Artistic creativity and whimsy: a reply to Costello

In Immaterial: Rules in Contemporary Art (2022), I grapple with a shift in artistic practice that has accelerated over the past several decades: artists are increasingly intervening directly in matters of display and conservation that were previously fixed by convention, thereby creating new possibilities for what their works can be and for the kinds of experiences audiences can have. They do this by articulating rules for display, conservation, and participation: rules regarding the configuration of objects that may or must be displayed for a work to be realized, whether and how those objects may or must be conserved or replaced over time, and whether and how audience members may physically interact with the display and/or alter the objects. These rules, I hold, are partially (and sometimes wholly) constitutive of contemporary artworks. The practice of articulating rules is now advanced enough that rules serve as an artistic medium, a resource and framework within which contemporary artists can maneuver to express artistic content.

The articulation of rules, like the fabrication of objects, is a creative process that unfolds over time. During an artwork’s early life, when the artist is directly involved in constituting the displays and conserving the objects, the rules may be largely implicit. The artist may not document or outwardly articulate the rules that are driving specific decisions; indeed, they may not be fully resolved in the artist’s own mind, leaving some aspects of the work unresolved. In addition, the rules may evolve as a given display is being constituted: the artist may make creative decisions in the moment about how to enhance a key aesthetic value or respond to environmental conditions.

As these works enter collections, there is a need to formalize the rules so that the works can continue to be appropriately displayed and conserved without the artist’s involvement. When a work is acquired, the institution typically goes through a rigorous process of collecting information about the rules. This may involve written instructions, artist interviews, and working with the artist on installation to document their understanding of the rules. Artists may also be consulted at subsequent moments when their work is newly installed or requires a significant conservation treatment. In practice, this means that the artist’s process of specifying the rules, and thus determining some aspects of the artwork’s nature, may be intermittent and sometimes continues long after any physical fabrication is complete.

As Diarmuid Costello (2024) rightly observes, this is where things get messy. It’s all well and good to say that artists constitute their works by articulating rules, but when we look at the details of the events through which the rules are (allegedly) pinned down, we sometimes find evidence of “artistic whimsy,” to use Costello’s term. Costello worries that my view makes the very nature of the artwork implausibly dependent on off-the-cuff statements that don’t deserve to be taken so seriously, with the consequence that the artwork may change with every creative breeze that sweeps through the artist’s mind and issues in a chance utterance—and that this may undermine the integrity of the work. I appreciate the fact that Costello’s concern is grounded in the empirical details of how these interactions unfold: he looks closely at two examples of artists changing their views in real time.

Although I’ll introduce distinctions that might resolve Costello’s concern in some particular cases, to a significant extent my response will consist of bullet-biting. But I hope to show that the bullet is made from tasty candy that many of us should swallow readily.

The first case Costello discusses is Jan Dibbets’s 1969 work All shadows that occurred to me in . . . are marked with tape. A display of All shadows is created by using masking tape to mark the borders of sunlight entering the exhibition space through windows or doorways. The shadows are marked multiple times throughout the installation day as the sun moves across the sky, with the consequence that the display consists of a series of overlapping shapes.¹ The work involves no persistent physical objects; at the end of each display event the tape is removed and discarded. The rules for display are the only persistent element of the work.

This work has a fascinating history, as discussed by Sanneke Stigter (2015). It was first displayed in 1969, and then twice in 1970. The well-known collector Giuseppe Panza acquired all of Dibbets’s other works from one of the 1970 exhibitions, sight unseen, but All shadows was not included in the sale, perhaps because at that time such an ephemeral work was not thought of as a candidate for acquisition. Dibbets later displayed the work in 1981, 2005, and 2006.

In 2007, Stigter, then head conservator at the Kröller-Müller Museum in the Netherlands, learned from Dibbets that the work had never been sold, and the Kröller-Müller seized the chance to acquire this early work by an important Dutch contemporary artist. This spurred a 2009 installation during which the Kröller-Müller team, including Stigter, worked with Dibbets to formalize the rules for display so that the work could be installed in the future without Dibbets’s involvement.

