In this paper I target the relationship between two prints that are roughly qualitatively identical and share a causal history. Is one an artwork if and only if the other is an artwork? To answer this, I propose two competing principles. The first claims that certain intentional relations must be shared by the prints (e.g., editioned prints vs. non-editioned prints). The second appeals only to minimal print ontology, claiming that the two prints need only be what I call 'relevantly similar' to one other. In the end, I endorse the second principle. There are no trumping features over and above relevant similarity, that is, for any pairwise comparison of relevantly similar prints, one print being an artwork is both necessary and sufficient for the other print being an artwork.
art, editions, ontology, practice, printmaking, prints
Printmaking as a skilled trade has been practiced for centuries, and for much of its history artists have employed printmaking practices or printmakers themselves to create print artworks ("fine prints"); this is especially so in the contemporary artworld. Despite this, printmaking has been given short shrift in contemporary philosophy of art. Other than brief treatment from Nelson Goodman and Nicholas Wolterstorff, inLanguages of Art and Works and Worlds of Art, respectively, printmaking has been afforded little philosophical attention. Such neglect is deplorable because print artwork enjoys widespread artworld presence and the nature of print artwork itself demands philosophical scrutiny. Print artwork has an interesting and prima facie unusual ontology¾prints are distinct, individual artworks, not reproductions of distinct, individual artworks. Print artworks, in addition to being individual artworks distinct from one another, can also bear interesting relations to other prints (e.g., tokens of the same process-type, being in the same edition). Unfortunately, determining what relations ought to count as substantive for print artwork is no easy task. This difficulty largely results from the artworld practice conforming only incongruously (often arbitrarily so) to standard printmaking practice. Perhaps philosophy can succeed where artworld practice has failed.
In this paper, I offer a coherent and plausible general principle governing the relationship between print artworks, or, more precisely, the relationship between two prints that are roughly qualitatively identical and share a causal history (i.e., they emanated from the same plate in the same manner—relief prints, lithography, intaglio). Only from such a principle can we then determine which relations ought to count as substantive as well as provide a principled answer to the question, "Is one an artwork if and only if the other is an artwork?" That is, I offer a general print-artwork principle that specifies the conditions under which, for any two prints, one is an artwork if and only if the other is an artwork. For example, if an artist prints an edition of twenty-five lithographs, and lithograph 1/25 is an artwork, are the other twenty-four lithographs likewise artworks, though all distinct, individual works? What factors, if any, either present or absent, could make it the case that only lithograph 1/25 is an artwork? What about non-editioned prints, such as artist proofs and trial proofs?
I argue that neither standard printmaking practice nor artworld practice can alone yield a workable principle out of which these questions can be coherently and consistently answered. I claim that certain intentional relations standard printmaking practice count as substantive for prints simpliciter shouldn't carry over as such for prints qua art. Moreover, artworld practice fails to be illuminating because artworld practice inconsistently counts those relations substantive. I do not purport to inform or correct standard printmaking practice. Rather, the view I defend should be seen, in part, as a view about how and to what degree standard printmaking practice ought to inform and correct current artworld practice.
2. Initial Remarks
I needn't be interested in what makes any particular print an artwork; I need only assume that a particular print is in fact an artwork (e.g., Sean Scully's BarcelonaDay, 2005, 31/40, aquatint on paper). Given this, the idea is to see what follows, given certain general principles, for relevantly related prints (e.g., Barcelona Day, 2005, 32/40, aquatint on paper), assess those results, and then decide which general principle to prefer. In the interest of simplicity and efficiency, the general principles I suggest operate only over roughly qualitatively identical prints, or, more precisely, what I call "relevantly similar" prints. For my purposes, two prints are relevantly similar to one another if and only if they share all constitutive appreciable properties in common in virtue of sharing a causal history. Two prints share a causal history if and only if they are printed from the same template (e.g., a particular etched copper plate), by the same process (e.g., intaglio), onto the same support (e.g., paper). I employ the term "constitutive appreciable properties" as a broad, theory-neutral stand-in for roughly those relevant descriptive physical (internal) features of the print (e.g., color, shape, size) as well as those relevant descriptive non-physical (relational) features (e.g., aesthetic, semantic, representational features) supervening on, and in part determined by, those relevant descriptive physical (internal) features of the print. The above, though broad, remains substantive, allowing the principles and examples I propose to be equally salient and applicable across disparate art theories.
