Mario Carpo

Publication History:

“Craftsman to Draftsman: The Albertian Paradigm and the Modern Invention of Construction Drawings,” in The Working Drawing, edited by Annette Spiro and David Ganzoni, 278-281. Zurich: Park Books, 2013.  (The same book also published in German)  

The text posted here is a preprint draft and it is different from the published version. Please only cite from copy in print

At some point in the early 1960s the American philosopher Nelson Goodman came to the conclusion that not all arts are equal, and a crucial difference exists between autographic arts, where artists make their own works, and allographic arts, where artists script works that must then be performed or executed by others.  Allography is in some cases a practical necessity: as no musician can play one hundred instruments at the same time, a symphony must be scripted so a whole orchestra may perform it.  The composer may still direct it, but the musical score is supposed to function even in the absence of its author—for example, after her or his death.  The main purpose of such scripts or notations, as Goodman called them, is to convey the same instructions to all, at all times and without ambiguity, so the executed work of art will be just what the artist had in mind, even when the artist is not there to illustrate or explain it, assisting and instructing each performer in person.  Notational arts famously require a great amount of learned skills from all parties involved.  Everyone today is more or less familiar with the thousand minute signs that populate a musical score, but many who did not study ballet at school are often surprised to learn that the infinite seamless movement of a solo dancer, or of a whole corps de ballet, can be almost as precisely recorded and enacted from script—using either the Laban notation that Goodman was familiar with, or others.

Goodman appears to have been less interested in architecture than in other arts—his wife was a painter, and he himself was mostly interested in the performing arts, as well as a noted and quirky art collector.  In his book on the theory of art languages, first published in 1968, he briefly remarks that although modern architecture is mostly executed from blueprints and other construction drawings, modern buildings are so complex that they can hardly be built entirely from notations, or allographically—thus suggesting that, when building, so many things may happen and so many choices must be made that no drawing, however rich or complex, can hope to include or anticipate them all. (1)  No architect who ever tried to build will disagree.  Indeed, throughout most of its history, architecture was not a notational art at all: in antiquity and in the Middle Ages buildings were made by craftsmen—and in the Middle Ages, by free craftsmen, or guildsmen—who, like artisans of all times, were expected to think and invent, solving problems and finding solutions often on the fly, within the ambit of a loosely defined building program.  In the fourteenth century no gothic vault could have been built from drawings—first because no construction drawing of the time could have explained how to cut stone in 3D; second because, not surprisingly, medieval stonecutters did not use scaled drawings to learn or practice stereotomy.  Even though exceptions must be made for Greek and Roman classical building (where some identical shapes, profiles and moldings where reproduced on site from full-scale templates or 3D mock-ups), most pre-modern architectural drawings were used to communicate some visual aspects of a building to clients, commissioners or to the general public, not to convey technical aspects of the buildings to the workers who would build it.  Indeed, today’s construction drawings, a notational tool for the conveyance of technical information from designers to makers, are a modern cultural technology (Kulturtechnik) invented by Leon Battista Alberti in his treatise On Building (De Re Aedificatoria), composed around 1452.

For it is around that time that Italian humanists of the early Renaissance thought that buildings should no longer be made by artisan builders, but by a new kind of artist and thinker, of which Alberti saw himself as the prototype.  Unlike the medieval master-builder, the Albertian architect does not make buildings; he just makes drawings of buildings—in Goodman’s terms, notations.  When the drawings are finished, they are sent to the workers.  They do the actual work—they cut stone, lay bricks and saw timber.  In Alberti’s famous definition (De Re Aed., I,1,2), architecture is conceived in the mind, expressed through drawings and models, then executed—but executed by others. As a corollary to this principle, the architect’s drawings must be complete and exhaustive, and must contain nothing more and nothing less than the physical building in its entirety.

Alberti’s novel way of making is in many ways a foundational paradigm of modernity: unlike the medieval craftsman, who was a maker and thinker in one, the modern thinker is not allowed to make, and the modern makers are not allowed to think.  Design, not making, carries all the intellectual added value in the modern world: thus a car is an act of design, and it is owned, intellectually, by the designers who invented it, not by the workers, and increasingly by the robots, that manufacture it.  Today, not only pharmaceutical drugs, but even soft drinks and cookies are acts of design—they exist first and foremost as formulas, or notations, irrespective of the factories around the world where Aspirin, Coca-Cola or Butterkeks are mass-produced, paying royalties to their respective authors and intellectual owners.  In architecture, construction drawings are the keystone of this new way of building, and the indispensable vehicle and tool for its implementation.  For the Albertian paradigm presupposes that the entire building, as built, should be the identical materialization of an act of design that must have been fully and entirely expressed by the author’s original notations.  If something went amiss or was added beyond the author’s original script, then the building is no longer the result of that design, and the designer is no longer the author of that building.  In short, let only one discrepancy occur between the author’s idea and the building as built, and the whole system breaks down.

Alberti knew that his whole theory of design was dependant upon functioning notations—drawings that designers could safely use to communicate with builders, and which builders could understand and would abide by.  He also knew that no such drawings existed at his time, hence he insisted that designers should only use a particular kind of pictures, which today we would call orthographic, scaled drawings in plan, elevation and side views (De Re Aed., II,1,4).  Unfortunately parallel (Mongeian) projections did not yet exist at the time, hence Alberti could not provide a simple geometrical definition for his new format of architectural notations.  Moreover, Alberti soon found out, to his detriment, that no construction worker of his time would build by notation—i.e., following the drawn instructions sent in by a remote and absentee designer.  That was simply not the way of building at the end of the Middle Ages, when workers mostly did what they were told to do by someone who oversaw them on site and in person.  To that end, oral instructions and viva voce argument was all that was needed—supplemented by contracts in writing when monies were at stake. 