By asking questions, Stigter shaped which rules Dibbets articulated for the work. Should the sunlight be marked only on the wall and floors, or should it also be marked where it was cast on the windowsills? Stigter observes, “Whereas the artist agreed to their inclusion, I am convinced that if we had not pointed out the sun on this spot, the area would not have been marked with tape” (2015, 112). After careful inspection of photographic documentation, she revived Dibbets’s memory of a detail that had been lost over the years—small white stickers marking the time of day at various moments during the installation—resulting in his declaration that the corresponding feature is permissible but not required (2015, 113). And, as Costello highlights, at one point the artist directly contradicts his own earlier statement. Having said that the shadows cast by furniture should not be documented with tape, he then sees “long narrow shadows cast by the antique display cases in the adjacent gallery” and states that these can be included, finally saying, “It is a bit to your own liking” (Stigter 2015, 110).

What should we make of the facts that (1) the content of the rules depends on which questions happen to be asked, (2) the content of the rules shifts over time, and (3) over a short period of time Dibbets reverses himself? While Costello worries about artistic whimsy and the undermining of the artist’s credibility, I see all of these as legitimate expressions of the artist’s creativity. Just as an artist may repaint a region of canvas multiple times and may be influenced by contingent matters such as a comment made in passing (or quite pointedly) by a curator, an artist may refine and even reverse earlier articulations of a rule for reasons having to do with aesthetic or artistic sensibilities and aims. The key difference is that the actual changes made to the painting have a concrete physical form and, by convention, cease when the work is acquired by an institution: the work is defined as finished, the artist is no longer invited to revise it, and any residual dissatisfaction is typically worked out on some other canvas. Rules, on the other hand, are revised through communication, and by convention this process may extend long after the work’s initial creation and sale. Stigter’s discussion provides direct documentation of Dibbets’s ongoing process of artistic decision-making and artwork creation.

My take on this differs in a key way from Costello’s: he describes Dibbets’s statements as having epistemic import, where I see them as work-constituting. Costello asks:

Should artists be taken as the final court of appeal in such cases irrespective of how unreliable they show themselves to be, or need they meet minimal standards of truth-telling, coherence and lack of self-contradiction that we might expect of any other epistemically and morally respectable form of testimony? (this volume)

But in my view, Dibbets is not here providing testimony. He is fashioning the rules that should be understood as work-constituting going forward, not simply offering a recollection. He clearly is not aiming to articulate rules that have governed every display from the outset: when he installed the work in 1981 he included shadows from other sculptures on display, but he saw this as the “least successful” display and now aims to articulate rules that will lead to better results (Stigter 2015, 110). He is engaged in active artistic decision-making and creation to articulate rules that lead to desired aesthetic effects.

But Costello’s remarks yield a second concern: namely, that allowing the artist’s incidental statements to carry such weight puts the integrity or value of the work under threat. Whereas the act of applying paint to canvas or chisel to stone involves an acknowledged physical commitment to altering the work, the act of saying, “Do it like this…. No, actually, do it like that,” seems perhaps too ephemeral to support the same level of commitment, and treating such statements as work-altering, especially long after the initial activity of creation, may undermine the integrity, value, or meanings of the work the artist originally made.

My response is twofold. First, social and institutional practices have emerged to shore up the treatment of such communications as work-constituting. Dibbets came to the Kröller-Müller to be interviewed and lead an installation of his work, with mutual understanding that this activity would generate official instructions for installing the work in the future. In other cases, artists fill out detailed questionnaires or provide many pages of elaborate diagrams and descriptions. Having reviewed many artwork files in a variety of museums, read accounts of these processes by conservation scholars, and talked to artists, curators, and conservators about their experiences, I have found that these communications are generally taken seriously by all parties as part of art-making.

Second, because the activity of articulating rules is work-constituting and not merely incidental, there is no substance, meaning, or value of the work independent of this activity. We can, it is true, offer a counterfactual assessment: how would the work’s nature, value, and potential for meaning be different had the artist articulated subtly or significantly different constitutive rules?² We can do the same for a painting or sculpture: how would the work’s aesthetic effects or meaning potential be altered if the artist had worked this region of the object a bit differently? Would it have been more coherent, or would a central idea have come through more clearly? We might feel strongly that there is room for improvement. But this does nothing to undermine the fact that the artist did make the work in a particular way, and the work has the values and meanings borne by its actual structure and not by some hypothetical alternative.

