Some architects such as Adolf Loos have excluded buildings from the domain of art, emphasizing on “function” as the distinguishing quality of architecture compared with other plastic arts: “So hätte also das Haus nichts mit Kunst zu tun und wäre die Architektur nicht unter die Künste einzureihen? Es ist so. Nur ein ganz kleiner Teil der Architektur gehört der Kunst an: Das Grabmal und das Denkmal. Alles andere, was einem Zweck dient, ist aus dem Reiche der Kunst auszuschalten.[1]” In this context, Loos defined tomb and monument as the pure architecture referring to their perceived lack of function enabling them to touch human beings in a way that artworks do.

From this non-artistic point of view, it can be argued that the evaluation of architectural works is strongly user-oriented, as buildings cannot exist independently of their users and any good architecture should be individually identified by its so-called “inhabitants”. Therefore the criteria for judging architecture are contingent to user’s interpretation and “function” is the most determined constituent element of architecture which ascribes meaning to the work. Functional analysis captures the essence of an architectural work in its materiality optimally satisfying user’s utilitarian “needs” and links conventional meaning to buildings. In this regard, architecture becomes a coherent entity, a unity where every element is perfect and nothing is redundant in it. But can we fully understand a building referring to its functionality and arrive at a correct interpretation by the physicality of its material properties? Is there really an objective property of a thing that constructs its meaning?

With an ambition surpassing even Loos’ argument, Hannes Meyer – the second director of the Bauhaus – showed a stronger affinity for function and defended willingly the facile performance as the essence of a building, articulating twelve concrete human needs for which architecture has to provide solutions: Sex life, Sleeping, Keeping Pets, Gardening, Care of the Body, Weather Protection, Domestic Hygiene, Car Maintenance, Cooking, Heating, Solar Exposure and Service.[2] Attributing a good architecture to what his client intends, Meyer undermined the dominant position of architects and severely curtailed architecture to a process of utilitarian organization.

But how far is the question of function relevant to the process of architectural evaluation? As a counterexample to what Loos and Meyer have proposed, Farnsworth House in Plano, Illinois (1945-1951) is, despite its general recognition among the most important monuments of all twentieth-century, an absolute failure in reference to user’s experience, affording functional satisfaction of Edith Farnsworth. Mies van der Rohe’s grandson wrote: “(The House) owes its stature as one of highlights of modern architecture to its spiritual rather than its functional values. … So unconventional is the house that every move and every activity in it assume an aesthetic quality which challenges behavior patterns in different surroundings.”[3] Edith Farnsworth, the owner of the house accused later Mies of fraud, alleging that he had falsely represented himself as a skilled, proficient and experienced architect.[4] Yet Farnsworth House is still a significant example of modern architecture memorized and widely admired by a large group of art historians and critics who grasped the value of the work in its context. The question arises over such an argument is quiet clear: Is an inhabitant really in position to judge the worth or quality of a building disregarding the higher theoretical, aesthetic, and philosophical aspects of architecture?

Sullivan’s coined phrase of “Form Follows Function” which predicates a modern building by its intended purpose, has largely penetrated throughout the physical realm of modern architecture making the borderline between function and architecture get somewhat blurred up to present time. But in case of Farnsworth House, functional concerns of the client are outstandingly repressed helping the outraged architect, coming down to the greatness of his design. Farnsworth house is a “pure” masterpiece without responding to the minimum need of its occupant for keeping a garbage can in her house.[5]

As we can see from the example, functional analysis cannot necessarily guarantee the legitimacy of any architectural reading. Function is neither the necessary (in case of tomb and monument) nor the sufficient condition (in case of Farnsworth House) of architecture and any question of function would rather result in defining architecture itself. How we define the concept of architecture determines how we define function. Thus, function can be in some cases more and in some other cases less important. For instance, Bruno Taut took the approach toward function for unjustified defending his decidedly non-functional projects as follows: “Yes, impractical and without utility! But have we become happy through utility? Always utility and utility, comfort, convenience – good food, culture – knife, fork, trains, toilets, and yet also – cannons, bombs, instruments of murder!”[6]