3. Two Competing Practices
Of course, many current art forms have historical origins in a skilled trade, but printmaking is one of the few art forms that has remained largely that, a skilled trade. Most printmakers are not artists. Likewise, many print artists are not printmakers. Some of the most famous print artists (e.g., Andy Warhol, Kiki Smith, Robert Rauschenberg, Chuck Close) never do the printing themselves; they hire master printmakers. Moreover, printmaking isn't and never has been principally devoted to art. For every Kiki Smith woodcut or Chuck Close mezzotint printed, printmakers produce a million wedding invitations, movie posters, advertisements, and T-shirts. Given this, I take standard printmaking practice to be independent from the artworld. What it is for something to be a print simpliciter has much to do with standard printmaking practice but nothing to do with the artworld. The question then is whether standard printmaking practice has anything to do with prints as art. My focus is then on the intersection of the printmaking world and the artworld, that is, when artists themselves employ the techniques and processes of printmaking, or employ actual printmakers, to create print artworks. Presumably, artworld practice (at least of the responsible sort) with regard to print artworks has a prima facie commitment to conform to standard printmaking practice. The question is whether artworld practice need do so in all respects or just some, that is, how should standard printmaking practice constrain artworld practice? I assume that print artworks must preserve minimal print ontology (i.e., relevant similarity), but either this preservation is itself sufficient or additional distinctions must also carry over.
4. Two Competing Principles
Two general principles suggest themselves, out of which more detailed versions can later be formulated. Let's call the first Permissive Print Ontology (PPO), that minimal print ontology is a sufficient constraint on print artworks, and the second Restrictive Print Ontology (RPO), that certain relations in addition to minimal print ontology must constrain print artworks.
PPO: Necessarily, for any pair of prints where x and y are relevantly similar, x is an artwork if and only if y is an artwork.
RPO: Necessarily, for any pair of prints where x and y are relevantly similar and both bear intentional relation r to A where A is the author of x and y, x is an artwork if and only if y is an artwork.
Although RPO may be prima facie the more intuitive ontological model given facts about standard printmaking practice, I show that only PPO can survive scrutiny. The plausibility of RPO depends on the specification of relation r, and I argue that the only plausible ways to flesh out relation r actually make RPO unworkable and counterintuitive. I argue that not only does the specification of relation r in standard printmaking practice fail to coherently carry over to considerations of prints as artworks, but that no such relation over and above relevant similarity counts as substantive for print artworks. Moreover, PPO can, I claim, both account for the very intuitions that make RPO look prima facie attractive as well as avoid the problem cases RPO accrues.
5. The Case For RPO and Against PPO
PPO appears to be a relatively simple and intuitive principle. Assume that Warhol's Flowers (1970), 1/250 (screenprint on white paper) is a work of art. If Warhol's Flowers (1970), 2/250 (screenprint on white paper) looks the same as 1/250, and 2/250 looks the same as 1/250 because both 1/250 and 2/250 were printed from the same template in the same way, then of course 2/250 too is an artwork. Why would one need to appeal to anything else? Fixing qualitative facts about the prints and the printing process ensures that whatever makes 1/250 an artwork likewise makes 2/250 an artwork. Additionally, PPO acts as a buffer against swamp-prints (objects that just look the same) and forged prints (printed from a forged template).
