to add body and fluidity to his brushstrokes" (Landesman 1999). American Mark Landis, a forger who successfully donated his works to prestigious museums, said of his method: I know everybody's heard about forgers that do all these complicated things with chemicals and what-have-you […] I don't have that kind of patience. I buy my supplies at Walmart or Woolworth -discount stores -and then I do it in an hour or two at most. If I can't get something done by the time a movie's over on TV, I'll give up on it. (quoted in Caffrey 2015) Far from doing their utmost to confound scrutinizing gazes or scientific probes, Landis and Myatt present forgery as superficial. The commonplace notion of the gaze of the discerning viewer, expert, etc. that needs to fooled is thereby rendered into a mere trope. Just as deceptive objects are not what they seem, neither might be the practices of deception.
Open Access This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.
The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.