The CSA Propylaea Project

Harrison Eiteljorg, II, 2013

Data copyright © Harrison Eiteljorg, II unless otherwise stated

This work is licensed under the ADS Terms of Use and Access.
Creative Commons License

Samuel H. Kress Foundation logo

Primary contact

Harrison Eiteljorg, II
Center for the Study of Architecture
Box 60
Bryn Mawr
PA 19010

Send e-mail enquiry

Resource identifiers

Digital Object Identifiers

Digital Object Identifiers (DOIs) are persistent identifiers which can be used to consistently and accurately reference digital objects and/or content. The DOIs provide a way for the ADS resources to be cited in a similar fashion to traditional scholarly materials. More information on DOIs at the ADS can be found on our help page.

Citing this DOI

The updated Crossref DOI Display guidelines recommend that DOIs should be displayed in the following format:
Sample Citation for this DOI

Harrison Eiteljorg, II (2013) The CSA Propylaea Project [data-set]. York: Archaeology Data Service [distributor]

Final Report, CSA Propylaea Project: General Introduction and Project Rationale



The Center for the Study of Architecture (CSA) operated the CSA Propylaea Project beginning in 2000; the project was concluded in 2008. The project arose from a combination of events, interests, backgrounds, and historical facts that cannot be explained in a few words. This somewhat lengthy and discursive document describes the path that lead Harrison Eiteljorg, II, to the CSA Propylaea Project. (Mr. Eiteljorg is the Director of CSA, the Center for the Study of Architecture.) This document will also set forth the reasons for the project and briefly describe the nature of the project.


CSA Director Harrison Eiteljorg, II, embarked on the path that led to the CSA Propylaea Project when he was a graduate student at the University of Pennsylvania in the 1970s. At that time he became interested in the work of the architect in the classical period; this may have happened because Mr. Eiteljorg had begun his college career intending to become an architect. Eventually Mr. Eiteljorg wrote his Ph.D. dissertation on the subject: The Greek Architect of the Fourth Century B.C.: Master Craftsman or Master Planner.

The path from Mr. Eiteljorg's interest in the planning done by ancient architects to the CSA Propylaea Project was not a direct one. Among the detours was Mr. Eiteljorg's excavation of the old entrance to the Athenian Acropolis, called the older propylon. Another detour was a deep immersion in the use of computers in archaeology. What follows is therefore a somewhat lengthy description of the journey through the various activities that led from the 1973 dissertation to the twenty-first century CSA Propylaea project.

The Basic Problem: The Role of the Architect in Antiquity

We know very little about the actual work of the architect in classical Greece. (The oldest surviving book dealing with Greek architecture is that of the Roman architect Vitruvius, but he wrote about 400 years after the great buildings on the Athenian Acropolis had been completed.) We are reasonably sure that scaled drawings such as those we take for granted today were not made. In fact, our only building plan from the classical period (from the fourth century B.C.E.) is a description, inscribed on stone and entirely in words. It describes a naval arsenal to be built near one of the harbors of Athens. The architect is named, and most of the features and dimensions of the building are specified. The inscription also indicates that some matters were yet to be decided and would be specified on site; those new specifications might be in the form of models of particular details or in the form of directions from the architect. There is other evidence that full-sized models of some parts of buildings (column capitals, for instance) were provided to guide the ancient masons as they worked.

There are indications in the arsenal inscription that the architect had determined the dimensions of small parts of the structure implicitly, because those dimensions were required by the canons of the day or were implicit in other construction specifications.

We have some inscribed contracts for construction work as well, and some of them also imply considerable specificity in advance of construction. The vast majority of those contracts date from the fourth century B.C.E. or later. Contracts for quarrying and transporting blocks also survive, and they are very specific as to sizes and numbers of blocks. Therefore, contracts and building needs indicate to most observers that careful planning by ancient architects was common.

Despite the evidence from the fourth century and later, we have little real evidence about the specifics of the planning process, especially in earlier periods. The etymology of the term architect, meaning master craftsman originally, seems to suggest that the architect was more a foreman than a planner when the term arose. Indeed, some have argued that building a Greek temple was similar to a barn-raising on the American frontier. Those American barns were not planned in any real sense; the overall dimensions were determined in advance, but the building process involved carrying out work according to tradition, not according to plan. As a result, the question that intrigued Mr. Eiteljorg was this: "When did the architect begin to plan more fully?"