It took a few centuries, as we know, for Alberti’s paradigm to be fully accepted, both technically and socially. (2)  Today, Alberti’s paradigm is the basis of the architectural profession around the world: it defines architecture as a global art of design, and is enshrined in the laws and customs of almost all countries—including many that never went through a Humanist revolution.  With it, the original paradox conspicuously inherent and embedded in Alberti’s theory has equally pervaded the theory and practice of all modern arts of design.  For it is evident—as it must have been to Alberti himself, and as it still was to Nelson Goodman half a century ago—that no construction drawing, no matter how rich, can ever hope to encapsulate all aspects of a physical object yet to be built.

Alberti was a visionary, and his theory was the expression of an overarching intellectual ambition.  As such, it has changed the history of building in the West—and more.  But if we translate that intellectual agenda into legal chapter and verse, and into the nuts and bolts of an actual design assignment, odd things start to happen.  If a building—any building that was built by design, and designed by notation, in the Albertian, Western way—must be the faithful copy and execution of its author’s drawings, that means that every atom of that building must have been notated by its designer.  Every atom?  And if that does not seem practical, where should the designer stop?  Should the designer choose the door knobs, the Venetian blinds, the glass of the window panes, the color of the carpets, the model and make of the electrical outlets?  And if not the designer, who else?  A designer could argue (and some have) that if you change the specifications of the glass in the windowpanes of a curtain wall facade, the building will no longer be the one she or he designed.  But which kind of construction drawing can indicate the architectural qualities of a certain make of tempered glass?  For that, other instruments should be used, as indeed today they often are.  In short, the ontological gap between design intentions, their notation through construction drawings, and their material implementation leaves an inevitable grey area of undecidability, argument, frustration, litigation and liability where all kind of ad-hoc personal interventions, approximations, improvisation, bullying, persuasion, implorations, machinations and subterfuge take the place of construction drawings and specifications, and haggling becomes the design instrument of choice.

Some star architects can, famously, control even the most minute aspects of a building—Mies van der Rohe was a case in point.  In the case of the Seagram, one feels that Mies’s office (in that instance, through the complicity of Phyllis Lambert, Mies’s inspired patroness) might as well have designed the haircut of each uniformed doorman standing at the entry.  Most practicing architects, however, do not have such power, and they can only maintain some hold over the endless minutia of actual building at the cost of a never-ending, titanic daily effort, which evidently contradicts both the spirit and the letter of the Albertian program: if all aspects of a building can and must be inscribed in a set of construction drawing, then, when the drawings are sealed and sent, the game should be over.  Yet as we all know, that game is never over.  Construction drawings are testimony to this paradox and conflict, and their opacities and ambiguities are the living proof of the original sin, so to speak, of the Albertian project, and of its indelible legacy: we make drawings to design things we know no drawing can and will ever control. 

In recent times, this Albertian paradox has migrated from paper to electronic media, and one of today’s most powerful tools for computer-aided design, known under the generic name of Building Information Modeling, or BIM, was developed specifically to facilitate the exchange of information between designers and makers.  The spirit of BIM posits that designers, builders, and theoretically other agents as well, such as customers or clients or users, should participate in the collaborative making of the digital model of a future building, and that contractors in particular, thanks to this new, interactive digital platform, may step in the design process from the very start, thus bridging the gap between design intentions and their execution.  Phillip Bernstein of Autodesk has recently suggested that this new participatory way of building invites a new business model as well as a new legal framework for project delivery, where authorship may no longer be the privilege and monopoly of traditional designers. (3) 

Given the unprecedented power of digital simulations, one may surmise that at some point virtual models may become perfect duplicates of, and substitutes for, the buildings they represent—embodying and enacting all and every aspect of them.  Their designers could then make a digital model just as builders would once have made an actual building, and the final translation from model to building would entail no intellectual (or informational) added value whatsoever.  As in Borges’s famous paradox of the map that becomes identical to the territory it portrays, this final culmination of the Albertian notational paradigm appears ontologically problematic.  Time will tell.  Meanwhile, regardless of the notational tools we use, physical or digital, the Albertian paradox will most likely live on for a while—and with it, all the messiness, uncertainty and drama that the notational way of building has engendered throughout the history of the design professions.

Footnotes  

1. Nelson Goodman, Languages of Art: An Approach to a Theory of Symbols (Indianapolis: Bobbs-Merrill, 1968; 2nd ed., 1976), 122, 218-221.

2. Mario Carpo, The Alphabet and the Algorithm (Cambridge, MA: MIT Press, 2011), 15-20, 71-79.  German transl., Alphabet und Algorithmus: Wie das Digitale die Architektur herausfordert. Transl. Joerg Gleiter, Jan Bovelet (Bielefeld: Transcript Verlag, 2012).

3. Phillip Bernstein, “A Way Forward? Integrated Project Delivery,” Harvard Design Magazine 32 (2010), 74–77.  Peggy Deamer and Phillip G. Bernstein, eds., Building (in) The Future: Recasting Labor in Architecture (New York: Princeton Architectural Press, 2010).  See esp. Bernstein’s essay, “Models for Practice: Past, Present, Future,” 191–98.