THE ALPHABET AND THE ALGORITHM

MARIO CARPO
2011

WRITING ARCHITECTURE SERIES

THE MIT PRESS

CAMBRIDGE, MASSACHUSETTS
LONDON, ENGLAND
SECTION OF CHAPTER 1: VARIABLE, IDENTICAL, DIFFERENTIAL

1.1 Architecture and the Identical Copy: Timelines

The history of architecture in the machine age is well known. As it has been written
and rewritten many times over by the militant historians of twentieth-century
modernism and by their followers, it is a tale of sin and redemption. Architecture
was slow in coming to terms with the industrial revolution. Throughout the
nineteenth century, most architects either ignored or reacted against the new
technologies of industrial mass production. Then came the pioneers of modern
architecture, and their wakeup call. As Le Corbusier and others began to claim in the
early twenties, mechanization was changing the world, and architecture had to rise
to the challenge. Architects should invent new architectural forms, made to measure
for the new tools of mechanical mass production; and town planners should invent
new urban forms, made to measure for the new tools of mechanical mass
transportation. For the rest of the twentieth century many architects and urbanists
did just that. Oddly, many architects and urbanists are still doing that right now, as
they ignore, or deny, that today's machines are no longer those that Le Corbusier and
his friends celebrated and sublimated almost a century ago.

Well before the industrial revolution, however, another mechanical revolution had
already changed the history of architecture. Printed books are a quintessentially
industrial product. They are mass-produced. Mass production generates economies
of scale, which makes them cheaper than manuscript copies. They are
standardized—each copy is the imprint on paper of the same mechanical matrix.
Early modern printed books were so much cheaper and better than coeval handmade
books that they soon replaced them in all markets, and the new architectural books
in print (manuals, treatises, pattern books, etc.) changed the course of architecture
first and foremost because of the printed images they contained. Before the invention
of print, manual copies of drawings were famously untrustworthy, and as a result,
images were seldom used, or altogether avoided, whenever precise copies were
required. In such cases, nonvisual media (alphabetical or alphanumerical) were
deemed safer. For centuries in the classical tradition (from antiquity to the Middle
Ages to the early Renaissance) most architectural descriptions were verbal, not
visual.

The advent of print reversed this relationship between text and images. All printed
images in the same print run are notionally the same, for all and in all places. 18
Both the makers and the users of images were quick to realize that, thanks to print,
technical information could be recorded and safely transmitted in new visual
formats. And a new architectural theory soon developed, made to measure for this
new technical condition. In a typical technocultural feedback loop, machine-made
identical copies prompted a cultural awareness of identical reproducibility: printed
images were put to task to illustrate visual models of famous buildings, old and new
alike, but also to disseminate new illustrated catalogs of architectural components,
both structural and decorative. These new models were deliberately designed for
identical copies in print, but in some cases also for reuse and replication in
architectural drawings, design, and buildings. The most successful spin-off of this
media revolution was the new “method” of the Renaissance architectural orders—
the first international style in the history of world architecture.

As I have recounted at length elsewhere, this was the new architecture of, and for,
the age of printing.19 The early modern making of the "typographical architect" left
an indelible mark on architecture, which henceforth has been permanently
confronted with the paradigm of exact repeatability. The only parts of the design
process that were actually machine-made and identically reproduced in the
Renaissance were the images printed in books; the architects' designs and
construction drawings were handmade, and so were the buildings themselves. But
the paradigm of identicality spilled over from books to visuality at large, and
prompted a culture of identical copies that became pervasive in the West well before
the industrial revolution, and the actual rise of mechanical mass production.
Standardized images preceded industrial assembly lines, and a culture of
standardized architecture was already well established at a time when all visually
standardized architectural parts (from moldings, columns, and capitals to windows,
chimneys, etc.) had to be carefully handcrafted in order to look identical to one
another.20 In the process, standardized images standardized the craftsman's gesture:
the free hands of artisans were coerced to iterate identical actions, working like
machines that at the time no one could imagine or presage—but which would
eventually come, churning out identical copies better and cheaper than any artisan
could or would. This is where modern Taylorism and mechanization took over, Le
Corbusier stepped in, and the second, and better-known part of the story began.