While PPO handles intra-edition prints perfectly well, not all prints are editioned prints. Also notice that being an editioned print seems to entail a certain intentional relation shared by all the prints in the edition. If x is a print in edition E, then, minimally, A (where A is the author) intended that x be a print in edition E. Just as being an editioned print entails a certain intentional relation, so to do many varieties of non-editioned prints (those prints not counted as part of the edition). For example, some prints, both editioned and non-editioned, are intended as gifts, and so get marked "H.C." (hors de commerce, not for sale). Often the artist employs a masterprinter to print the edition, and permits the printer to keep a few non-editioned prints (printer's impressions). Trial proofs are those prints printed during the proofing process in order to examine and refine the image. Artist proofs are those proofs used as a printing guide and quality standard for the edition or those prints leftover from the printing of an edition.
For a print to be any one of the above kinds of non-editioned prints is, minimally, for that print to bear a certain intentional relation to the artist. One might plausibly think that relation bears upon whether or not that print is an artwork, despite that print being roughly qualitatively identical to and sharing a causal history with an editioned print artwork. PPO fails to distinguish between an editioned print and a non-editioned artist proof, printer's impression, or even a trial proof. The editioned prints are intended for public reception and consumption, while trial proofs, master proofs, and artist proofs are intended to be corrective tools, methods of calibration, and archival references rather than objects for critical reception or sale. One might claim that though these kinds of prints are importantly related to the final product (the print artwork), they are not themselves artworks, ¾despite being relevantly similar; this is perhaps analogous to relationship between rehearsals and performances. Since PPO clearly ignores wholesale what appear to be substantive relations, the presence or absence of which might plausibly trump considerations from relevant similarity, then so much the worse for PPO.
Furthermore, PPO can't account for the basic notion of permissibility, that is, PPO fails to distinguish between artist-sanctioned prints and unsanctioned prints. Imagine that an enterprising and devious art collector sneaks into an artist's studio and presses a few prints for his own collection. Is this person now an art thief? Did he steal an artwork or merely a relevantly similar print, guilty only of thieving paper on which he printed an impression relevantly similar to the prints that are in fact artworks? Further imagine that the artist had completed her or his edition and intended to destroy the plate (a common practice) the next morning. Or imagine that the thief made off with the plate itself, pressing thousands and thousands of prints from the stolen plate. Lastly, consider an instructively bizarre case. Imagine that our burglar isn't interested in art or its kin, only expensive tools and equipment. Unfortunately, our burglar is also a bungler. Upon entering the studio, he stumbles and, by sheer accident, crashes into the ink, paper, tarlatan cloth, copper etching and printing press such that the physical process of pressing a print occurs, resulting in a relevantly similar print. Spooked, he runs off with the resultant print. Did he abscond with an artwork or merely a fortuitously pressed non-art print relevantly similar to the prints that are in fact artworks?
Whether a print is licit or illicit depends upon what relation it bears to the artist. Licit prints are those prints brought into existence with the direct or indirect (explicit or implicit) permission of the artist; illicit prints are those lacking the artist's permission. Some might plausibly contend that being illicitly printed trumps considerations from relevant similarity. Should one be even slightly leery of regarding as artworks the thousands of prints illicitly pressed from the stolen plate, PPO will cease to be attractive. Likewise, PPO's apparent accommodation of accidental prints might reasonably be cause for concern. Accidental prints per se, no matter the degree of relevant similarity, aren't artworks, especially when, as featured in the bungling burglar case, the print is a non-intentional, accidental print. Notice that one needn't claim that accidental, non-intentional prints are impossible, only that, minimally, the kind of general principle under discussion must at least structurally accommodate the claim that a print being accidental trumps concerns from relevant similarity, which PPO apparently cannot do.
