1

On the Origin of the Doric Order

Jacob Reidel

“But he must acquaint himself with many narratives from history; for architects often incorporate many ornamental features in the designs of their works, of which they must be able to give a reasoned account, when asked why they added them.” (Vitruvius, I.i.5)

Of the three primary classical architectural orders—Doric, Ionic, and Corinthian—which gradually developed in Greece beginning (with the exception of the later Corinthian) sometime around the eighth century BCE, the Doric is perhaps the most quintessentially Greek. Reaching its most refined state in the fifth century BCE with the construction of the Parthenon, the Doric order soon thereafter fell out of fashion (Chitham: 1985: 21). Whether this occurred due to the order’s famously difficult and exacting proportional requirements (in particular the resolution of the corner triglyphs) or simply a change in taste is open to conjecture, but the use of the Greek Doric declined after the fifth century, and changed drastically once the Romans assumed control of architectural production. However, a resurgence of the classical vocabulary in architecture brought about by the Renaissance, combined with a rising popular taste in the early-to-mid nineteenth century for Greek (as opposed to Roman) architecture, especially in the United States, has left us thus surrounded by these products of classical Greece. In light of this, the above quote from Vitruvius, perhaps the single most important ancient writer on architecture, brings to light an important question. If one is to use these forms, or by extension, live among them, would it not be best to have a notion of their origins? As the following will demonstrate, there are numerous causes posited for the forms of the Doric order, and while there may be no definitive answer to where, for example, the Doric capital came from, the forms found in classical architecture were by no means derived via sheer whimsy.

Vitruvius’s work, written under Augustus hundreds of years after the fact yet nevertheless the oldest and best source to have survived on architectural matters, claims that the Doric originated when Dorus, ruling the Peloponnese, built “a temple in this style at the old city of Argos, in the sanctuary of Juno, and, afterwards, in the other cities of Achaea after the same style, when as yet the determination of the exact proportions of the order had not begun” (Vitruvius IV.i.3). In time, the Athenians began to found colonies in Ionia, and supposedly, when the time came to build a temple in this new land,

they called it Doric because they had first seen it built in that style. When they wished to place columns in that temple, not having their proportions, and seeking by what method they could make them fit to bear weight, and in their appearance to have an approved grace, they measured a man’s footstep and applied it to his height. Finding that the foot was the sixth part of the height in a man, they applied this proportion to the column. Of whatever thickness they made the base of the shaft they raised it along with the capital to six times as much in height. So the Doric column began to furnish the proportion of a man’s body, its strength and grace.(Vitruvius IV.i.5-6)

This story is interesting and has had a lasting impact, whether for better or for worse, on later western theories of classical architecture, but Vitruvius makes fairly clear errors in his tale. First of all, the column proportion of six diameters to the height was actually not reached by the Doric until much later in its development (closer, in fact, to the time in which Vitruvius was writing) (Warren, 1919: 188). In addition, his chronology is at times faulty. Nevertheless, there was supposedly a very ancient wooden Doric temple located at Argos, which burned in 423 BCE (Rykwert, 1996: 249). It would appear that while Vitruvius’s account of the origins of the Doric should at time be viewed with suspicion, it should not be discounted entirely.

However, we must look beyond Vitruvius. One theory that became quite popular in the early to mid-nineteenth century was that the Doric order originated in Egypt. This view was initially put forth by Champollion in 1833 after examining the site at Beni-Hasan on the east bank of the Nile. Here he had found a number of tombs featuring one or two polygonal stone pillars supporting an abacus capital. Since Campollion dated the tombs to the ninth century BCE, and moreover thought he could identify wall paintings inside the tombs as being depictions of captured Greeks, tracing the Doric back to Egypt suddenly became a very promising possibility (Rykwert, 1996: 158). Sites farther south at Deir-el-Bahari outside Thebes and Karnak also appeared to have examples of the stone column with abacus, and this Egyptian type would soon come to be called the proto-Doric. Significantly, this type was believed to have been derived as a pure-stone design, meaning that its forms were not thought to reflect older wooden architecture. This directly contradicts Vitruvius’s theory on the origins of the Doric, which state that the forms of the orders as realized in stone took their cues from the earlier features of traditional wooden construction. As he wrote in De Arch IV. ii. 3. “just as triglyphs came by the treatment of the beams, so from the projections of the rafters the detail of the mutules under the cornices was invented…[thus] in the Doric style the detail both of the triglyphs and of the mutules arose from this imitation of timber work.”