1.2 Allography and Notations

As it happens, at the very moment printed images were revolutionizing the

transmission of architectural models, another media revolution was crucially
changing the way architects work. Alongside the images of eminent buildings of the
past or present, and the new sets of ready-made visual models that would
characterize early modern architectural books in print, another class of architectural
drawings and models was fast rising to prominence: the project documents that
Renaissance architects produced in growing numbers and forwarded to increasingly
distant building sites—a physical distance that went hand in hand with the growing
intellectual and social estrangement between architects and builders. New
reproduction technologies were of no consequence for project drawings, as these
technical documents destined for builders were not meant to be mass -produced:
each drawing could be hand-drafted as precisely as needed before being shipped to
the site where it would be used, without any loss in precision or other risks that
would have come with copies. The only technological innovation in Renaissance
project drawings may well have been their very invention—or the invention of their
mode of use.

According to Nelson Goodman, all arts were born autographic— handmade by their
authors. Then, some arts became allographic: scripted by their authors in order to be
materially executed by others.21 When did architecture evolve from its pristine
autographic status as a craft (conceived and made by artisan builders) to its modern
allographic definition as an art (designed by one to be constructed by others)? The
traditional view, which attributes to early modern humanism the invention of the
modern architect, and of his new professional role, rests upon some famous
narratives: Brunelleschi's legendary struggle for the recognition of his role as the
sole conceiver and master of a major building program; Alberti's radical claim that
architects should be not makers but designers, and his definition of a modern
notational system of scaled architectural drawings in plan and elevation that were
the indispensable means to this end.

Counter to these clear-cut stereotypes, it is easy to point out that the separation
between design and building (and between designers and workers) is a matter of
degrees. Architectural notations of some kind have almost always existed. It seems
that at the beginning of historical time Egyptian architects already used fairly precise
architectural construction drawings.22 But the history of design processes in antiquity
is a difficult and controversial subject, as archaeological scholarship on the matter
must build on slender evidence. Indeed, the evidence is at times so thin that some
archaeologists have concluded that Greek architects of the classical age did not use
scaled drawings at all; other known ancient notational systems, such as textual
instructions, three-dimensional models, templates, or full-size diagrams, sometimes
incised on stones or walls or otherwise sketched on site, all imply or require some
presence of the designer on the site of construction. The use of scaled project
drawings would have arisen only in the Hellenistic period, alongside the growing
estrangement between designers and craftsmen that the introduction of a more
advanced notational system suggests.23 The controversy is compounded by the
ambiguity on this issue of the most important extant source, Vitruvius's treatise.
Vitruvius's famously obscure definitions of three kinds of architectural drawings
(ichnographic, orthographic, scaenographia) in his first book seem to take some
practice of architectural drawings for granted. Yet his own design method never
refers to, and does not require, any kind of scaled drawing.24

Beside archaeologists and classical scholars, medievalists have also weighed in on

the matter. Something similar to proportionally drawn plans and elevations can be
dated to the thirteenth century, and more convincingly to the fourteenth and fifteenth
centuries (the famous drawings from the workshop of Peter Parler in Prague are
coeval to Brunelleschi's work on the Florence dome). These, and other textual
documents, have led some to suggest that "construction by remote control" was
common among Gothic master builders, and that adequate notational tools, and
social practices, already existed to support such design methods well before, and
unrelated to, the new architectural theory of the Italian humanists. This thesis has
been corroborated by an unusual blunder by the eminent scholar Wolfgang Lotz,
who in his seminal 1956 essay on Renaissance architectural drawings misread a
crucial passage in Alberti's De re aedificatoria, wrongly concluding that Alberti
encouraged architects to draw in perspective, and that Raphael's “Letter to Leo X”
(1519), rather than Alberti's treatise, should be credited with the modern “definition
of the orthogonal projection.”25 Although Lotz eventually corrected himself, that
essay is one of the sources of a persistent tradition according to which the pictorially
oriented Renaissance architects of the South, far from having developed the "orthogo
nal" notational format, would in fact have delayed its rise due to their penchant for
perspectival, illusionistic, nontechnical drawings.26

Recent scholarship has pointed out that Raphael's passages on architectural drawings
are little more than an amplification of Alberti's theory on the matter, 27 but the idea
that "orthogonal projections" may have been invented by Gothic builders, or even
by Renaissance architects, is problematic on other counts. Orthogonal, or parallel
projections, as defined by Caspard Monge's descriptive geometry (1799), posit a
center of projection located at infinity (the only possible point of origin for rays,
beams, or vectors, that must all be parallel to each other on arrival): in today's
projective geometry, central and parallel projections differ only in that the projection
center is a proper point for the former, and an improper point (i.e., a point at infinity)
for the latter. The drawing of "orthogonal" ground plans may not require projections
of any kind, as the ground plan of a building may simply be seen as its imprint or
trace on a real site (if necessary, redrawn to scale). But "orthogonal" front views, or
elevations, are a trickier matter.