The First Detour: The Older Propylon

During the Late Bronze Age, ending about 1200 B.C.E., the Athenian Acropolis served as a citadel, and a massive fortification wall was built around it near the end of that period. The principal gate of the fortification wall was at the western end of the Acropolis, where the approach was less steep, and the gate there remained in use long after the end of the Bronze Age. In fact, the fortification wall survived into the fifth century B.C.E., and parts of it are still standing today. A long, sloping approach path seems to have been added in the sixth century B.C.E., but the gate itself was not altered at that time. The standard view (in 1975) was that after the Battle of Marathon (490 B.C.E.) the simple gate was replaced by a new gate building. The new building was assumed to have been a colonnaded entrance building, and that new entrance structure has been called the older propylon. It was thought to have been started after the Battle of Marathon in 490 B.C.E. and completed before the Persian invasion of 480 B.C.E. It must have been partly destroyed by the Persians.

The older propylon remained the entrance building for the Acropolis until it was demolished to make way for the Propylaea (the grand Periclean entrance building to the Athenian Acropolis) beginning in 437 B.C.E. Of course, there must have been some repairs after the Persian invasion of 480 B.C.E. Therefore, scholars believed that the older propylon was the entrance building for the Acropolis from some time between 490 and 480 B.C.E. until 437 B.C.E.

Mr. Eiteljorg had included some information about the older propylon in his dissertation and had studied the building. His study indicated to him that the prevailing views of its size did not seem to fit with the minimal known remains. Therefore, Mr. Eiteljorg undertook a short project to excavate the remains of the older propylon in 1975. (The older propylon had been excavated many times before; so this project was to re-excavate the older propylon.) He received permission for the work with the kind help of the American School of Classical Studies at Athens.

What Mr. Eiteljorg found was quite unexpected. A previously undiscovered cutting in the bedrock of the Acropolis was uncovered, and it was an enormous surprise. The cutting showed that the interior plan of the structure had to be reconsidered. Therefore, the overall plan also had to be reconsidered. The older propylon was a far smaller and simpler building than had been thought. However, it also must have gone through more changes than had been thought. There were three separate phases of the building that could not be precisely dated; the first phase must have fallen in the ten-year period after the Battle of Marathon (490 B.C.E.). The second phase must have been the result of repairs after the Persian sack in 480. The entrance structure was re-worked again before its final demolition, apparently because of instability in the repaired version of the structure. The date for that re-working cannot be determined. The building was finally removed to make way for the grand entrance building of the Periclean Age, the Propylaea, in 437 B.C.E.

None of the three phases of the older propylon involved a colonnaded building. Instead, the successive entrances amounted only to rather decorative courtyards; none would have required the work of an architect in the modern sense. There was nothing more complex than the erection of decorative blocks of veneer for the old fortification wall, the installation of simple steps, the repair of various walls after the Persian attack, and the erecting of a short wall atop the steps in the last phase.


Eiteljorg reconstruction of older propylon

The Entrance to the Acropolis — the older propylon — of 480 B.C.E. as
reconstructed by Mr. Eiteljorg. The entrance consists only of two courtyards. The upper
courtyard is on the east side of the steps and ramp; the lower one is to the west. The area was
unroofed. The gate itself was probably the old gate in the fortification wall at the east
side of the upper courtyard. (Actual remains in black; restorations in red,
and Propylaea outline in light gray. Drawing from the CAD model.)


Dinsmoor reconstruction of older propylon

The Entrance to the Acropolis — the older propylon — of 480 B.C.E.
according to another scholar, William B. Dinsmoor, Jr. This drawing was
scanned from Plate 16 of Mr. Dinsmoor's book, The Propylaia to the Athenian Akropolis:
The Predecessors
(Princeton: 1980). The original drawing was simplified for presentation
here. This older propylon is a colonnaded and roofed structure with a gate wall in the
middle. (No visual distinction between actual remains and restored materials; a
portion of the Propylaea shown with broken-line outline.) Other scholars have
suggested wider or narrower entrance buildings, but the general idea shown here
was the accepted view before 1975.