Costello highlights some of the details of these interactions to make the case that I am overstating the seriousness of the artists’ engagement, and thus overstating the claim that their statements are seen, even by themselves, as work-constituting. When Dibbets reverses himself on which shadows to include, and then says, “[i]t is a bit to your own liking,” isn’t he simply deflecting responsibility? But I understand this statement not as a throwing up of hands, but instead as the conscious articulation of a rule that delegates a degree of aesthetic decision-making to the installer. As I discuss in chapter 2 of Immaterial, rules for display lie on a spectrum, with very precise specifications at one end and wide latitude for the installers at the other. Ghanaian artist El Anatsui, for instance, invites the installation team to exercise creative autonomy rather than copying earlier displays or deferring to an understanding of the artist’s vision (Irvin 2022, 44ff). Dibbets has articulated a rule that installers should engage in aesthetic decision-making as they encounter the installation conditions.

Through engaging in dialogue about their works, installing them, and looking at the results of installation or conservation treatments, artists learn about the aesthetic and artistic effects and possibilities opened up by the rules they have articulated. This may lead to revisions in the rules, just as a composer might be moved to revise a composition after performing it or hearing it performed. Now: might an artist make bad decisions as part of this process? Might they be flighty, or impulsive; or quite serious, but in a way that undermines aspects of the work’s value and coherence? Certainly, just as a painter or a composer might. In my view, this does not prevent the artist’s decisions from being work-constituting. But it might motivate us to use a strategy that is available for rules (and musical compositions) though not for paintings: namely, to roll back to an earlier version rather than allowing the artist’s current articulation of the rules to carry the day.

As I discuss in “Museums and the Shaping of Contemporary Artworks” (2006), Canadian artist Jana Sterbak attempted to change the rules associated with her work I Want You to Feel the Way I Do . . . (The Dress) (1984–5) years after it was acquired by the National Gallery of Canada. National Gallery curator Diana Nemiroff rejected this change on the grounds that it weakened the work and undermined its connection to the historical moment of its creation. In my view Sterbak did create a new version of the work (which may or may not ever be exhibited) by articulating a different set of rules.³ However, the National Gallery was within its rights to decline to trade in the previous version for the new.⁴

So: the museum is not bound to comply with statements the artist makes long after the work’s initial creation or acquisition: they may have good reasons to stick with an earlier version. But I don’t see this strategy as desirable or viable in the case of All shadows. In 2009, Dibbets is engaged for the first time, at the museum’s invitation, in formalizing the rules that should be understood as constituting the work going forward. There is no clearly articulated earlier version that could be operationalized within the institution. I see Dibbets’s engagement with the Kröller-Müller as constituting a 2009 version of the work, but I don’t see his creative activity as objectionably impulsive or as degrading the quality of the work, and I thus see no grounds for rejecting the 2009 version (even if there were an earlier version to fall back on).

There are (at least) two other strategies an institution might use if the artist articulates a rule that seems artistically catastrophic or pragmatically inviable. The first is negotiation. The process whereby an artist articulates rules in an institutional context often involves back and forth: both the artist and the curator or conservator may make proposals, raise concerns, and weigh options. When the Dallas Museum of Art acquired the complex multimedia work Drum Solos by Brad Tucker, Tucker specified a robust rule for participation: audience members should be able to play the two unique records included in the work on the unique turntable. Because this created risk that essential objects would be damaged, he then agreed that it would be acceptable to limit participation to museum staff, and even at times to present the work with no participatory element (Irvin 2022, 186–7).

Another option the institution may consider is to violate the rules. While this is certainly not ideal and may be something of a nuclear option in dealing with the work of a living artist, there can be good reasons for an institution to mount a non-compliant display, to restrict audience participation, or to conserve objects in ways that don’t strictly comply with the work-constituting rules the artist has articulated. Just as an artwork from an earlier period may sometimes be exhibited despite physical damage, it may sometimes be necessary to mount a non-compliant display of a contemporary artwork for reasons having to do with safety, cost, or unforeseen changes in exhibition or object conditions. To the extent possible, the audience should be informed about the non-compliance and assisted in understanding what the experience of a compliant display would be like (Irvin 2022, 205–25).

We are thus not entirely hostage to an artist’s whimsy: a concern that the artist has weakened their work through changes in rules can be addressed through dialogue (somewhat as a patron or curator might give feedback on a painting in progress) or by mounting a non-compliant display. I don’t believe anything about the Dibbets case warrants using these strategies, but I hope it is helpful to know we have them in our back pocket.