According to late medieval optics, and to Alberti's own geometrical perspective,

"orthogonal" front views would have required an observer's eye to be physically
pushed back to an infinite distance, which, as a Renaissance mathematician
famously remarked, is actually "nowhere."28 Late medieval and early modern
geometries, owing to their Aristotelian framework, did not allow for such insouciant
appropriations of infinity, nor could they supply the homogeneity and continuity of
space that parallel projections from infinity would have demanded. Piero della
Francesca drew at least one famous head in plan—actually, two plans, an elevation,
and a side view, where all the views are connected by parallel projection lines, more
than three centuries before Monge's Descriptive Geometry.29 Likewise. late
medieval and Renaissance architects used, and depended upon, simpler sets of
"orthogonal" plans, elevations, and side views (and later, sections) which, however,
no mathematician at the time could have defined, nor formalized, for lack of any
workable notion of geometrical infinity. Parallel projections were for centuries a
practice without a theory.30

Yet in this matter too Alberti scored a major breakthrough. Alberti could, and did,
codify central projections, which represent infinity (as a vanishing point) without
defining it; but he could not codify parallel projections, which would have posited a
physical eye (the perspectival point of view) in a nonphysical place (infinity).
However, precisely because he had already defined central projections in his treatise
on painting, when a few years later he wrote his treatise on architecture Alberti could
for the first time ever lay out precisely what architects should not do: architects
should avoid perspective, as from foreshortened lines one cannot take precise
measurements (in Raphael's slightly later wording). As Alberti mandates in a key
passage in the second book of De re aedificatoria, architects' drawings, unlike
painters' perspectival views, require "consistent lines," "true angles," and "real
measurements, drawn to scale."31

One needs perspective to have been invented in order to tell architects not to use it.
As a side effect of his invention of geometrical perspective, Alberti could provide
the first (albeit negative) geometrical definition of modern proportional and
orthogonal plans and elevations—at a time when geometry did not allow for any
definition of parallel projections. This may appear to be a fine point of geometry
(and it is, as it is tantamount to defining parallel projections as "noncentral"
projections, without a corresponding center of projection at infinity); but, at a more
practical level, Alberti's strategy was also consistent with the basic need to explain
why scaled elevation drawings should not include foreshortened lines (as such
drawings often did before Alberti, and occasionally kept doing after him).

Alberti's distinction between building and design (linea menta) is spelled out in
various but unequivocal terms in the first, second, and ninth books of De re
aedificatoria, and it is one of the foundational principles of his entire architectural
theory.32 His new geometrical definition of architectural project drawings (and
models) provided a consistent set of notational tools suited to his new, allographic
way of building. As previously mentioned, the distance between designers and
building sites is an historical variable, and it ebbed and flowed for centuries before
(and after) Alberti's theoretical climax. With these ebbs and flows, the need for, and
availability of, reliable notational tools varied over time, and the evidence that
several ways of building by notation existed, and were variously implemented,
before Alberti should not be discounted. But, in addition to the sharpness of its
conceptual proclamation, the Albertian way differed from all precedents in another,
essential aspect—one that has stayed to this day.