Note that both of these reconstruction drawings show part of the old fortification
wall in the area north of the central portion of the Propylaea. That
area has been thoroughly studied and shows no remains of the
fortification wall. Both are therefore likely to be incorrect
as to the northern extension of that wall.


The Second Detour: Computing in Archaeology

Mr. Eiteljorg published his finds concerning the older propylon immediately (1976). However, he was unable to create a full and complete analysis of their meaning at that time. He returned to Athens twice in the 1980s to study the remains further, since the analysis required additional data. The final publication appeared nearly 20 years after the initial excavation, in the 1993 book, The Entrance to the Athenian Acropolis Before Mnesicles (Boston: 1993). When he worked on the analysis, Mr. Eiteljorg needed to make drawings that would permit him to deal with all the remains that had been found, but standard paper drawings were not adequate. Fortunately, computer-aided drafting programs — sometimes called computer-assisted drawing and generally referred to as CAD — were coming into use as the time. Mr. Eiteljorg was able to begin using a CAD program. It permitted him to work with all the material he had found and provided important advantages over hand-drawn plans. Over time, Mr. Eiteljorg became more and more expert in the use of CAD. He developed very sophisticated ways to use CAD programs to assist with the analysis; for instance, he was able to examine various scenarios for the different phases of the older propylon without drawing any parts of the building a second time.


Eiteljorg plans of older propylon remains   Eiteljorg plans of older propylon remains

Two views of the plan data for the older propylon, one (left)
showing all plan information and the other (right) showing only the
material belonging to the earliest phase.


As Mr. Eiteljorg worked with the CAD programs, they became more and more complex. The programs became more and more capable as well, and they even permitted him to create fully three-dimensional "models" of architectural material, not just plans. The 3D CAD programs offered two significant advantages. The first was the possibility of visualizing ancient structures in 3D. The older propylon could be "seen" from virtually any vantage point, making it far easier to visualize the actual remains. Of course, it was also much easier to visualize a suggested reconstruction. Mr. Eiteljorg was especially pleased that he could make good, understandable 3D views for those who were not familiar with the building.

The second advantage was critical to scholars but not necessarily to others. The data in a CAD model was far more precise than data in a drawing. The CAD model could retain the full precision from survey work. That is, a point identified as having an x-value of 1.234 m., a y-value of 2.345 m., and a z-value of 3.456 m. would always retain those values in the digital file containing the model. (Note that the example presumes measurement precision to the nearest millimeter. Modern survey instruments often provide far more precise measurements, but Mr. Eiteljorg routinely rounds all such measurements to the nearest millimeter. If more precise measurements were used, the CAD program would retain them.) By comparison, the scale of any paper drawing places definite physical limits on precision. The smallest measurable distance is determined by the scale chosen; for instance, at the unusually large drawing scale of 1 unit in the drawing to 10 units in the real world, one millimeter in the real world would be only one-tenth of a millimeter (about the thickness of an average human hair) on paper. At a more usual scale of 1:50, a mm. in the real world would be only one-fiftieth of a mm. on paper, too small to be seen, much less measured.

The limits on precision apply both to the draftsman and to the user of a paper drawing. The draftsman may only draw as precisely as using pen or pencil on mylar or paper will permit. Then, regardless of the draftsman's skill, the user can only interpret the drawing to a certain level of precision. For instance, if a line has thickness (as it must to be seen) does one measure to one side or the other or the middle of the line (assuming the middle can be determined)?

A CAD model uses — and retains — the coordinates provided by survey, regardless of precision. (The program will treat 1 as equivalent to 1.000.) All drawings are created from the numbers. CAD drawings may be marginally more precise than hand-made drawings, but any user is still limited by the problem of interpreting the drawings. The great advantage of the computer model is that the coordinates associated with a point are retained in the CAD model; they can be retrieved (with full precision) at any time via a simple command. Thus, the precision with which a building has been surveyed is retained by the CAD model while it cannot be retained in a paper drawing. (For more information about CAD software and its application to archaeology, this old and somewhat outdated booklet, Computer-Assisted Drafting and Design: New Techniques for Old Problems, remains a good introduction. Other resources available from this page may also be of interest; they are more technical but also more current.)

