I have been thinking about this question since reading the special issue of ANY (7/8) on Colin Rowe, especially because I think of myself, at least on one level, as a formalist. But clearly there is disagreement and uncertainty today as to what formalism is, and the opinion of its relevance to an advanced architecture tends to be alternately noncommittal or pejorative. Moreover, there is obviously no consensus as to whether Rowe himself is a formalist. Peggy Deamer, who alone mentions the Russian formalists, comes closest to defining formalism in suprapersonal terms. These terms are relevant, though I would argue with some of the finer points of her explanation that seem to imply that strange-making, or defamiliarization, is more a psychological phenomenon than a formal one. As Victor Erlich points out in his classic study Russian Formalism: Theory and Doctrine, the Russian Formalists believed that "before trying to explain anything, one should find out what it is." And if this has a tautological ring to it, it is also rather nicely consistent with Rowe’s dedication to the idea of the scholar-as-detective, which Paulette Singley recalls in her fascinating article in the same issue. Anyway (as I believe Colin Rowe would say), I appreciate being provoked into re/investigating this question.
An important source for the definition of formalism is Rosalind Krauss’s brilliant essay in Houses of Cards, "Death of a Hermeneutic Phantom: Materialization of the Sign in the work of Peter Eisenman."
The essay firmly convinces the reader of five things: (1) 20th-century formalism had its origin in literary theory, specifically Russian formalism; (2) 20th-century formalism was therefore inextricably linked at its point of origin to the avant-garde, namely modernism; (3) 20th-century formalism was the "strategic conversion of transparency into opacity" (the former related to everything that was not art and the latter to everything that was) and relied on a taxonomy of devices for defamiliarizing the artistic object; (4) Rowe, who stressed that architecture is a form of text and concerned himself (with Robert Slutsky) with the issue of false versus a true transparency, is a formalist/modernist; (5) given this and given the fact that Eisenman’s House I and House II are paradigmatic examples of 20th-century formalism in architecture, formalism can’t be all bad. Moreover, Krauss’s identification of the transition from formalism to structuralism in Eisenman’s work—"the dispersal of unities into a field of differences"—is significant, for it introduces the question: Irrespective of what formalism may actually turn out to be, once we do find out what it is, is it still relevant? That is, if "post-formalism" is now in, isn’t formalism out? Can formalism really function today as an operative intellectual construct for an advanced architecture? I think yes.
If, as Alan Colquhoun writes, "the problem of architecture is part of a larger problem involving the whole notion of art," then the potential for meaningful formalist research in architecture is inexhaustible. It addresses the timeless problem of form and content—or, as is suggested by the chess analogy that fascinated both Victor Shklovsky and Ferdinand Saussure, the problem of move and meaning. Here, it is clear that the Russians working during the cultural upheaval of 1910s, whose principal organization was Opojaz (The Society for the Investigation of Literature), reinvented the nature of this relationship. Shklovsky’s attempt to define a non-objective literature in terms of devices and techniques applied to materials (which he initiated the same year, 1915, as Malevich’s revolutionary exhibition of non-objective paintings and publication of his manifesto "Living in a Non-Objective World") has come down to us as a dilemma of the unity versus the separability of form and content. Is form content? Do the elements of content have an independent existence, exempt from the adopted laws of aesthetic structure? Is there content—in a Mondrian painting, for example—that sustains what might be called the "truth" of the aesthetic object (so as to different it from a forgery), but that is categorically invisible?