1.3 Authorship

As Alberti repeatedly emphasizes in his treatise, architects must work with drawings
and three-dimensional models throughout the design process, as various aspects of
the project cannot be verified unless they are visualized.33 Drawings (albeit,
apparently, not models)34 will also be used as notational tools when the project is
finalized and construction drawings are sent to the builders; but the two functions,
visualization and notation, remain distinct. Designers first need drawings and models
to explore, nurture, and develop the idea of the building that, as Alberti states at the
outset of his treatise, is "conceived in the mind, made up of lines and angles, and
perfected in the learned intellect and imagination."35 Alberti insists that models
should also be used to consult experts and seek their advice; as revisions, corrections,
and new versions accumulate, the design changes over time; the whole project must
be examined and reexamined "not two, but three, four, seven, ten times, and taking
breaks in between."36 The final and definitive version is attained only when each part
has been so thoroughly examined that "any further addition, subtraction or change
could only be for the worse."37 This is when all revisions stop, and the final blueprint
(as we would have said until recently, both literally and figuratively) is handed over
to the builders. Thenceforward, no more changes may occur. The designer is no
longer allowed to change his mind, and builders are not expected to have opinions
on design matters. They must build the building as is—as it was designed and
notated.

At various times and in different contexts Alberti insists on this ideal point of no
return, where all design revisions should stop, and construction begin speedily and
without hesitation (and, he adds, without any variation or change during the course
of the works, regardless of who is in charge of the site).38 Alberti famously advised
architects against directing the actual construction: in his view, building should be
left to the workers and to their supervisors.39 He allows that “to have others' hands
execute what you have conceived in your mind is a toilsome business”,40 and indeed
documents related to the building history of the Tempio Malatestiano in Rimini 41
prove that the throes of allography did not fail to take a toll on Alberti's career as an
architect.
Local workers, craftsmen, and master builders might not have been easily persuaded
to comply with the drawings and models sent by an absentee designer. Building by
design was most likely not an absolute novelty for the craftsmen of the time.
Building by someone else's design may have been less common, but, crucially,
Alberti's new way of building left the builders no leeway. Craftsmen in Rimini
around the mid-fifteenth century may have resented Alberti's complaints as the
caprice of a scholar dabbling in construction matters. In fact, much more was at
stake.

At the close of the ninth book of De re aedificatoria Alberti muses that the lifespan
of major building programs may be longer than that of any architect, and that many
incidents may occur during the construction to alter or pervert the original de sign.
Yet, he concludes, "the author's original intentions" should always be upheld.42 This
remark is slipped in, inconspicuously and seemingly inadvertently, at the end of the
ninth book (which some scholars consider the real conclusion of the treatise, or at
least of its systematic, theoretical part). By bestowing upon the architect this
unprecedented "authorial" status, Alberti emphasizes the scope and ambition of his
new vision of the architect's work—but he also raises new questions concerning the
"author ship" of architecture's end product.

An original, autographic work (for example, a painting made and signed by the
artist's hand) is the unmediated making of its author. But in the Albertian, allographic
way of building the only work truly made by the author is the design of the
building—not the building itself, which by definition is made by others. The only
way for Alberti to claim an extension of authorship, so to speak, from the drawing
to the building was to require that the building and its design should be seen as
perfectly identical. This requirement, however, was bound to be difficult to enforce,
and technically problematic. Until very recent times, scaled models were not used
for notational purposes, due to the difficulty of measuring them (and Alberti himself
implies they should not be, presumably for the same reason). Alberti's apparent
preference for drawings, rather than models, as the primary notational tool marked
a significant departure from the late medieval tradition.43 But drawings, unlike
buildings and models, are two-dimensional, and in most cases proportionally smaller
than the building itself.44 Consequently, a building and its design can only be
notationally identical: their identicality depends on a notational system that
determines how to translate one into the other. When this condition of notational
identicality is satisfied, the author of the drawing becomes the author of the building,
and the architect can claim some form of ownership over a building which in most
cases he does not in fact own, and which he certainly did not build—indeed, which
he may never even have touched. The transition from Brunelleschi's artisanal
authorship ("this building is mine because I made it") to Alberti's intellectual
authorship ("this building is mine because I designed it") is discussed in more detail
in section 2.6 below. The notion of an architect's intellectual "ownership" of his work
is not spelled out by Alberti in so many words, but it is inherent in the notion of
authorship that Alberti borrowed from the humanists' arts of discourse and applied,
for the first time ever, to the art of building.

