Previous |Contents |Next

Dinsmoor’s Theory

Around the problem of the dimensions of the Parthenon, William Bell Dinsmoor built a general theory about the architecture of Greek temples. This theory consists of two elements, a negation and an affirmation.

To begin with, Dinsmoor denies that the dimensions of the rectangle of the stylobate were of major importance in the planning of Greek temples: the rectangle of the stylobate was “an artificial plane which to the Greeks was purely secondary.” The fact that through the nineteenth and the twentieth centuries scholars of Greek architecture focused their attentions on the dimensions of the stylobate when investigating temples, would have been an unfortunate result of the attention paid to the dimensions of the stylobate by Stuart and Revett in their survey of the Parthenon: “Probably no theory has had more baneful effects on later archaeologists than this.”

Having denied that the dimensions of the stylobate were of major significance, Dinsmoor affirms that the architects of Greek temples began their planning from the diameter of the columns. Then he proceeds to try to prove this theory in relation to the Parthenon with the understanding that thereby the theory would be valid for all Greek temples. He claims that the theory of the importance of the stylobate in Greek temples, originated with Stuart and Revett’s survey of the Parthenon, so that, if this theory is refuted in the case of the Parthenon, it will be refuted for all Greek temples.

In order to prove that the Parthenon was planned beginning with the diameter of columns of the peripteros, Dinsmoor presents the argument that follows.argued that “we have only to remember” that the foot of 12  1/8 English inches (the geographic foot) never existed. He declares that “the mainland Greeks used the Attic foot of 12 7/8 inches [327.0250 mm]; and that the Asiatic Greeks used the Ionic foot of 11 5/8 inches [295.2750 mm], which was by them transmitted to the colonies of Magna Gręcia, and thence to Rome.” On the basis of this fantasy, for which there is not a scrap of evidence, he claimed that the Parthenon was planned by a foot of 327 mm.

The diameter of the columns was 5 5/6 feet of 327 mm., that is, 1907.5 mm. Starting from this basic datum, the architects calculated the clear space between the columns as equal to the diameter of a column plus the width of a man’s shoulders, which, says he, is 1½ feet (a cubit). Hence the clear space between columns was set at 5 5/6 + 1 ½  = 7 1/3 feet; but this figure was changed to 7 7/24 feet, in order to keep a relation 4:5 between the diameter of the columns and the clear space between the columns. Hence, the normal intercolumnium of the Parthenon would be 5 5/6 + 7 7/24 = 13 1/8 feet.

The clear space between columns was reduced to 5 ½ feet in the case of the space between a corner column and the next column. As a result the corner spaces (intervals between the axis of the last column that is not a corner column and the edge of the stylobate) were fixed at 14 ¼; feet, that is (1 ½ x 5 5/6 feet) + 5 ½ feet. Then the diameter of the corner columns would have been increased by 1/40, reducing thereby the clear space between the columns to 5 17/48 feet, but keeping the corner spaces at 14 ½ f; feet. By multiplying these intercolumnia by the number of columns and making these adjustments for the corner columns, one would have arrived at a stylobate of 94 5/12 x 212 13/24 feet. The dimensions of the stylobate would have resulted merely from the additions of intercolumnia and corner spaces so calculated, and, hence, did not need to be expressed in round figures.

One could try to follow the argument of Dinsmoor and point out its arbitrariness and lack of logical sequence. For instance, one could observe that in order to explain why the alleged width of a man’s shoulders was included in the reckoning of the clear space between columns, he declares that this was the “fruit of two centuries of experience.” This would imply that the assumed formula “width of column plus width of a man’s shoulders” was a customary one, whereas in reality if it were true that it applied to the Parthenon, it cannot have applied to any other temple of Athens. The spacing of the columns of the Parthenon was unusually narrow, since the columns occupy half of the width of the temple and very close to half of its length, being closer to each other even more than in the type of temple called pycnostylus, “thick-columned,” by Vitruvius. There are very few temples in the Greek world, including Magna Gręcia and Asia Minor, in which the columns are so near each other as in the Parthenon. In any case, Dinsmoor speaks of “two centuries of experience,” but does not quote a single instance. But there is no need to waste efforts in trying to follow his contorted logic since his theory must be rejected for more essential reasons: it is based on dimensions that are quite different from the actual dimensions of the Parthenon.

To begin with, the normal columns of the Parthenon do not have a diameter of 5 5/6 feet or 1907.5 mm. and the corner columns do not have a diameter that is 1/40 more than this, 1955.2 mm. According to my reckoning, the normal columns have a diameter of 1926.7 mm. and the corner columns, of 1965.2 mm.; according to the survey conducted by Magne, which in Dinsmoor’s judgment is the most reliable one, the diameters are 1928 and 1968 mm. The normal intercolumnium is not 13 1/8 feet or 4,291.9 mm but 4,300.5 mm. the corner spaces are not 14 ¼ feet or 4,659.17, but 4,716.6 mm.

Dinsmoor was not satisfied with taking these liberties with the actual measurements of the Parthenon, but felt free to tamper with his own figures. According to his reckonings the temple should have the following dimensions:


5 normal intercolumnia of 13 1/8 feet = 65 5/8 feet
2 corner spaces of 14 ¼ feet

= 28 ½ feet

Total width

= 94 1/8 feet



14 normal intercolumnia of 13 1/8 feet = 183 ¾ feet
2 corner spaces of 14 ¼ feet

= 28 ½ feet

Total length

= 212 ¼ feet

But, without a word of explanation, Dinsmoor changed the results to 94 5/12 x 212 13/24 feet (30,874.3 x 69,501.1 mm.). In a later publication of his, for equally unaccountable reasons, these figures appear as 94 ½ x 212 5/8 feet (30,901.5 x 69,528.4 mm.). When the dimensions of the stylobate are described according to this second version, Dinsmoor explains that width and length relate “exactly” as 4:9. This statement, contrary to fact as usual, appears to have been added as a barb directed at Stuart and Revett for having said that the length is slightly more than 9/4 of the width.

