Lise M. Hetland

The correct date is the first half of Hadrian’s reign. The building was not begun before 117, and probably dedicated about 126–8.... In the body of the Pantheon there is a preponderance of brickstamps of the 120’s, and it is upon this fact, more than any other, that the dating of the building is based.

— William L. MacDonald, The Pantheon, 1976¹

1. Problems with Dating the Pantheon

In the opening quotation from one of the most authoritative scholars on Roman architecture, William MacDonald, the Hadrianic dating of the Pantheon is presented as a proven fact. Yet this became the established view only late in the nineteenth century, when scholars turned away from variations on the previously accepted Augustan date.² This took at face value the main inscription fronting the portico: M. AGRIPPA. L. F. COS. TERTIVM. FECIT (Fig. 3.1), which seems to claim the building to have been erected by Agrippa in his third consulship (probably in 27 BC).³ Dio Cassius presumably took the inscription literally when he described the building in the second decade of the third century AD:

... in addition [Agrippa] concluded the construction of the building called the Pantheon. ... Agrippa ... wanted then to place there also [a statue of] Augustus and to bestow upon him the honour of having the work named after him; but since the prince did not accept either of these two honours, he had placed in the temple a statue of [Julius] Caesar pater, while in the porch he put statues of Augustus and himself.⁴

3.1. View of entablature and tympanum at night. (Courtesy of Fulvio Santus)

In effect, this text harks back to Agrippa’s creation of the first Pantheon on the same spot as the present structure. In AD 80, while the emperor Titus was absent in Campania, a fire damaged the whole region around and including the Pantheon. Again it is the Roman consul Dio Cassius who describes the devastations:

It consumed the temple of Serapis, the temple of Isis, the Saepta, the temple of Neptune, the Baths of Agrippa, the Pantheon, the Diribitorium, the theatre of Balbus, the stage building of Pompey’s theatre, the Octavian buildings together with their books, and the temple of Jupiter Capitolinus with its surrounding temples.⁵

The term “consumed” may exaggerate the damage done to the Pantheon itself, in the sense that scant trace has been found of repairs that may be attributed to the reign of the emperor Domitian (see Chapter Two).⁶ Another fire, caused by lightning, devastated the building in 110; an ancient source states simply that “The Pantheon was struck by lightning and burned.”⁷ This time a total rebuilding was required, resulting in the edifice that survives today.

The emperor Hadrian’s involvement with the Pantheon is briefly noted in two ancient sources. The first of these, once again by Dio Cassius, informs us:

He [Hadrian] transacted with the aid of the senate all the important and most urgent business and he held court with the assistance of the foremost men, now in the Palace, now in the Forum or the Pantheon or various other places, always being seated on a tribunal, so that whatever was done was made public.⁸

Written some hundred years after the present Pantheon was built, Dio’s Roman History does not in fact connect its rebuilding with Hadrian, since Dio attributes the building to Agrippa, as is clear from the passage by the same author cited earlier.⁹ However, theScriptores Historiae Augustae explained “Agrippa’s” inscription in another way, as an example of Hadrian’s modesty:

He [Hadrian] built public buildings in all places and without number, but he inscribed his own name on none of them except the temple of his father Trajan. At Rome he restored the Pantheon [instauravit Pantheum], the Saepta, the Basilica of Neptune, very many temples, the forum of Augustus, the Baths of Agrippa and dedicated all of them in the names of their original builders.¹⁰

For more than a century, this statement has been interpreted to mean that the Pantheon was built entirely during Hadrian’s reign, from the foundation to the dome. But it is intriguing that this emperor’s involvement is described in almost identical terms to the works of repair carried out by his successor, the emperor Antoninus Pius, who, as the Scriptores Historiae Augustae states: “... instauratum ... templum Agrippae.”¹¹ (These repairs, evidently not that extensive, are confirmed by brickstamps.)¹²

Another inscription, situated on the architrave directly under the more prominent Agrippan one, records repairs carried out by the emperors Septimus Severus and Caracalla, as has also been confirmed by some brickstamps found in the dome and in the intermediate block.¹³

No known ancient source, then, states that the emperor Hadrian was behind the actual construction of the Pantheon. It has been the brickstamps found in situ and around the monument that have been taken to indicate the Hadrianic date – yet it is precisely on the evidence of brickstamps that this will now be challenged.