When Mr. Eiteljorg worked with CAD, he also worked with other computer technologies and became convinced that computer data — data in digital form — had significant values for scholarship. Data in digital form could be far more easily and cheaply shared with colleagues. Data could be sent on a floppy disk in the early years of this work on the older propylon, then on CDs, and now shared on the Internet. Digital data also made it possible to share the information in forms that were more useful. A 3D CAD model is more complex, precise, and useful than any drawing; a database is far more useful than tabular information on paper. Finally, color photographs are too expensive to publish on paper but are valuable for the extra information they convey; they can easily be shared in digital form. In short, the use of digital data permitted scholars to work more collaboratively than ever before.

The Propylaea

The Propylaea replaced the older propylon as the building guarding the entrance to the Acropolis starting in 437 B.C.E., and it seems to hold some answers to the questions about the work of the architect in antiquity.

The Propylaea was a grand entrance building for the Acropolis; it was the second of the buildings erected while Pericles lead Athens through its most glorious years. The Propylaea was built between 437 and 432 B.C.E. (but left unfinished), directly after the construction of the Parthenon. The Propylaea led people toward the heart of the Acropolis where the Parthenon stood. (For more information about the Propylaea, see this page for the Wikipedia article about it and the CSA Propylaea Project page with general information about the building.) Construction of the Propylaea commenced before all the sculptural portions of the Parthenon had been completed; however, the structural portions of the Parthenon had been finished and the temple had been dedicated. The Propylaea was a grand building, as it had to be for its use as the gateway to the Acropolis. It has been used ever since as a model for grand entrance structures. Its central core has been used as a model not only in antiquity but in more recent structures such as the Brandenburg Gate in Berlin.


Fish-eye photo of the Propylaea

The Propylaea from the West. This fish-eye photo
distorts the building, but such a wide-angle view is the only way
to see the entire structure from the west without intervening structures.
(Original Kodachrome (25) photograph; August, 1994)

Propylaea from West

The Propylaea from the West.
(Original Kodachrome (25) photograph; March, 1975)

Propylaea from the East

The Propylaea from the East. This is a relatively
old photograph, used because more recent ones all include
scaffolding used when the building was being restored.
(Original Kodachrome (25) photograph; March, 1975)


The Propylaea is not only large and grand; it is also complex. It is apparently the first monumental building of the classical period that was designed to be more complex in plan than a simple rectangle or a circle. It was, from the very beginning, planned that there would be a central core plus four wings. Two of the wings were to be on the west side (facing toward one another); the other two wings were to be on the east side (not facing one another but facing east, like the central core, toward the heart of the Acropolis).


Propylaea plan

The Propylaea plan. Circles enclose wall stubs indicating the
intention to build eastern wings, which were never
constructed. The wall stub in the SW wing seems to exist
only to provide a symmetrical façade for visitors.

The complexity of the Propylaea required a depth of planning not needed for buildings designed in simpler shapes. For instance, the smaller-scaled wings on the west had to be joined to the central core at the level of the roof to cover the floor of the building. But that made the roofing design very difficult. In addition, the typical grand structures of the day — temples and colonnaded stoas used for commercial, governmental, or religious matters — had many ancestors and fixed notions of their final appearances. Therefore, the advance planning required for such buildings was minimal; relatively gross dimensions limited everything else and permitted experienced masons to work with few specifications. The architect could rely upon the experience of all involved to take care of most problems. In fact, the most difficult aspects of construction were so subtle that a modern architect — even one using the latest equipment — would find it very difficult to design and document those construction matters. (For example, temple floors were intentionally designed with subtle curves. The edges of the floors rose from the corners to the center of each side. The shape of a temple floor has been likened to that of a handkerchief held down at its four corners while a gust of air is blown underneath, causing some lifting of the handkerchief throughout, with the highest point being the center. The change in the height of the floor, even in the Parthenon, is so slight that a modern architectural drawing would not effectively show that rise at a scale needed to draw the side of the Parthenon on a 6-feet-long piece of drafting paper. A scale of 1:40 would permit the north or south elevation of the Parthenon to be drawn on a single sheet of drafting paper 72 inches long. At that scale, the rise in the long sides of the building from corner to center would be 0.108 inches - less than 1/8 of an inch and less than 3 mm. Thus, the change in height from one side of a stone along the floor to the other side of the same block would amount to less than 5 thousandths of an inch or about .25 mm., much less than the thickness of the lead in a good drafting pencil.)