Let us turn now to the other case Costello discusses, Sarah Sze’s Migrateurs (1997). Sze participated in the Artists Documentation Project, a series of in-depth interviews of contemporary artists performed by conservation scholar Carol Mancusi-Ungaro. Because contemporary artists often use everyday materials with variable lifespans and sometimes articulate rules for conservation as part of their artistic practice, longstanding conventions of artwork conservation can’t be straightforwardly applied to many contemporary works. Mancusi-Ungaro recognized that it was important to document how work-constituting objects should be conserved: for rules that shape the evolution of materials over time contribute to aesthetic effects and are actively used by many artists as an artistic resource.

Sze often makes large, complex sculptural installations, but Migrateurs is a small sculpture made up of diverse everyday materials, some of which will need to be occasionally replaced due to deterioration in color and form. Sze’s discussion with Mancusi-Ungaro is geared toward identifying which elements should be replaced and at what point in their evolution. Faded green Tic Tacs, Sze says, can be replaced to restore their color; moreover, they can be replaced either with new green Tic Tacs or with simulacra, since Sze consciously combines real and fake elements throughout her artistic practice. The sagging gum inside the chewing gum packets can be replaced with a stiffer material to maintain structural integrity, since the gum isn’t seen anyway. The cotton swab can be replaced if it starts to look scruffy. And then there is, as Costello notes, the matter of the aspirin. Some of the aspirin tablets are nearly pristine, while others have deteriorated markedly, to the point that Mancusi-Ungaro describes one of them as “just the skin.” Though Sze previously articulated a rule that elements should be replaced when they have badly deteriorated, she likes the look of the aspirin and ultimately states that it shouldn’t be replaced yet (Irvin 2022, 73). Costello asks, “does this bottom out in anything more principled than how the artist happens to feel about a particular aspirin on a particular day?” (this volume).

I argue that the answer is yes. Fine-grained decision-making about materials and their configuration and evolution, to achieve very precise formal and aesthetic effects, is Sze’s bread and butter (Irvin 2022, 73–5). Both rules and physical materials function as artistic medium, and it makes sense that Sze’s deployment of rules would be as quirky and specific as her use of materials. Like Dibbets, she is making aesthetic assessments in real time as particular considerations become salient, with an eye to indicating how a conservator might respond to conditions that arise in the future. Sze has repeatedly said that she wants to stimulate material curiosity in her audience, such that viewers come to have the sensibilities of conservators: noticing the different rates of decay of different materials, questioning whether an orange or a house plant is real or fake, inferring that certain objects must be receiving some sort of care (Irvin 2022, 72–3). The object, then, should show signs of having been cared for rather than allowed to fall decrepit, but this doesn’t mean that everything should be restored to a pristine original state. What better to stimulate material curiosity than an aspirin that has hollowed out but not yet collapsed?

The conservator now has a nuanced rule to work from, which acknowledges the need to exercise their own aesthetic sensibilities. The rule goes something like this: “replace it if it looks scruffy, but if you notice something really cool that reveals the passage of time and the potentiality of materials without compromising the overall structural integrity, let it stick around for a while.” Rules, then, can be quite complex and fine-grained, as befits an artistic medium. And the artist’s process of refining them can have the same spontaneity and responsiveness to aesthetic perception as the artist’s application of a paintbrush or chisel. To my mind, the fact that my account allows for such exercises of artistic creativity, even whimsy, is a feature, not a bug.

I’ve argued that Dibbets and Sze are engaged in reasonable acts of artistic decision-making and work creation. But of course, artists are sometimes truly unreasonable, and they may sometimes make terrible choices. Where the artist uses rules poorly, the work is unlikely to end up in a significant collection, just as a writer who uses words poorly is unlikely to be published by a major press. In my view this does not prevent the rules from being work-constituting; it simply prevents the work from being good. So my account allows for this form of artistic failure.

Does it allow for other forms of failure? Yes: the artist may fail to sanction the rule they intended if they misspeak, provide a misleading diagram, leave out a key detail, or otherwise communicate poorly. This may or may not amount to a defect in the work; sometimes accidents and mistakes are compatible with or even contribute to artistic merit. The artist may complete the work and later come to wish they had articulated the rules somewhat differently. Depending on the circumstances, this may or may not lead to an event in which the artist revises the work or produces a new version. But the mere fact of the artist’s dissatisfaction, or ours, does not undermine or change the rule.