Moreover, functional elements are still partly unbeknownst to both architects and users of the buildings. There is no way to determine by observation whether a person has a particular need or not, then the concept of need is too ambiguous to end up with some objective functions for an architectural work. Inhabitants could acquaint with concepts such as comfort and convenience in different manners based upon their personal preferences. Hence, the chance for an architect to prescribe a singular design which could gratify the desire of its every user seems to be scarce. Christopher Alexander and Barry Poyner have argued, even if you could firstly define a functional programme and “state clearly what a building has to do, there is still no way of finding out what the building must be like to do it. The geometry of the building is still a matter for the designer’s intuition; the programme doesn’t help with the geometry. Secondly, even if you state clearly what the building has to do, there is no way of finding if this is what the building ought to do. It is possible to make up a very arbitrary programme for a building. There is, at present, no way of being sure that programmes are themselves not arbitrary; there is no way of testing what the programme says.”[7]

Function is a contingent matter and cannot develop once and for all solutions to the question of need. Nor architecture can be reduced to its performative use. Therefore, it is doubtful whether we can determine the value and significance of a building by evaluating its utilitarian activities and establish our interpretation on the concept of use. Once we acknowledge the deficiencies in clarifying the definition of function, we have to admit the failure of Hannes Meyer’s theory in turning architecture into a science by its functional means and retrieving information about the user’s intention. Furthermore, architecture as a scientific entity could be developed by skilful engineers rather than architects and artists.

Following Immanuel Kant’s aesthetics and it will to purity, some architects have made analogies between architecture and art bringing function to the center of their arguments. Being identified independently of user’s intention, Loos tried to distinguish architecture as an artwork by dispensing it from any ”Zweck”. Contrary to Loos, John Ruskin tried to appreciate the buildings as art by highlighting ornament as the essence of architecture. He believed that the most artistic part in architecture is the least necessary and the most dispensable element in it[8], writing: “I suppose no-one would call laws architectural which determine the height of a breastwork or the position of a bastion. But if to the stone facing of the bastion be added an unnecessary feature, as a cable moulding, that is architecture.”[9] In a similar vein, Kant had separated artworks from objects he categorized them preliminarily under the concept of having a particular end.  Art should set itself free of any utilitarian interest that involves an ulterior motive or a concept by which the object is evaluated. Self-subsisting beauties of this kind is the subject of pure art. Thus, two types of beauty are distinguished by Kant: Free beauty (freie Schönheit) which corresponds with the spectator’s pure judgment of taste and appendant beauty (anhängende Schönheit) which is ascribed to those objects like most architectural works described as beautiful under the condition of a definite concept. ”Flowers are free beauties of nature” because they appear to be “purposive without purpose” and nothing underlies our judgment calling a flower beautiful, “but the beauty of a building (such as church, palace, arsenal, or summer-house) presupposes a concept of the end that defines what the thing has to be, and consequently a concept of its perfection; and is therefore merely appendant beauty.” [10]

Architecture conspicuously keeps distance from pure art according to Kantian notion of beauty unless it could liberate itself from its so-called users. Whatever a user intends, including the will to fulfillment of his/her needs, would substitute transcendent reality of architecture with a physical one. In other words, function will contaminate user’s aesthetic judgment about a building by the concept of its purpose.

Ruskin strongly believed Kant saying: “Much might be added to a building that would immediately please the eye, were it is not intended for a church.”[11] Any exclusive contribution to functional requirements of a building hinders purity of user’s aesthetic experience.

To see if a correlation between art and architecture exists, one should examine limits of independency architecture represents. In other words, we might focus on things architecture has in common with other arts, rather than things like function which is subject to their differentiation. Thus, for truly constructing an architectural work, a functional analysis seems to be irrelevant. Instead of linking architecture to function by constructing a causal explanation concluding function as the cause, it is more reasonable to make our predication in favor of architecture, for the building cannot  be reduced to its functional contingent construction and function in turn might be constructed by the buildings. Moreover, in case a particular function is abandoned in a building it could be realized in the next building and the idea of function as the immutable essence of architecture cannot work in every situation.

To sum up, human interpretations in terms of functional aspects of a building cannot take a satisfactory position in evaluation of architectural works. The effect a building has on its inhabitant is too variable to be ever measured experimentally and functional requirements of a building vary from one case to the other. Architecture is more a process of continual negotiation and therefore it should remain open in reference to every interaction between architect, his/her clients and the cultural context articulated by critics of the discipline. What goes on in these interactions would shape interpretations applicable for “constructing a building”. As a result, the criteria for judging architecture are contingent to all individual interpretations involved in designing, living and understanding a building.