Because it was so complex, the Propylaea had no real predecessors, no models on which the problems of of its design had already been worked out. (Even those scholars who believe that the predecessor of the Propylaea, the older propylon, was as large as the central core of the Propylaea itself do not suggest that there were wings on the earlier structure.) As a result, the Propylaea seems to be a key structure in the history of Greek architecture. It requires planning both because it is unique and because it is complex. It seems to mark a new kind of structure — one without predecessors to act as guides to its design and construction. (We know the name of the architect who designed it: Mnesicles. His name comes to us from Pausanias, the Greek traveler who visited Athens and wrote about its monuments in the second century C.E., but we know nothing else about Mnesicles.)

During the 1980s, as Mr. Eiteljorg was still working on the analysis of the older propylon, he met the man in charge of modern work to restore parts of the Propylaea, the Greek architectural historian, Dr. Tasos Tanoulas. They became friends, no doubt in part because of their mutual interest in the entrances to the Acropolis.

Mr. Tanoulas began work on the Propylaea restoration project in 1984. In 1994 he published an important work about the project, and that work included carefully drawn and dimensioned elevation views of two of the walls of the NW wing of the building. Mr. Eiteljorg saw those drawings and the three-dimensional survey information noted on them. As a result, Mr. Eiteljorg thought he could create a full, three-dimensional, block-by-block computer model of the Propylaea with a CAD program. With that much information (three coordinates for every corner of every block), he envisioned the real possibility of understanding the design of the Propylaea — what the architect had intended, what he had foreseen, and what he had not expected because the building was new, different, and complex.


Computer model of SE corner of Propylaea

A computer model of the SE corner of the Propylaea. Note that
cuttings can be clearly shown. Broken blocks, final surfaces, and
the temporary surfaces that had not been trimmed away when
construction stopped are also shown. Each block is complete
and could be illustrated in isolation from the others.

Model of one block of Propylaea

A computer model of a single block from the Propylaea. Note that
portions of the block that were roughly trimmed are shown (in blue).
Portions that were broken away are also indicated (in red).

The CSA Propylaea Project

Mr. Eiteljorg, Mr. Tanoulas, and Maria Ioannidou defined the CSA Propylaea Project. (Ms. Ioannidou was the engineer for the Propylaea restoration work. She later moved to another position and had too little time to participate fully.) The aim of the project was to create a block-by-block CAD model of the Propylaea. Mr. Eiteljorg, Mr. Tanoulas, and Ms. Ioannidou created a grant proposal for the CSA Propylaea Project, and funding was gratefully received from the Samuel H. Kress Foundation beginning in June of 2000. The computer model was the ultimate aim, and included in the grant proposal was the critical idea that all information about the building — the CAD model, the data files from survey work, and all photographs, in color — should and would be freely available via the web. The intention to make all data readily available via a website was an integral part of the plan from the beginning.

The team started to work immediately, but there was a problem that appeared very quickly. The three-dimensional survey work shown in the drawings of two walls of the NW wing was abnormal for the Propylaea. Most walls of the Propylaea had not been surveyed in all three dimensions. Instead, the height and length of each block had been measured, not the complete position of each corner. As a result, the project became, to a significant extent, a technical one. It was necessary first to survey the building to obtain the survey information for the blocks in the building. (There were some other fully-surveyed portions of the building, and computer models could be created of those portions, for example, the wall shown above. However, the building as a whole could not be modeled with the data available.)