2. Dating Roman Buildings by Brickstamps

A walk in Rome today reveals that bricks were quintessential for Roman Imperial buildings, all the more so since so many of them have been stripped of their marble coverings.¹⁴ A portion of bricks – we do not know the percentage – were imprinted with stamps (Fig. 3.2).¹⁵ These tend to name the owner of the brickworks (often an aristocrat, sometimes even a member of the imperial family), the foreman of the brickworks, very often the name of the operative (usually a slave or a freedman), and the place of production. Such information, frequently heavily abbreviated, could be featured individually or combined, and sometimes accompanied by ornamental symbols. The shape of brickstamps also changed, from the earliest simple rectangular ones used in the first century BC, via the open crescents in vogue toward the end of the first century AD, to crescents that later gradually became more closed, eventually becoming a complete circle during the later empire, when the rectangular-shaped stamps reappear again as well (Fig. 3.3).¹⁶

3.2. Brickstamp in situ, Caseggiato del Serapide, Ostia, from the Severan period, ca. AD 193–198. The stamp corresponds to CIL XV 371. (Courtesy of Lynne Lancaster)

3.3. Selected brickstamps, counterclockwise from top left: CIL XV 966.7 (first century BC); CIL XV 315 (Trajanic period, from the Pantheon); CIL XV 20 b (AD 115, from the Pantheon), made from the brickworks owned by M. Rutlius Lupus (hence “MRL”), dated by the name of the consuls Messalla and Pedo; CIL XV 801 (AD 123). (Drawings courtesy of Cesare Mecatti, compiled by Georg Herdt)

Dated brickstamps, that is, ones which mention the names of two consuls (who changed every year), are known from as early as 76 BC on a series of bricks from Velia.¹⁷ This was not a geographically isolated case, and dated brickstamps have also been found in other parts of the Italian peninsula, from Bologna in the north to Vibo Valentia in the south.¹⁸ Some time in the second decade of the second century AD, brickmakers in Rome also began to adopt this practice.¹⁹ The year 123 is particularly prolific for brickstamps; about 280 known types cite the consuls in office that year.²⁰

One of the first major studies on Roman brickstamps was carried out by Gaetano Marini in the late eighteenth century.²¹ But even before this, stamps from the Pantheon had been described in a less scientific manner by figures such as Ottavio Falconieri and Giovan Battista Piranesi (see Fig. 4.7).²² In a monograph of 1807 on the Pantheon, Carlo Fea concluded that the building was erected by Agrippa, as stated by the inscription, and had later undergone repairs during the reigns of the emperors Domitian, Hadrian, Antoninus Pius, Septimius Severus, and Caracalla.²³ This conception was challenged in 1892 when the French architect Georges Chedanne examined the structures and concluded that the whole building had been constructed in Hadrian’s reign.²⁴ Later the same year, Lanciani referred to Chedanne’s discovery as if it were already well known: “non si tratta veramente di scoperta, ma di conferma da fatto già conosciuto.”²⁵ One of the main objectives of the ensuing archaeological excavations, headed by Luca Beltrami in the period 1892–1893, was to check the dating of the rotunda and the portico, and to confirm whether or not the rotunda had been the laconicum of Agrippa’s Baths, an idea advanced by Fea and later replicated by Lanciani. Beltrami’s ensuing publication verified that there were no traces of heating that could have supported the laconicum theory, and more importantly, it concluded that the whole Pantheon was Hadrianic, and sooner or later scholars came to accept this as certain.²⁶

Just before these developments, Heinrich Dressel had been conducting the first methodologically based study on Roman brickstamps, which he published in 1891 as the fifteenth volume of the Corpus Inscriptionum Latinarum (conventionally abbreviated as CIL XV). In the 1930s and 1940s, his work was further developed by the German scholar Herbert Bloch.²⁷ Bloch systemized the corpus of dated brickstamps, and then applied them to the chronology of a long list of buildings in Rome and its surroundings.²⁸ Even if I advance here a different interpretation for the Pantheon to that of Bloch, it remains directly dependent on his method, and on the important discoveries made by him (and by Dressel before him).²⁹ Bloch’s work remains indispensable for the dating of many Roman buildings in Rome, in Ostia, and in Tivoli.