So far, three relations have been proposed for which, intuitively, a general principle should count as plausibly trumping relevant similarity: 1) what purpose the artist intended the print to serve (trial proof, master proof, artist proof); 2) a print's licitness or illicitness; and 3) whether the print is intentional or non-intentional. According to PPO, nothing can trump relevant similarity; relevant similarity exhausts the general principle. Recall that RPO really just is PPO with an additional requirement, that is both prints must bear intentional relation r to the artist. Nothing demands that RPO specify this relation in any one particular way, only that the relation can in fact be specified in such a way as to make RPO workable. Additionally, one needn't find all of the relations specified above compelling. As long as RPO can accommodate at least one of the relations found compelling, then insofar as one finds that relation compelling, one ought to prefer RPO to PPO. That's the point. The general argument for RPO is that most, if not all, will find at least one relation compelling, so most, if not all, ought to prefer RPO to PPO. If in addition to this, RPO appears supportive, or at least consistent with, standard printmaking practices, then again, so much the worse for PPO.
In the main, printmakers and denizens of the artworld alike regard editioned prints far differently than non-editioned prints. Master proofs typically receive rough treatment rather than a delicate touch, ending up ink-stained, tacked to a print shop wall, or they simply get destroyed. Trial proofs, at least those not destroyed, are seen as merely interesting historical items, snatches of the creative process behind the edition. Artist proofs get tucked away as records, archived epistemic access points. Clearly then, insofar as one sees printmaking practice as a rough and ready guide for print artwork ontology, RPO, both consistent with and supportive of standard printmaking practice, wins the day. Moreover, PPO claims not only that those relations counted substantive by standard printmaking practice fail to be substantive for print artworks but also that nothing over and above relevant similarity counts as substantive, thus rejecting considerations from both licitness and non-intentional printing. So, PPO appears bereft of even prima facie plausibility. Appearances, in this case, I contend, are deceiving.
6. PPO and Editioning
What exactly is motivating standard printmaking practice? The practice of editioning surely results from material and financial considerations (e.g., templates degrade and scarcity determines value). Many printmaking procedures are such that only a handful of relevantly similar prints may be culled from the plate. Gum printing is notorious for being so unpredictable that only a few relevantly similar prints can be reasonably expected. Even in the normally fruitful Intaglio technique, a drypoint on Plexiglas will yield only half a dozen relevantly similar prints. Moreover, printmakers rely financially on editioning prints; prints from smaller editions ceteris paribus are worth more money (can command more money) than those prints from larger or open editions. Often, material considerations aside, the relationship between monetary value and the demand determine the size of the edition. The practice of editioning prints, in the main, looks to be entirely motivated by material and financial factors (both historically and contemporarily).
Being so motivated, however, doesn't mean that PPO can blithely dismiss the practice of editioning prints. On the contrary, PPO, in its appeal to relevant similarity, looks to be at least consistent with, if not supportive of, the practice of editioning. Since the kind of materials and printing techniques employed largely determine the number of relevantly similar prints likely to result, then the number of relevantly similar prints likely to result acts as an upper bound for the edition. When printmakers, at least of the reasonable sort, decide to print an edition of n prints, they believe at least minimally that the printing process can reasonably yield at least n+1 relevantly similar prints; otherwise they invite disaster. The very notion of editioning, in order to be substantive, seems to depend on the notion of relevant similarity. So, PPO runs parallel with, rather than counter to, the practice of editioning.
Of course, one might argue that PPO nevertheless fails to distinguish between editioned prints, open or limited, and non-editioned prints, such as artist proofs and trial proofs. Whether or not a print is an artist proof or a trial proof or part of an open or limited edition is largely intention-determined, that is, what purpose the print was intended to serve, especially whether or not the print was intended for public reception. Surely such distinctions must be preserved when considering prints as artworks. Unlike RPO, PPO fails to preserve such distinctions, so RPO remains the clear choice. Notice, however, that PPO doesn't claim that such intentions are irrelevant to standard printmaking practice. PPO need only either claim that such intentions fail to have their relevancy carried over to prints as artworks or show that counting them as relevant to prints as artworks leads to counter-intuitive results.