Naturally, by taking such liberties with the actual dimensions, one can defend any theory. With even less tampering with empirical data, one could prove that the surface of the earth is concave and not convex. Among the dozens of theories about the dimensions of the Parthenon that I have examined in detail, none shows such a disregard for the actual dimensions of the temple. Nevertheless, Dinsmoor’s theory has been widely accepted not only as far as it concerns the Parthenon, but also as having demonstrated that the dimensions of Greek temples were not calculated taking the stylobate as a starting point. For instance, in the general survey of Greek architecture by Helmut Berve and Gottfried Gruben, the theory of Dinsmoor is accepted as valid for temples of the classical period, even though earlier temples would have been planned beginning with the rectangle of the stylobate. These German archaeologists declare (p. 376):

During the sixth century, most of the external dimensions of temples were conceived in round numbers, from which the other measurements were, as far as possible, derived, whereas the early classical architects started out with a definite basic measurement, the interaxial, and developed from it a clear, regular groundplan and elevation, the building’s external measurements now becoming derivative quantities.

They quote as examples not only the Parthenon, but also the Temple of Zeus at Olympia. Elsewhere I show that in the case of the latter temple the dimensions of the stylobate were fundamental and expressed in round numbers.

The reason why specialists of ancient architecture have embraced Dinsmoor’s theory, even though it is obvious that it does not agree with the empirical data, is revealed by another statement of Dinsmoor, which is equally incorrect factually, but puts in clear light his ideological preferences. Dinsmoor prefaces his analysis of the dimensions of the Parthenon by a lengthy profession of principles in which he declares that Greek temples should be studied in the light of the rules formulated by Vitruvius. Then, after this declaration, with which few would have any quarrel, he proceeds to assert that Vitruvius “tells us that columnar buildings were designed to the full lower diameter of the column.” But Dinsmoor, having proclaimed the necessity of following the text of Vitruvius, does not provide any reference for his alleged quotation for the simple reason that Vitruvius does not say anything of the sort. What Dinsmoor has in mind is a rule formulated by Andrea Palladio who claims to have imitated Vitruvius.

Isaac Ware’s translation of Palladio’s Four Books of Architecture (p. 13; Book I, Ch. XIII) reads:

For the better understanding of which, and to avoid my repeating the same things often, it is to be observed, that in dividing and measuring the said orders, I would not make use of any certain and determined meawure peculiar to any city, as a cubit, foot, or palm, knowing that these several measures differ as much as the cities and countries; but imitating Vitruvius who divides the Dorick order with a measure taken from the thickness or diameter of the column, common to all, and by him called a module, I shall therefore make use of the same measure in all the orders.

The passage of Vitruvius to which Palladio refers reads as follows (IV, 3, 3-4):

Let the front of a Doric temple, at the place where the columns are put up, be divided, if it is to be a tetrastyle, into twenty-seven parts; if hexastyle, into forty-two. One of these parts will be the module (in Greek modoulos); and this module once fixed all the parts of the work are adjusted by means of calculations based upon it. The thickness of the columns will be two modules and their height, including the capitals, fourteen.

Vitruvius certainly does not say that the dimensions of the stylobate are not the starting point, as Dinsmoor claims. Vitruvius favors a wide spacing of the columns and condemns as undesirable a close spacing of the columns (III, 3, 1), gives examples of temples in which the columns occupy 3/27 or 12/42 of the width of the stylobate. These examples are far from the case of the Parthenon in which the columns occupy ½ of the width, but unquestionably prove that the diameter of the columns was determined by the dimensions of the stylobate.

It was Palladio who made the column the main element in the architecture of a building. This bias of Palladio was further emphasized by the Palladianists. And ever since the growth of Palladianism, the main notion of academic architecture has been that the way to achieve architectural beauty is to draw some Greek columns and paste them on any building whatsoever.

For the last two hundred years all major reform movements in architecture, beginning with that initiated by Stuart and Revett, have been directed against this degenerate Palladianism. But Palladianism is close to the heart of most professors of architecture and seems to bloom again and again with irresistible vigor, as proved, for instance, in our age by Hitlerite and Stalinist architecture. It is a fact that most cities in the world are adorned, or, one could better say, deturpated, by this kind of architecture. It is the natural choice for those who prefer rhetorical display to logical coherence. Hence, when Dinsmoor interpreted the architecture of the Parthenon and of Greek temples in general, as if it were that of the campus of Columbia University, his theory was greeted with enthusiasm by most of his colleagues. They could easily see that Dinsmoor’s figures do not agree with the actual dimensions of the Parthenon. They could also see that he misquotes Vitruvius, since any scholar of history of architecture is familiar with this basic author; but any scholar of history of architecture is also familiar with Palladio and with Palladian writers and, hence, can realize that Dinsmoor had his heart in the right place. His theory of how Greek temples would have been planned contains the quintessence of academic taste in architecture, and if the remains of Greek temples do not agree with it, so much the worse for them.

Previous |Contents |Next