I want to acknowledge two situations where we may not be able to rely on the artist’s pronouncements as fixing a rule. First, someone might engage an artist in an exercise of pure recollection about the rules constituting a work created decades earlier. (This is not how I understand the Dibbets case: I believe Dibbets is engaged in activity that is mutually understood as work-constituting, not merely reconstructive.) In such a case, I would see the artist’s pronouncements as having epistemic rather than ontological import, and to that extent it would be only one source of information about the rules constituting the earlier work. Some artists have detailed and accurate records and recollections of their earlier decision-making, while others may offer a revisionist history grounded in subsequent artistic sensibilities. We may be better off relying on sources like the artist’s writings at the time, the testimony and notes of studio assistants and curators, documentation of installations, and so forth.

Second, an artist might continually flip-flop or waffle on some matter in ways that defy the attribution of a settled rule. This connects to a broader issue, which is that the archive of documents and communications relevant to our grasp of the rules may be complex, and we may be left with an interpretative challenge of reconstructing the rules based on this archive, a matter I discuss at length in chapter 8 (2022). There is no settled notation for rules in contemporary art, and practices of collecting information are in flux and vary from institution to institution (though there have been some initiatives to systematize this process). Where the artist’s process of articulating the rules has been extended in time, unsystematic, and dispersed across various media of communication, there may be tensions or internal contradictions that cannot be settled by appeal to evidence of what the artist finally meant. The museum may have to resolve some matters that the artist left unresolved.

Fortunately, for matters having to do only with rules for display, a straightforward solution is often available: if the artist has continually flip-flopped between X and Y, the museum can treat the work as governed by a rule such that displays exhibiting either feature X or feature Y are permissible. More significant conundrums arise with rules for conservation, since these may have implications for the long-term disposition of unique objects constituting the work. Conservator and conservation scholar Hanna Hölling (2017) describes a case in which a physical component of Canopus, a work by multi-media artist Nam June Paik, was damaged when the work fell off the wall. The component was a mass-produced hubcap with a unique inscription. Because it was not feasible to restore the hubcap to an undamaged appearance, Hölling advocated acquiring a new one and replicating the artist’s inscription, on the grounds that Paik had often affirmed that various components of his works (though not this specific one) could be replaced. Her colleagues, with reasoning more grounded in longstanding conventions of object conservation, felt that this would undermine the material authenticity of the work. In my view, Paik’s work is indeterminate on the replaceability of the component in question given unrepairable damage, and an institution could justifiably go either way, with consequences for the work’s material nature (Irvin 2022, 180–1).

Similarly, at some point, a conservator will have to decide whether to replace some components of Sze’s Migrateurs, with Sze no longer available to consult. The matter may not be settled by the rules Sze has articulated and the examples she has weighed in on; the conservator will have to exercise their judgment about the aesthetic effects of various options, and the work’s future material nature will depend on the outcome.

At one point, Costello draws an analogy with philosophy: when whimsical artists change their minds, he suggests, it is rather like a “philosopher who declared ‘today, by p, I have decided to mean not-p.’” “One would,” Costello notes, “put very little store” in such an unserious thinker (this issue). I would revise the analogy. I believe Dibbets and Sze are more like the philosopher who argues for p and then, having entertained a compelling objection, revises their position to “p except in circumstances like C.” Since the circumstances can’t be exhaustively specified, whether the exception applies on a given occasion may be a matter for interpretation. We thus have an ineliminable element of aesthetic casuistry.

But other artists, I must admit, are more like the philosopher who continually vacillates between p and not-p, or who changes their view from p to not-p with seeming arbitrariness, or who says, “p, except, um, sometimes,” or who simply buries us in a mass of statements that are hard to untangle. Such mushy decision-making may be a significant weakness, though we may still value their works due to other strengths. I acknowledge that my account gives even these people extensive authority over the nature of their works, and I have no argument against any exasperation this may give rise to. But I hope I’ve given you a sense of the resources and strategies we have for dealing with such artistic whimsy.⁵

REFERENCES

Footnotes