Thus reformed, architecture ideally acquires a fully authorial, allographic, notational

status. Insofar as a building and its design are considered notationally identical, one
can identify an architectural work either with the design of the building or with the
building itself (a step that Nelson Goodman still hesitated to take in 1968).45 Around
1450 Alberti's claim to architectural authorship (as well as his new way of building
by notation) must have appeared outlandish or worse, culturally as well as
technically. Yet Alberti's authorial ambitions and concerns were common among
writers, rhetoricians, and scholars of his time.
Alberti, the humanist, was painfully aware of the inevitable destiny that awaited all
texts and images when severed from the hands of their authors and caught in the
unpredictable drift of scribal production. Catullus could well "smooth with dry
pumice" his brand-new papyrus roll of poems before presenting it to his first, and
possibly fictional, dedicatee; but that original finishing touch was but a frail seal,
and fifteenth-century humanists knew that most extant classical texts were mosaics
of citations, interpolations, additions, subtractions, and plain copy errors. Modern
philology was developed precisely to reconstruct, as much as possible, the original
text of the author—the one the author would have "smoothed with dry pumice" on
the day he considered his manuscript finished. Contemporary philologists and
linguists have also suggested that in the late Middle Ages the awareness of the
technical variability (mouvance, or drift) of scribal copies prompted new modes of
textual interaction, where variances were not only tolerated, but actually expected,
encouraged, and sometimes exploited.46 The very notion of an "original" would
hardly apply in such a context, as the so-called originals would be too many, and
none more relevant than any other.

There is additional evidence that some of the early human ists (Poggio Bracciolini
in particular) availed themselves of this potentially interactive format to circulate
manuscripts that invited feedback, comments, and additions.47 Alberti himself may
have engaged in this practice—and indeed, it would be fascinating to see Poggio and
Alberti as active wikipedists of the late scribal age. 48 This process of multiple
revisions of, and possibly interactive feedback on, the successive drafts of a literary
text corresponds to the fluid state of architectural design during the "versioning"
phases of its development, on which Alberti insists so emphatically.49 But Alberti
also evidently thought that when revisions stop, they should stop for good—and
forever. Alberti was so anxious about scribal errors that he took the unusual step of
flagging passages where he thought that copyists of his manuscripts might be more
at risk of being led astray, and devised some ploys (as well as some full-fledged and
bizarre devices) to limit that risk, and contain potential damage.

A wealth of evidence proves that, when the final version of a given text was attained,
Alberti aimed at having it copied and reproduced as faithfully as possible, and with
as little external intervention as was conceivable at the time. The best technical
means to this end would have been for Alberti to have his texts and illustrations
printed—which, however, he could not or would not do, mostly for chronological
reasons, although he may have considered the option toward the end of his life.
Indeed, Alberti's pursuit of identical copies is exactly coeval to the development of
print technologies, and this parallel chronology is certainly not a coincidence.
Alberti's insistence on an ideal, but drastic, authorial cutoff—the point at which all
revisions stop and identical replication starts—curiously anticipates a practice that
eventually became common in the printing industry, and survives to this day in the
technical term bon à tirer (good to print).50 Originally, the author's bon à tirer
(normally dated and signed) written on the last proofs validated the final version of
his or her text, and "authorized" its identical replication in print. Thenceforward,
readers could expect exactly the same words in each copy as in the author's original,
even though the author never printed, nor necessarily signed, any individual book.
Thanks to the cultural and technical logic of mechanical replication, authorship was
extended from the author's original to all identical copies of it.

For intellectual and ideological reasons, which should be seen in the context of the
humanists' invention of modern authorship, and perhaps in the larger context of the
humanists' contribution to the shaping of the modern self and of the notion of
individual responsibility, Alberti anticipated this division between an author's work
and its mechanical reproduction. But Alberti tried to impose this authorial paradigm
within the ambit of a manual production chain, where no machine would deliver
identical copies, and scribes could be reasonably expected to produce just the
opposite—randomly changing, individual variations. Also, but crucially for the
history of architecture, Alberti extended his precocious bon à tirer paradigm from
literary to architectural authorship, asserting that the same conditions and the same
consequences should apply. The fact that in most cases the architect's design should
beget only one building (and not a series of copies, as would a printing press, or a
late medieval scriptorium) is irrelevant in this context.51 What matters is the relation
of identicality between the original and its reproduction. Alberti's entire architectural
theory is predicated on the notational sameness between design and building,
implying that drawings can, and must, be identically translated into three-
dimensional objects. In Alberti's theory, the design of a building is the original, and
the building is its copy.