Though Shklovsky tried to expose the fallacy of the notion of separable content, he was tripped up by the double problem of philosophical and semantic complexities and ultimately failed in his attempt to articulate a cogent, mature position on the issue. He thus made it possible to consider the problem of the unity of form and content under the rubric of formalism in two very different ways. As Erlich wrote: "Was he implying that all that matters in art is form, or was he simply saying that everything in the work of art is necessarily formed, i.e., organized for an esthetic purpose?" I am currently more interested in the latter proposition—but, to the degree that the architectural equivalent of forgery is avoided; I do not reject the validity of the former. I am also trying to sort out the degree to which my acceptance of this formalist position is really at odds with Meyer Schapiro's counter-argument against the unity of form and content. In "On Perfection, Coherence, and unity of Form and Content," Schapiro wrote: "In practice, form and content are separable for the artist who, in advance of the work, possesses a form in the habit of his style that is available to many contents and a conception of a subject or theme rich in meaning and open to varied treatment. In the process of realization these separable components of his project are made to interact, and in the finished work there arise unique qualities, both of form and meaning, as the offspring of this interaction, with many accords but also with qualities distinctive for each."
All architecture seems to be a conscious or unconscious commentary on this larger problem of art. So though it is a popular idea that formalism is to poetics as syntax is to meaning (see Mario Gandelsonas in ANY), I am persuaded by the Russians that not only is syntax simply one of the devices of art, but that formalism is not situated on one side of the virgule in the form/content, move/meaning dialect. Rather, this dialect is at the very center of formalism's philosophical construct.
I have begun to sort out formalism’s identity and relevance in the following way: as it has a descended from the post-cubist contemplations of the Russian literary avant-garde, formalism is not a sterile aesthetic purism, a narrowly focused, perhaps even formulaic, obsession with syntax or composition. Nor is it a simplistic "art for art’s sake" doctrine that promotes form over content, or even simply form as content. Instead, it is a far more serious, multivalent, and equivocal proposition. If it centers on three major ideas—devices, conventions, and density (opacity)—and if its goal is toward art as strange-making, which involves a formal procedure whereby the object is transferred to a sphere of new perception, then formalism asks us to consider the possibility of an architecture whose aim, like poetry, is "to make perceptible the texture of the world in all aspects," as Boris Mikhailovich Eikhenbaum asserted. And if poetry is "a complex transaction involving the semantic and morphological, as well as the phonetic, levels of language," as mature Russian formalism ultimately concluded, if the aesthetic dimension of art lies not in the absence of meaning but in multiplicity/density of meanings, then this recognition is still important for the contemplation of an advanced, poetic architecture today.
Moreover, the degree to which architecture of such multivalent density is realized raises the critical distinction between the visible and the visual. I would argue that the interrelation of the visible (the aesthetic) and the invisible (the poetic) constitutes the visual, and that formalism, therefore, ultimately addresses the problem of vision. As such, it questions what the critical intelligent eye sees. Mondrian and other De Stijl artists understood vision to be first, both optical and plastic (the latter refers to seeing underlying relations or abstraction), and second, the intellectual "seeing" of nonaesthetic ideas. This complex network of visible /invisible interrelationship calls to mind Rosalind Krauss’s image of "the infrastructure of vision."
Thus I would hold that an architecture may be said to be part of a 20th-century formalist enterprise to the degree that (1) the techniques and devices of the visible, as applied to materials (both two-dimensional graphic notations and three-dimensional constructions) comprise a rigorous suprapersonal system; (2) the result is an architectural object that is strange with respect to everyday building (if not also with respect to the current advanced style); (3) this strange-making comprises a dense "infrastructure of vision." In other words, modern formalism, dialectical in nature because it includes the problem of form/content, move/meaning, is an unfamiliar nexus of the aesthetic/poetic (morphologic/semantic) that operates at a heightened order of difficulty and multivalence and is ruled by the adopted conventions (i.e. the suprapersonal system) of aesthetic structure.
Perhaps 20th-century formalism is found somewhere in the visual infrastructure established by four familiar architectural paradigms: Le Corbusier’s Villa de Monzie at Garches, Giuseppe Terragni’s Danteum, Eisenman’s House I (and/or House II), and Michael Grave’s Snyderman House (as well as/better yet: Le Corbusier's Palace of the Assembly at Chandigarh). And though we will soon move into the 21st century, we are far from exhausting the lessons of these and similar formalist/modernist models.