Surveying the building sounds simple, but it is not, especially with thousands of tourists passing through the building every day. Working on the best ways to accomplish the surveying engaged the team from 2001 until 2007; meanwhile, the CAD work that could be done without additional survey data moved forward. (See this link for articles about the project, including many concerning the work to solve the survey problems.) Various modern survey techniques were tried. Finally, in December of 2007, a system that offered the best possible survey results was successfully tested in the Propylaea. It required a portable scaffolding structure (sent from the US to Athens); the scaffolding could be erected where needed in the building and moved from place to place. Using the scaffolding system, a team member could reach any point in the structure and hold a survey target at any relevant point. (More permanent scaffolding could not be used for this project; it would have obscured parts of the building while in position, not only for the tourists passing through the building but also for the surveyors who needed clear sight-lines to any survey point. Thus, the portable scaffolding was critical. The small size of the survey team also required that the portable scaffolding be small and light. It had to be put into position by two or three people.) A member of the team could stand on the scaffolding to hold a survey target on any point of interest, and another member of the team could survey the designated point from the ground with a modern surveying instrument, called a total station. The system developed was not fully capable in 2007; no method for raising and lowering the scaffolding with only two or three people — and without ropes going over fragile wall blocks — had been created. Additional pieces were created during the following spring and summer. The scaffolding could then be raised and lowered by one or two people, and the system could function as required. The full system was ready to go by the fall of 2008.


Scaffolding test in the Propylaea in 2007

The scaffolding system as tested in the Propylaea in December of 2007.
(Digital photograph; December, 2007)

By the time the new system was ready to be put to work, however, Mr. Tanoulas had changed his position about the website (the CSA Propylaea Project website and any successors). He had decided that there could not be open access to the project's data. Instead of having all information available via the web, any information gathered in the future — at the project's expense — would belong to the Greek Ministry of Culture. Such data would be made available to the team (and/or others) under whatever terms and conditions the Ministry of Culture might impose from time to time. As a natural consequence, Mr. Eiteljorg and the Board of Directors of CSA recommended that the project be terminated. Officers from the Samuel H. Kress Foundation agreed. One of the most basic aims of the project could no longer be met. Remaining funds were returned to the foundation; work on the building was terminated; the digital materials were put into final form to the extent possible; the website was completed to make those materials available; and the work of preparing the digital materials for archival preservation was begun, leading to this material now residing on the ADS website.

The CSA Propylaea Project has stopped. Nevertheless, Mr. Eiteljorg remains committed to learning more about the work of the architect in ancient Greece. He hopes that the knowledge gained, though it is incomplete, will permit him to continue the search for the real work of the architect that underlay the Propylaea. At the moment, for instance, Mr. Eiteljorg believes that the length of the blocks of the structure are the most likely building module. In addition, Mr. Eiteljorg believes that the Propylaea could not have been completed without several modifications to the design. For example, the floor level in the northeastern wing would have been far too low for its position on the Acropolis. There are other, equally serious difficulties with the intended design; these problems should not surprise anyone, given the novelty and complexity of the Propylaea.

This ADS website provides access to the data from the work of the CSA Propylaea Project, including information about the survey and computing processes used by the team. Since the project was not completed, however, interpretive results are not presented here. (There are a blog and related web pages concerning architectural planning in antiquity. They are available via the CSA Propylaea Project website but may or may not remain there over time.)

The Wikipedia page about the Propylaea has more information about the building; so does the General Information page here. The History (of the project) should be read before moving on to more detailed materials.



About this document:

  • Title: "General Introduction and Project Rationale
  • Author: Harrison Eiteljorg, II and the staff of CSA, Box 60, Bryn Mawr, PA 19010, (e-mail: user nicke at (@) the domain; tel.: 484-612-5862)
  • Original file name: indexns.cfm
  • Revision history: Since this document is part of the CSA Propylaea Project Final Report and has been archived with the Archaeology Data Service, changes should not occur. Serious mistakes may be corrected; if so, clear indications of corrections will be included. While preliminary versions of this document were on the project website, the version represented here was not previously posted. It was prepared in 2011. Prepared for ADS archival storage June, 2013.
  • Internet access: This document was first prepared for, operated by the Center for the Study of Architecture and Harrison Eiteljorg, II. It has been turned over to the Archaeology Data Service for archival preservation.
  • Long-term availability: This document or its successors will be maintained for electronic access indefinitely.