7. PPO and the Purpose Relation
PPO does not claim that intentions do not matter tout court. One could easily hold PPO and also hold that intentions are descriptively necessary features of any artwork, prints or otherwise. A clever advocate of PPO need only claim that the relevant intentions are already in play prior to employing PPO, that is, whether a pair of prints share a causal history and are roughly qualitatively identical in virtue of that shared causal history is largely intention-determined. The particular plate material (copper, Plexiglas, wood), the particular process employed (intaglio, woodcut, lithography), the particular image on the plate, and the particular impressions printed are all largely intention-determined. PPO relies on this fact. PPO merely entails that further intentions over and above those involved in relevant similarity simply fail to figure in a print's being an artwork. An artist's intention that a print be an artist proof is no more relevant to that print being an artwork than is the artist's intention that a print be used as a napkin to absorb a spill. If that print comes out as an artwork, according to PPO, then the artist is using an artwork as an artist proof (or an artwork as a napkin). While some may perhaps find it distasteful to use an artwork as a napkin, surely employing an artwork as a guide to the production of relevantly similar artworks should be unproblematic. Again, PPO doesn't jettison intentions; PPO merely claims that intentions over and above those fixing relevant similarity count neither for nor against a print's being art.
Furthermore, appeals to artworld practice won't strengthen the case against PPO. Artworld practice is horribly inconsistent with respect to the treatment of non-editioned prints. While many artists archive their artist proofs, some artists, especially those employing printmaking techniques that yield few usable prints (e.g., gum printing or monotyping), happily display and sell any relevantly similar artist proof along with those in the edition, ¾especially in monotyping cases where the artist proof may be the only workable print. In fact, whether or not artist proofs or trial proofs get regarded as artworks often depends either on whether or not they are relevantly similar to those prints in the edition or on the presence or absence of any workable editioned print. Moreover, any artworld resistance met by non-editioned prints seems to be inversely proportional to the fame of the artist. Artist proofs and, even more strikingly, trial proofs of Warhol's Flowers (1970) get displayed, sold, and regarded just as those in the edition (though in most cases, but not all, they are more affordable than those in the edition).
The above examples shouldn't count as evidence for artworld practice endorsing PPO. On the contrary, most artist proofs and master proofs lie unseen in studio flat-files, and trial proofs often get destroyed. The point is that artworld practice, with regard to print artworks, inconsistently conforms to and frequently and arbitrarily departs from standard printmaking practice. Even were we to assume that all departures from standard printmaking practice are prima facie justified in virtue of the presence or absence of reception-intentions, there could nevertheless be a printmaking equivalent of Emily Dickinson or Franz Kafka¾an artist who never intends that her prints be displayed, sold, or even received by the artworld public. Just as those cases aren't worrisome for poetry and literature, they shouldn't trouble PPO. This all assumes either that non-editioned prints are never intended for reception or that non-editioned prints cannot coherently be intended for reception. Both assumptions, however, are quite clearly false. While being an artist proof entails a certain intentional relation, as the above examples amply demonstrate, being an artist proof needn't also entail the exclusion of reception-intentions.
Given all this, PPO shouldn't be burdened with underwriting wholesale artworld practice with regard to prints as artworks. Rather, PPO ought to correct artworld practice, telling us what features of standard printmaking practice count as substantive for prints as artworks. PPO claims that only relevant similarity matters, and, therefore, artworld practice ought to reflect this. Notice that PPO needn't also entail that a print's being an artist proof or a trial proof is irrelevant tout court. PPO claims that a print's being an artist proof or a trial proof simpliciter counts neither for nor against that print's being an artwork, allowing, at least in principle, for artworld practice to make substantive evaluative distinctions between editioned and non-editioned prints. Of course, one can still argue that artworld practice, if guided by PPO, would end up a counter-intuitive mess. On the contrary, I argue that artworld practice consistent with RPO is far more pernicious.
