NEW YORK—Though Richard Serra’s retrospective doesn’t open at the Museum of Modern Art until June 3, his sculptures are so big, and so time-consuming to install [see video], that two of them, Intersection II (1992–93), which measures about 13 x 52 x 18 feet, and Torqued Ellipse IV (1998), at about 13 x 27 x 33 feet, have been standing in the MoMA sculpture garden for several weeks now, giving visitors something of a sneak preview of the show.

It’s interesting to see how people have been relating to them. For many visitors, the works seem to have taken on the character of playground entertainments: the four curving plates of Intersection II (1992–93) form a simple maze for people to walk through, and Torqued Ellipse IV creates an enclosure with a single narrow opening that people take pleasure in occupying. Other visitors try to ignore the Serras altogether. They see them as frustrating obstacles, hard to walk around, casting unwelcome shadows in the normally sunny garden, and seeming to do nothing more than take up a lot of space.

Back in the 1970s, when Serra started making work of this size, there was talk that they might challenge the notion of what sculpture could be. Nowadays, of course, we know that a sculpture can be anything (and Serra himself told me in a recent interview that he is no longer interested in sculpture per se), but it occurs to me that the way people are responding to his works in MoMA’s garden brings us back to that question of what the basic requirements of a sculpture are.

The issue here is that the works are so enormous that you can’t see either as a single, separate, discrete thing unless you get a long way away from them, which at MoMA would mean retreating from the garden into the building to look down from the museum’s second-floor corridor. Add the fact that there is nothing in the work at the human scale to engage us, other than the chance patterning of their rusted surfaces, and you begin to understand why Serra’s art is so perplexing, and so powerful.  

Up at the Metropolitan Museum of Art, there are two related exhibitions of Frank Stella’s work, "Frank Stella on the Roof" and "Frank Stella: Painting into Architecture." Both are dreadful disappointments, and the reason is not hard to understand: by contrast with Serra’s work, the bigger Stella’s sculpture gets, the more unsatisfactory it becomes.

Chinese Pavilion (on the roof)—dated 2007 and termed “in progress”—is big by anybody’s standards. It’s made up of slender curved uprights that support a thick, heavy black tracery of curves and angles. At 15 x 34 x 30 feet, it is pretty much the same size as Serra’s Torqued Ellipse IV, but it is far less powerful. In part, this is a because of where it is. Up on the Met’s roof the Stella is no match for the vast city seen in every direction. But just as problematic is the fact that Stella wants his piece to be interesting in a way that Serra does not.

The scale of his curved tracery is pretty much human. You can stand within it and, if the security guard looks the other way, wrap your arms around bits of it. So, unlike the Serras, which are so uninteresting at human scale, the Stella invites us to examine it. And when we do, we discover its real weaknesses. The sculpture is too obviously a mock-up. Its finish is just horrible, and the careless variations in its surface—shiny and mass-produced here, tarry and worked-over there—along with the rough junctions between the components and the poorly fitting surface veneer means that it cannot possibly evoke the weight and density—the substance—that Stella intends. As a consequence the work's size appears flabby rather than impressive.

Stella attempts what are apparently two contradictory things at the same time—huge size and interest at human scale—but in the end achieves neither.

Just up the road from this sorry spectacle, at the Jewish Museum, a wonderful show of Louise Nevelson’s sculptures offers a lesson that this apparent contradiction need not necessarily be.