Sullivan's Eagle: Form and Function Artistically Considered

Function as code for formal preferences

Sullivan's specific explanations of how the "function" of the office building influenced the formal articulation of his elevations is rarely critiqued, in part because both critics and architects generally prefer to argue about the subjective merits of particular modes of expression rather than about the actual functionality of the works that they examine or design. Of course, there are exceptions. Christopher Alexander's A Pattern Language and Stewart Brand's How Buildings Learn are both concerned with the actual functionality of buildings, as are handbooks like Neufert's Architects' Data and Littlefield's Metric Handbook: Planning and Design Data.

Jordy's description of a technical sense is also imprecise. For example, in what sense does structural or mechanical form "follow" the "needs of the building"? In many cases, the "needs" (functions) of the building are generic enough to support rather "generic" structural or mechanical forms. On the one hand, many different "functions" could occur within a generic structural/mechanical scheme; on the other hand, many different structural/mechanical "forms" could accommodate a particular building's needs. In the final analysis, invoking "utilitarian" and "technical" functions to explain the form of buildings is often just a cynical maneuver to justify an expressive (formal) function.

Identifying a "function" provides the architectural form with a "reason" that allows both the immediate stylistic culture in which the work was conceived, as well as the overarching basis for its subjectivity—i.e., the never-ending and cut-throat competition that animates all artistic production—to escape scrutiny. Function, for both critics and architects, is most often a code word that allows entry into the subjective sphere of taste.

The Lamarckian fallacy

Sullivan's extrapolation of this Lamarckian model of species evolution—a model where form is alleged to evolve in order to satisfy a function previously determined—to the relationship between architectural form and function provided perceptive critics with an easy target. But it is also true that architectural theorists and historians were most interested in the formal consequences of Sullivan's theory and practice, and were therefore often willing to overlook what Sullivan actually thought about the "function" of tall office buildings. Thus, when Sullivan characterizes his division of a tall office building into three parts (top, middle, and bottom) as a formal response to a functional condition (form purportedly following function), the specific formal and functional consequences of this claim are generally not challenged.

On the other hand, and unlike the evolution of biological species, architectural form often really does "follow" function, just as Sullivan argued: Many parts of buildings are designed (i.e., their form is configured) in direct response to previously existing functional programmatic interests. A toilet is designed to accommodate a known function, as is a bathroom, or a kitchen. Handles are meant to be gripped by human hands; heights of surfaces are meant to allow washing, cutting and so on. Auditoria are configured so that humans can be efficiently packed into spaces where they can sit, watch, and listen. In other words, humans did not randomly stumble upon spaces configured like auditoria or bathrooms and then subsequently design or invent activities (e.g., lectures, musical performances, defecation) that would fit in such spaces.

However, just because the phrase "form follows function" accurately describes the process of architectural design in many cases, the evolutionary or co-evolutionary variant—where function follows or co-evolves with form—can also be found. For example, it is likely that the corner office, now a symbol of power and prestige within many corporate settings, was not invented in order to give form to that corporate function. Rather, the form of the corner office preceded that function. Buildings had corners long before those corners were "discovered" to have the potential for hierarchical significance because they afforded greater privacy, more windows (views in two directions), and greater size. In other words, the "corner" was a formal outcome, but not of the desire to satisfy the function of articulating a corporate hierarchy. Rather, it emerged as an inevitable byproduct of rectangular building geometry that, in turn, came about both because of rectangular street grids which created uniform—and rectilinear—land parcels, and construction techniques whose orthogonal geometries corresponded to floor plates spanning between parallel load-bearing masonry walls and girders or, later, skeletal steel matrices of columns, girders, and beams. In this case, the function (of identifying hierarchical patterns of spatial organization) followed the form (of the rectangular building).

Expression of internal function

Figure 3. Leonardo da Vinci: Anatomical Drawings from the Royal Library Windsor Castle, The Metropolitan Museum of Art (New York: 1983)

Even so, this translation of internal structure onto external "skin" only applies to material literally stretched, or painted, over an underlying form and no larger implications about the expression of internal functionality on outward appearance should be inferred. The visible human form provides very little information about internal organs or even the underlying structure of bones.

Sullivan claims that this "law" of nature—that "form ever follows function"—applies to inorganic as well as organic form. Yet the idea that inorganic assemblages of stone or water have an intrinsic "function" is teleological in the extreme. There is no consistent sense in which the underlying structure of inorganic matter expresses itself in outward appearance. Sullivan's superficial view of "ever-brooding hills" tells us nothing about the earth's geology, since the underlying geological structure is hidden by the external form. Of course, geologists familiar with particular features on the surface of the earth may be able to infer what lies beneath based on their expertise and study, just as Renaissance painters learned about the appearance of human form by dissecting bodies.

None of this, however, has anything to do with the claim that these forms "follow" function. Both the internal "content" and external form of natural things obey their own laws and, however interrelated the former's behavior may be to the latter's form and appearance, this interrelationship has no bearing on the expressive qualities of one with respect to the other.

Moreover, the obligation to express an object's internal structure on its surface is something that—contrary to Sullivan's poetic claims—has no basis in organic or inorganic conditions found in nature, and certainly cannot be validated by examining his own works of architecture. In fact, Sullivan's tall building designs represent only a small variation on the prevalent mode of expression that he found so abhorrent (Figure 4).

Figure 4. The Home Insurance Building by architect William Le Baron Jenney, 1885, is expressed as "stacked" buildings (left) while the Old Stock Exchange Building by Adler & Sullivan, 1894, has a tripartite formal arrangement with nine uninterrupted office floors sandwiched between a base and attic; Jenney's "wedding cake" façade may be taken to represent the type of tall building design that Sullivan reacted against.

The idea that either of the designs shown in Figure 4 is better or worse at "expressing" the functions occurring within cannot be sustained with a logical argument. Both buildings provide a matrix of windows that actually do have a necessary function, i.e., to provide light, air, and views for the variable occupancies contained within the buildings. Both buildings have gratuitous decorative elements on their façades that draw upon the subjective aesthetic frameworks valued by their architects. Beyond that, nothing useful can be concluded about the relationship of the building's form to its internal function. The "artistic" consideration of tall buildings that has evolved over more than a century following Sullivan's pronouncement on form and function offers nothing to support Sullivan's vision: Formal expression has remained stubbornly independent of the skyscraper's internal function.

Influence of Semper and Bötticher

Figure 5. Typical floor plans of Jenney's Home Insurance Building (left) and Sullivan's Old Stock Exchange Building (right), both built in Chicago at the end of the nineteenth century, both show conventional relationships among structural bay, windows, light shafts and courts, offices, corridors, elevators, and stairs shafts, all of which influence the actual functioning of the building In other words, it is not the actual functioning of the building that engages Sullivan's passion.

Thus, the problem with Sullivan's formulation about form following function is not that it isn't true (it is, in fact, often true for buildings) but that, on the one hand, its argument on the basis of biological and inorganic analogies is flawed and, on the other hand, it abstracts from many of the functions that actually influence architectural form. Most architectural functions, for Sullivan, are not worthy of the architect's intellectual attention, both because they are the routine "product of the speculator, the engineer, the builder" and because they are "crude" and "materialistic." Instead, one small aspect of functionality—in this case, the tripartite functional division of an office building—is singled out because it supports Sullivan's argument that tall buildings should look, well … like Sullivan wants them to look.

Endnotes