Assume that RPO has it right; that is, assume that the intentions behind artist proofs, trial proofs, and master proofs are able to trump considerations from relevant similarity. One of two things must be then be the case: either (1) artist proofs, for example, cannot be artworks or (2) artist proofs can be artworks, but what makes them artworks must be distinct from what makes relevantly similar editioned prints artworks, rendering inert any pairwise comparison. Consider an editioned print numbered 1/25 and a relevantly similar artist proof (A.P.) Assume for the sake of argument, that 1/25 is an artwork in virtue of possessing feature F. Since 1/25 and A.P. are relevantly similar, 1/25 has F if and only if A.P. has F, so A.P. also has F. If (1) is correct, then somehow in virtue of being an artist proof, A.P. gets disqualified from being art despite having F. If (2) is correct, then if A.P. is an artwork, it cannot be an artwork in virtue of having F, even though A.P. has F in the same way that 1/25 has F and having F is what makes 1/25 an artwork. So, if A.P. is an artwork, it can't be an artwork in virtue of having F. Given that the only relevant difference between A.P. and 1/25 is that A.P. is an artist proof and 1/25 is an editioned print, what then could possibly make A.P. an artwork that is distinct from what makes 1/25 an artwork? For (1) to be correct requires only that being an artist proof disqualifies A.P. from being art (e.g., being editioned is at least necessary). For (2) to be correct, however, requires being an artist proof to be sufficient for A.P.'s being art; ¾it couldn't be anything else. So, if RPO is correct, then either (1) or (2) is true. The latter looks wildly implausible, so it must be the former: artist proofs cannot be artworks.
Assuming this, imagine the following case. An artist gathers her editioned prints with the intention of numbering them 1 to 25. Unbeknownst to her, she has mixed up a trial proof with a print from the edition. She then numbers the trial proof "1/25" and labels the print from the edition "T.P." I assume that labeling a print "T.P." doesn't in fact then make that print a trial proof and that a trial proof labeled "1/25" is still a trial proof and not from the edition. Given this, should we regard the future buyer of "1/25" as an unfortunate RPO dupe who falsely believes that she has purchased an artwork? Should we regard the future buyer of "A.P." as unknowingly lucky for getting an otherwise expensive artwork for a bargain non-art price? Clearly such labeling mishaps shouldn't be relevant to prints as artworks. What matters for prints as artworks is that the prints are relevantly similar. That is, if "1/25" is relevantly similar to "T.P.," then if the editioned print is an artwork, then so too is the trial proof. PPO makes sense of this while RPO only makes a mess.
8. PPO and Licitness
Recall our enterprising print thief. Imagine that this time, the thief makes off with both plate and press. This thief, well-versed in printmaking techniques, proceeds to run off thousands of prints from the stolen plate ( assuming no plate degradation occurs). PPO tells us that as long as these prints are relevantly similar to those printed by the artist for an edition, if those editioned prints are artworks, then so too are those thousands illicitly printed. Intuition tells us, however, that the illicitness of those prints trumps considerations from relevant similarity. RPO accommodates this intuition, PPO does not, so RPO is preferable.
Again, appealing to artworld practices won't strengthen the case against PPO. If our intuitions track artworld practice and artworld practice is inconsistent, then we shouldn't be surprised when our intuitions begin to conflict. If intuition tells us that licitness matters in the thief case, then our intuition ought to tell us that licitness matters in analogous cases; otherwise, why think running afoul of this intuition counts against PPO. Consider prints made from a Rembrandt plate. The artworld distinguishes between "lifetime impressions" (those made during Rembrandt's lifetime) and "late impressions" (those made after his death), but this distinction serves only as a function of value rather than art status. Late impressions are regarded as artworks just as much as lifetime impressions, though less valuable in virtue of less provenance. Dozens and dozens of original Rembrandt plates still exist, and prints are still made from them, all presumably lacking Rembrandt's explicit or implicit permission. PPO, unlike RPO, allows for both the "life" and "late" impressions to be artworks.