Bloch’s extensive research relied on prosopography, the study of the lives of historical figures, in this case the careers of brickmakers, who can sometimes be tracked as their status changed from slaves to foremen, and even to becoming the owners of brickworks. The information condensed in the stamps changed constantly to reflect such changes of status, and consequently individual molds were used for limited periods only. By cross-referencing this kind of information for stamps dated by virtue of the consuls named with other evidence about the buildings in which they were found (ideally, datable inscriptions), Bloch was able to map out a chronology for undated brickstamps. This important work was later supplemented by the researches of scholars such as E. M. Steinby, and more recently by Janet DeLaine.³⁰

Fundamental as Bloch’s methodology was to the business of dating buildings by means of the brickstamps they may contain, this does not mean that he should never be questioned. In the following pages, I reexamine the basis for dating the initiation of the works on the Pantheon to 118 or 119, near the beginning of Hadrian’s reign.³¹ Much of the structure of the building is brick-faced concrete (Fig. 3.4), and it is to the stamps on some of its bricks that we now turn.

3.4. Detail of exterior brickwork. (Courtesy of Maxim Atayants)

3. Bloch’s Interpretation

The theory of a Hadrianic date for the Pantheon has been left practically unrevised since 1937–1938. Bloch recorded a total of 184 brickstamps in and around the building.³² Excluding the brickstamps found not in situ or in parts not connected with the Pantheon proper (the so-called eastern wall, most probably part of the neighboring Saepta Julia, the Basilica Neptuni, and all of those found around the building), this still leaves 70 examples.

Just five of these in situ brickstamps bear consular dates. Four of them belong to the years 114, 115 and 116, each of which mentions the same brickmaker, M. Rutilius Lupus. Only one is dated to Hadrian’s reign, in particular the prolific year for brickstamps AD 123, as shown in Table 3.1.

Table 3.1. Dated brickstamps (consular dates) from the Pantheon

CIL XV no.	Date	External rotunda	Intermediate block	Total
19 a	114	1		1
20 b	115	2		2
23	116		1	1
549 a	123		1	1

While the consular dated brickstamps represent absolutely dependable evidence for the dating of the building, there are 62 brickstamps that have been dated with reasonable confidence by prosopography to varying time ranges. Of these, 19 brickstamps can be defined within a relatively tight time frame, and all of these are Trajanic/late Trajanic (100–117).³³ A larger group of 39 exemplars have been dated to a broader time range, the late Trajanic/early Hadrianic period (these are also dated by prosopography, but it has not been possible to pinpoint them any more precisely).³⁴ Then there are 4 brickstamps that have been dated to the Severan period, which can be attributed to the repair works already mentioned. Finally, there are just 3 brickstamps that have not been dated at all (neither by consular dates nor by prosopography), as shown in Table 3.2.

Table 3.2. Synopsis of the brickstamps from the Pantheon (rotunda, dome, intermediate block)

Date	Nos. in situ	% in situ
Trajanic	23	32.9
Late Trajanic/early Hadrianic	39	55.7
Hadrianic	1	1.4
Severan/post-Hadrianic	4	5.7
Undated	3	4.3
TOTAL nos.	70	100.0

In interpreting this material, a striking pattern emerges. Of the stamps that can be dated to either a relatively tight range or a specific year there is a great preponderance of Trajanic examples (4 + 19, equaling 23).³⁵ While it is theoretically possible that many of those in the Trajanic/Hadrianic group could tend toward the later period, the fact remains that there is only a single stamp with an absolutely secure Hadrianic date. Obviously, this creates difficulties for the theory of a Hadrianic dating for the Pantheon. Bloch confronted the problem by repeatedly maintaining that the Trajanic bricks should be understood as vecchie rimanenze dal periodo anteriore – “old remains from the preceding period.”³⁶ He argued that the Trajanic brickstamps could represent superfluous material that had been transported to the site of the Pantheon from other earlier projects in Rome, or that had been stockpiled. This may seem to be a satisfactory explanation, and for almost 70 years it has been accepted as such. However, Bloch was evidently concerned that the hypothesis of “old remains” was not an unassailable justification, for he felt obliged to elaborate his explanations further. He did this by focusing on two particular groups of in situ brickstamps.³⁷ One group comprised the 4 consular dated brickstamps from the years 114 to 116 made by M. Rutilius Lupus. The other group was made by one Anteros Severianus, the name that occurs on 21 brickstamps.