There are a number of difficulties with locating the source of the Greek Doric in Egypt however. First of all, Champollion had misdated the tombs at Beni-Hasan, which actually appear to have been built in the Twelfth Dynasty, c. 2500 BCE (Rykwert, 1996: 456, n. 58). In fact, archaeological evidence now suggests that these so-called proto-Doric columns were only in use between the Eleventh and Eighteenth Dynasties, thus placing these Egyptian models at least six hundred years before the earliest stone columns appear in Greece (Dinsmoor, 1950: 56). Moreover, formally the Greek and Egyptian types bear important differences. For example, the earliest Greek stone columns are far squatter (four diameters to the height of a column) as compared to the examples found at Beni-Hasan (five to six diameters), there is no echinus in the Egyptian version (which had been a primary component in even the earliest Greek examples), and the Egyptian abacus is barely wider than the diameter of the column shaft (in stark contrast to the wide Greek abacus). In fact, the Egyptian versions are quite different from the columns used by contemporary Minoan and Mycenaean builders (Rykwert, 1996: 160). Furthermore, it has been noted that there were a number of other column types at Beni-Hasan and Thebes (Dinsmoor, 1950: 56 n.). For example, lotus bud capitals and bell-shaped capitals abound, yet never are to be found in Greece. If the Greeks derived their Doric forms from these Egyptian examples, would it not be expected to find instances of other Egyptian types in Greece as well? In any event, the significant chronological gap, combined with important stylistic differences, weaken the case for an all-stone, Egyptian origin of the Greek Doric.

Thus rejecting the Egyptian model as the originator of the Greek Doric, let our attention turn back to Greece itself. While there was a significant disjunction between Minoan/Mycenaean architecture and later Greek developments, the Doric nevertheless bears greater resemblance to the columns of these earlier civilizations than to those found in Egypt. Perhaps the possibility should be considered that the Doric column was developed within Greece itself. In fact, this may be supported by Vitruvius’s aforementioned notes on the timber origins of Doric stone architecture, which are perhaps worthy of consideration. First of all, it is almost assured that Greek construction began working in wood and only later moved on to stone. Some of the oldest examples of Doric columns are known from depictions in vase-paintings, and appear to be made out of wood (Robertson, 1945: 65). It seems reasonable to consider that perhaps the simplest form of a ‘column’—indeed the natural form—is a tree, and hence the first columns were probably little more than tree trunks sunken into the ground. This idea is amusingly reflected in Greek epic when in the Odyssey, Book XXIII, Odysseus speaks of

a young olive growing within the precincts of the house, in full vigour, and about as thick as a bearing-post…I cut off the top boughs of the olive tree and left the stump standing. This I dressed roughly from the root upwards and then worked with carpenter's tools well and skillfully, straightening my work by drawing a line on the wood, and making it into a bed-prop. (Homer, Odyssey XXIII).

Unlike earlier Minoan columns, Greek Doric columns taper towards the top. It is worth noting that the forests of Crete were predominately composed of cypress, while mainland Greece was populated with oak (Allsopp, 1965: 45). Since oak trunks taper upwards, it would appear that the form of the native Greek stone column may thus have derived virtually from the land itself. With the rough form of the Doric column hence explained, what about the capital and base? Once again, early timber construction can reveal the source. The exposed top end grain of a cut log placed vertically will be susceptible to water penetration and rotting. Moreover, as the trunk dies, the end grain has a tendency to open up, thus producing star shakes which weaken the structural integrity of the trunk (Allsopp, 1965: 46). It is therefore necessary to close off the top end of the trunk. A stone or timber slab would accomplish this task neatly, and thus the invention of the abacus of the capital. Many more elements typical of the Doric order can be explained in similar fashion—for example, the mutules, which in timber construction correspond to the ends of wooden roof rafters; the triglyphs, which would have been the protective facing placed on the ends of the roof beams; and the metopes, which correspond to the terracotta wall facing between the ends of the beams. When considering the timber origins of the Doric forms, it becomes much more likely that this order is simply an extension—albeit, a highly refined one—of traditional Greek wooden building practices.

Hence, similarities in form observed between Egyptian and Doric columns such as the presence of the abacus can be viewed with this knowledge not as an indication of some imported Egyptian influence, but rather as a natural symptom of building styles with roots in traditional wooden construction techniques, common across many cultural and geographical divides. As Joseph Rykwert pointed out in his monumental work The Dancing Column, at Beni-Hasan “small ribs project along the top edge of the main beam, as if they imitated the rafters of a wooden construction, while the ceilings of the tombs of Khnum-Hotep and Amenemhat are painted with a checkered pattern, clearly intended as representations of rush matting; some of the stone beams are even painted to simulate wood” (Rykwert, 1996: 158). Thus Beni-Hasan, the model so often cited as a pure-stone example adopted by the Greeks, appears to have in fact derived its forms from wooden construction techniques, just as the Doric order was later to do on its own.

But much of this discussion regarding the roots of Doric architecture has been grounded in theory since there are no surviving examples of traditional wooden Greek construction. When examining the few surviving physical examples of pre-Doric architecture, the progression from wood to masonry Doric building becomes slightly more muddled. Examples of this are the two early Archaic temples at Corinth and Isthmia, both having been constructed sometime in the seventh century BCE. What remains does not contain any specifically Doric elements. However, following the argument which posits timber construction as the source of the Doric, the temple at Isthmia was initially reconstructed by completing the building with wooden Doric elements, the assumption being that in the seventh century these elements had not yet made the transition to stone and consequently did not survive. A later reconstruction, though, was made without the use of conjectural Doric wooden elements, for it was found that such a reconstruction could rely solely on the stone and tiles which remained. According to this reconstruction, the temple was not Doric, but rather consisted of “a simple, rectangular, solid-walled building with a hipped roof, probably at both ends” (Rhodes, 1987: 477). The temple at Corinth was found to be almost identical, and a sixth century building at Nemea proved to be very similar. It has long been believed that the Corinthia was a pivotal region with regards to the creation of monumental architecture, and hence a possible birthplace of Doric architecture (Rhodes, 1987: 479). But then, how could the seventh century temples at Corinth and Isthmia and the sixth century building at Nemea not display any characteristic features of the Doric?