Perhaps, in defense of RPO, someone might claim that "late" impressions could be artworks if, for instance, the printing was sanctioned by a legally recognized executor of Rembrandt's estate. Such a claim should appear prima facie absurd because it employs only the thinnest notion of licitness¾legal sanctioning¾and an even thinner notion of Rembrandt¾the executor of his estate. Again, imagine a Franz Kafka of printmaking commanding his printmaking Max Brod to destroy all his templates. The same notion of licitness is in play here as in the plate-thief's case. PPO rightly regards these intentions as superfluous precisely because all of the substantive intentions have already done the work required of them. In the Rembrandt case, Rembrandt's intentions clearly matter because the intentions fixing relevant similarity for the prints are Rembrandt's. The intentions of executors, curators, or plate thieves shouldn't figure in whether or not the prints are artworks.
Now imagine that our thief steals a Rembrandt plate, or steals the plates from our printmaking Max Brod. Just as we shouldn't think notions of legality and estate executors can substantively stand-in for robust licitness with regard to Rembrandt's intentions, we shouldn't likewise think that the resultant prints from the stolen Rembrandt plate fail to be artworks even though their licitly (legally) printed, relevantly similar cousins are. Although our thief may be forced to sell the Rembrandt prints on the black-market to avoid capture, the thief nevertheless is selling artworks, a fact her or his buyers, as art collectors, understand perfectly. Likewise for the original stolen plate case. Notice that none of these cases are forgery cases, that is, our thief isn't trying to pass off the artist's work as her own. If our thief runs a print, in the right sort of way, of a Rembrandt plate, then the result is a Rembrandt print, and Rembrandt prints are artworks. That's the point of lifting the Rembrandt plate in the first place. PPO captures this, RPO,(so described, cannot. What we ought to be worried about is prints from forged plates, prints from faded Rembrandt plates that have been re-etched, or just prints from faded Rembrandt plates. PPO needn't claim that prints in these cases satisfy relevant similarity.
9. PPO and Accidental Prints
Much the same can be said for accidental prints. Recall that the objection against PPO wasn't about purely accidental objects that look like prints, only appearing to be relevantly similar to other prints. The objection against PPO had to do with prints that were accidentally printed. If it turns out that something's being a print entails that it could not have been accidentally printed, then PPO has no worries since it ranges over only pairs of prints. Let's assume then for sake of argument that prints can be accidentally printed, and that being accidentally printed (non-intentionally printed) trumps considerations from relevant similarity such that one ought to prefer RPO. This objection really looks to be merely a subspecies of licit/illicit objection. I do think, however, a brief response instructive.
Imagine that Smith readies everything to do a large relief print. In order to accommodate the size, Smith decides to employ a large steamroller rather than a printing press. Having laid down the plate and paper, Smith goes to fetch the steamroller. Through sheer coincidence and faulty wiring, the steamroller starts up and lurches forward unmanned, running squarely and firmly over Smith's plate and paper, producing a print identical to the one Smith would have produced (ceteris paribus) were he to have piloted the steamroller. Smith should deem this a fortuitous accident rather than shake his fists at the heavens. What matters to Smith is that enough pressure gets exerted to transfer the ink adequately; that he drives the steamroller, his sister drives it, or it goes unmanned matters not to him insofar as printmaking is concerned. Moreover, we could imagine while Smith is futilely attempting to start the steamroller, a high school marching band unknowingly tramples over his plate and paper. As long as the marching band exerted enough pressure to transfer the ink, Smith should be satisfied. PPO needn't unnecessarily fine-grain causal history; PPO can retain its strength even while coarse-graining causal history to capture mere range of pressure-exertion as sufficient, for certain kinds of printmaking techniques. Again, certain relations that initially appear to be substantive for print artworks, upon closer scrutiny, fall away, leaving only relevant similarity. If only relevant similarity remains, then RPO too falls away, leaving only PPO.