4. Testing Bloch’s Thesis – the “Special” Case of Rutilius Lupus

Bloch assigned the Pantheon to a group of buildings that contained brickstamps made by Rutilius Lupus during the period 114–117 but that were not erected until early in Hadrian’s reign.³⁸ In addition to the Pantheon, this group comprises buildings in Ostia, the so-called Le Quartier des Docks (better known as Portico di Pio), part of the Capitolium Group, the Piccolo Mercato, and Insula I.ix (excluding the Curia), as well as two further buildings in Portus, the Portico di Claudio and the so-called Palazzo Imperiale. In normal circumstances, the presence of numerous bricks stamped with consular dates in the years 114 to 117 would point to a construction date around this time, and so here was a problem for Bloch to negotiate.

That the brickmaker Lupus and M. Rutilius Lupus, the holder of the lucrative office of praefecti Aegypti for the period AD 113–117, were one and the same had already been affirmed by other scholars.³⁹ However, it is Bloch’s idea that while he was away in Egypt, his brickworks would have continued to produce bricks for stockpiling, with none (or few) being actually sold. While the identity of this Rutilius Lupus as the “MRL” on the stamps seems practically certain, the second element in Bloch’s hypothesis, the stockpiling during his stay in Egypt, has never been proved.⁴⁰ It makes little sense from the point of view of economics. Bloch suggested that Rutilius Lupus was speculating in the hope of obtaining higher prices for bricks that had been cured for longer than normal.⁴¹ However, the question of whether any such higher price would have covered the expense of storage, plus a premium for the risk, is left without answer. Bloch assumed that the absence of the emperor Trajan (who was away on a campaign against the Parthians) caused a temporary reduction in building activity in the capital and, moreover, that Rutilius Lupus expected to sell bricks at an inflated price due to increased demand once public building works took off again.⁴²

This may seem plausible, but if so, why did the owners of other brickworks not do the same as Rutilius Lupus? In fact, it seems that Bloch’s hypothesis was above all created to account for the brickstamps of 114–116 found in situ in the Pantheon, while maintaining his theory that the building only started in 118 or 119.

Bloch’s stockpiling theory was endorsed by Axel Böethius, and in particular the claim that bricks cured for a longer period may have obtained higher prices. As support, Boëthius invoked Vitruvius’s treatise on Roman architecture, which recommends a prolonged storage period of two years for sun-dried bricks.⁴³ Vitruvius even reported that the people of Utica only used bricks if they had been stored for five years.⁴⁴ It is understandable that sun-dried brick masonry could benefit from seasoning, but it is important to point out that they are structurally different from fired bricks.⁴⁵ Once bricks have been fired, prolonged storage will have brought little additional benefit, and besides, there is at least one documented example of bricks being used a very short period, months, not years, after they were made.⁴⁶ Through the inscription found on the Serapeum in Ostia, we know that this building was dedicated in AD January 127, yet its fabric contains some brickstamps with the names of the consuls of the year 126 – proving that they must have been used more or less immediately.⁴⁷ What is perhaps surprising is that it was Bloch who made the discovery. Even so, he maintained the benefit to bricks of being seasoned for a long period; he suggests years. The case of the Serapeum should have led him to question his dating of the whole group of “early Hadrianic” buildings, but evidently this was not something he was prepared to do. Instead, he argued that the Serapeum was an extraordinary case.⁴⁸

Bloch went even further in defense of his postponement of the use of bricks made in 114–117 by arguing against the chronological distribution of Rutilius Lupus’s bricks in the group of “early Hadrianic” buildings.⁴⁹ This is a significant point, since a sequence of utilization that more or less matches a chronological progression of stamps implies the direct take-up of production. On the other hand, the jumbling up of stamps of different dates could mean that earlier bricks awaited the later ones before being put to use. In a recent article on building activity in Ostia in the second century AD, Janet DeLaine has pointed out that there is indeed a chronological distribution of brickstamps in the buildings that Bloch had highlighted.⁵⁰ As DeLaine shows, the buildings situated around the so-called Curia, west of the Capitoline, were built sequentially, as indicated by both the building material, that is, dated brickstamps, and the building technique. The first to be completed (after the Curia) was an industrial building (I.ix.2), situated on the northwestern part of the insula. All of the in situ dated brickstamps are from the year 114, with none later, which indicates construction around 114–115. Clearly built up against this structure on the southwestern side is the Caseggiato del Larario (I.ix.3), and this has brickstamps found in situ only from the years 114 and 115, but none from the years before or after; this would be consistent with works being finished no later than 116. No dated brickstamps have been found in situ in the third building, the Casa Basilicale (I.ix.1), but since it is built up against I.ix.3, DeLaine concludes that construction took place around 117–118. She therefore dismisses Bloch’s claim that these buildings were all built circa 120. Furthermore, DeLaine has been able to show that another of Bloch’s early Hadrianic buildings, the complex commonly known as the Portico di Pio, likewise consisted of several separate bodies of fabric that had been built in distinct phases from 114 onward.