The lack of typical Doric features in these buildings such as triglyphs and other elements has been seen by some scholars as evidence that the Doric order did not develop in the gradual fashion that has been suggested by a Vitruvian timber-stone progression. As Robin Rhodes noted

[i]n the Corinthia, a kind of petrification of half-timbered rubble and mudbrick buildings occurs in the eighth and seventh centuries B.C., but not primarily for the purposes of decoration or monumentality. Rather, it is for structural purposes that cut stone begins to appear with frequency in retaining walls, as thresholds, as beds for wall timbers or socles for mudbrick, as sturdy corner foundation blocks or quoins, and, finally, as a sound cornice for the heavy new tile roofs. (1996: 479-80)

Thus, it would appear that at Corinth and Isthmia, the architectural use of stone is dictated by function alone and hence lacks the Doric décorative elements which hearken back to the days of all-wood construction. Does this then mean that these examples from the Corinthia, located in a period and region pivotal to the development of Greek architecture, cannot be viewed as part of the tradition which eventually led to the Doric order? Or conversely, do these buildings devalue the concept of there being a connection between the Doric forms and precedents in wood? I would be led to conclude that the answer to both questions is no. Given the location and date of these buildings, it would be inappropriate to suggest that they simply exist outside the progression of Greek architecture towards the Doric. On the other hand, there is simply too much evidence to deny a long-standing connection over time between the Doric and older wooden forms. Instead, the evidence at Corinth and Isthmia suggests revising the Vitruvian version, but not scrapping it entirely.

It has been long noted that in Doric peripteral temples, triglyphs and mutules are found on all four sides. Of course, since these elements represent wooden rafters, it is a strange fact to find them on four sides of a pedimental structure (of which the Doric peripteral temple is an example). This is not the case, however, with the double-hipped buildings such as those found at Corinth, Isthmia, and Nemea. Such structures would indeed feature sloping rafters on all four sides (Rhodes, 19987: 480). Thus, what can be seen is an example of a functional element of this transitional, pre-Doric architecture finding its way into the finished Doric form as decoration. Indeed, the buildings in the Corinthia do have a place on the evolutionary ladder of the Doric order, the lack of other characteristic Doric details in these buildings only meaning that those missing characteristics must have developed somewhere else. This would in turn suggest a somewhat looser interpretation of the Vitruvian / wood precedent scheme which allows for variation and innovation throughout the development of the Doric order— a variation, one should note, which is wholly in keeping with the geographical, political, and social divisions that have at times throughout history kept pockets of Greeks in relative isolation from one another.

In conclusion, when combining Vitruvius’s account of the Doric order’s origins with archaeological evidence and an analysis of the forms typical to that order, it becomes clear that the development of the classical orders by the fifth century BCE was a complicated affair. It is unlikely that any one source can be pinpointed as the prototypical Doric building. Still, it is strongly possible that the Doric derived its forms from elements of traditional wooden construction combined with other functional advances which were made along the way and featured in building such as the temples at Corinth and Isthmia. Eventually, these disparate developments coalesced into the refined architectural order that we know today, comprised of forms which were developed largely in Greece. While, as the Egyptian examples demonstrated, the language of the Doric order may not exclusively be found in Greek architecture, the degree to which this language was used, refined and perfected by the Greeks has been matched by no others. Indeed, the presence of the Doric in the built environment today stands as testament to this remarkable achievement.

References

Allsopp, Bruce. 1965. A History of Classical Architecture. London.

Chitham, Robert. 1985. The Classical Orders of Architecture.

Dinsmoor, William. 1950. The Architecture of Ancient Greece. London.

Homer, Odyssey. Translation by Samuel Butler. (n.d. and lines unnumbered.)

Rhodes, Robin. 1987. “Early Corinthian Architecture and the Origins of the Doric Order,” American Journal of Archaeology, p. 477.

Robertson, D.S. Robertson. 1945. A Handbook of Greek & Roman Architecture. Cambridge.

Rykwert, Joseph. 1996. The Dancing Column. Cambridge, MA.

Vitruvius. De Architectura. IV.i.3. Translation by Frank Granger (London 1931)

Vitruvius. De Architectura. I.i.5. Translation by Rose, quoted in Hugh Plommer “Vitruvius and the Origin of Caryatids,” JHS (1979) 97-102.

Warren, Herbert Langford. 1919. The Foundations of Classic Architecture. New York..