Basic print ontology isn't itself a problem. What it is for two prints to be relevantly similar isn't itself a problem. The fact that prints can be artworks isn't itself a problem. The real problem lies with combining these so as to create an informative principle about the relationship between relevantly similar prints as artworks, a principle that tells us what conditions must be met such that, for any two relevantly similar prints, one is an artwork if and only if the other is an artwork. Standard printmaking practice alone can't ground such a principle. Neither can artworld practice because artworld practice is inconsistent with regard to the relevance to prints as artworks certain intentional relations standard printmaking practice counts substantive. To be consistent, artworld practice must either respect only relevant similarity (PPO) or must also count as substantive at least some intentional relation over above those involved in fixing relevant similarity (RPO). The persuasiveness and plausibility of RPO over PPO relies on there being a persuasive and plausible specification of intentional relation r, such that artworld practice consistent with RPO is preferable to artworld practice consistent with PPO. I have shown that there are no such intentional relations. PPO is the clear choice, and, therefore, relevant similarity is exhaustive.
Christy Mag Uidhir
Sage School of Philosophy, Cornell University
Ithaca, New York
Published February 26, 2009
I owe special thanks to Aaron Meskin, Derk Pereboom, Earl Conee, and the anonymous referees for their suggestions and criticisms and to print artists Jeffery Stone and Jamie Davis for sharing their printmaking knowledge.
 Nelson Goodman, Languages of Art (Indianapolis: Hackett Publishing 1976) and Nicholas Wolterstorff, Works and Worlds of Art (Oxford: Claredon Press 1980). For a rare contemporary example see John Dilworth's article "Pictorial Orientation Matters," British Journal of Aesthetics 23 (2003), 39-56.
 RPO differs from PPO in that RPO claims as additionally determinative some relational (intentional) property that does not supervene on, or is not in fact largely determined by, a relevant descriptive physical (internal) feature of the print. Also notice that I offer no definition of what it is to be an author (individual or collective). I need only appeal to common sense, and I leave it up to the reader to fill in the details.
 When artists themselves print editions, they almost never print more than the edition requires, as the process is really is labor-intensive. So the artist proof in this case functions as the master proof or Bon-a-Tirer proof ("good to print") and typically gets archived by the artist. Printmakers in the artist's employ, however, almost always print more than is required. Those leftover are returned to the artist and typically labeled "A.P." for artist proof. My arguments are such that which sense gets used needn't matter (unless otherwise specified).
 Notice that PPO deals perfectly well with extreme cases such as monotyping (transferring ink from a smooth surface to paper by pressing). The process behind monotyping yields a unique print; not enough ink is left on the smooth surface for another impression. Subsequent printing attempts result in vastly inferior and incomplete impressions, violating the relevant similarity condition set by PPO.
 Whether a print is an artist proof depends on a certain intentional relation, but presumably being an artist proof does not in fact supervene (or is not largely determined by) features internal to the print (in contrast to say a print's being a representation of Abraham Lincoln or having this or that aesthetic property).
 Notice that trial proofs in the main won't be problematic for PPO given that most trial proofs (save master proofs) likely fail to be relevantly similar to those in the edition. The same goes for the points made in Nigel Warburton's peripherally related article "Authentic Photographs," British Journal of Aesthetics 37 (1997), 129-37.
 Notice that PPO can account for the latter—art status may be overdetermined.
 I think that this result may have some interesting implications for other art forms. For example, I suppose one could argue for a PPO-style principle for performances (e.g., Permissive Performance Ontology) according to which if relevant similarity conditions are satisfied, rehearsals count as performances (or artworks). I make a similar claim about recordings in my article "Recordings as Performances," British Journal of Aesthetics 43 (2007). Of course, this is all for another project.