Another of Bloch’s early Hadrianic group is the so-called Palazzo Imperiale at Portus. He records 21 brickstamps in situ; of these 11 were made by M. Rutilius Lupus and dated by consular names to the year 115, while others are datable to the years 114 and 116.⁵¹ There is no reason why the construction of the Palazzo Imperiale was not concluded in or around 117, which increases the likelihood that Bloch’s early Hadrianic buildings were all built (or were substantially under way) in the last years of Trajan’s reign.

There is at least one aspect of Bloch’s theory that seems correct, namely, that the Pantheon was contemporary with the group of early Hadrianic buildings that also contain bricks from Rutilius Lupus’s works – only these buildings were in fact Trajanic.⁵²Thus, this revision of the dates of the buildings at Ostia and Portus to before 118 (Hadrian returned to Rome for the first time as emperor in July of that year) strengthens the case for bringing forward the dating of the Pantheon, too.

5. Another Special Case?

Bloch maintained that the series of brickstamps belonging to the Pantheon and other buildings that mention the brickmaker Anteros Severianus also represents a special group.⁵³ Those from the Pantheon represent by far the largest group found in situ (CIL XV nos. 811 a–f) in Table 3.3. Twenty-one brickstamps of this type come from the Pantheon proper (internal and external parts of the rotunda, the dome, and the intermediate block), that is, 30 percent of the total. Bloch assigned the 811 d and 811 f stamps to the second decade of the second century AD (these being frequently used in Trajanic buildings), while the series CIL XV 811 a–c he judged to be Hadrianic because they were found in the Pantheon and in Hadrian’s Villa, both of which he believed to have been from this time.

Table 3.3. Distribution of brickstamps CIL XV 811 a–f

CIL XV no.	Trajan’s Baths	Trajan’s Markets & Forum	Basilica Ulpia	Atrium Vestae	Pantheon
811 a–c					18
811 d	1	12	3		1
811 e			3	1
811 f	1		5	10	2
811 e or f		6	1
TOTAL	2	18	12	11	21

It is not in fact possible to be precise about the dates of CIL XV 811 a–c, which could span the late Trajanic and early Hadrianic periods.⁵⁴ What is clear, though, is that CIL XV 811d–f are earlier. Here, then, is another group of “problematic brickstamps” for Bloch’s dating of the Pantheon to the Hadrianic period, which he again explains as “old remains.”⁵⁵ Contrary to such a hypothesis is also the fact that the oldest types, CIL XV 811 d–f, have only been found in the lower parts of the Pantheon, which must necessarily have been built before the dome, where instead the types CIL XV 811 a–c are dominant (making up 12 out of the total of 21 brickstamps found). Once again, here is a chronological distribution, which argues against Bloch’s stockpiling or old-remains theory in that it points to the progressive release of the bricks to the market.

6. Revising the Date of the Pantheon

The main thrust of the preceding discussion, supplemented by the researches of Steinby and DeLaine, is to revise in some respects the dating proposed by Bloch for the Pantheon. The spread of dates for consular brickstamps is summarized in Table 3.1.

This presentation of the facts invert’s Bloch’s scenario. Instead of 23 problematic Trajanic brickstamps for a Hadrianic dating of the building, this now changes to 1 problematic Hadrianic brickstamp for a Trajanic dating. This stamp, CIL XV 549, is dated by the consuls named to the year 123, but its location at the junction with the columnar portico suggests that it saw use only after the rest of the building was already complete (see Chapter Seven). By contrast, 23 of all the brickstamps found in situ (or almost 33 percent of the total), may be dated to the reign of the emperor Trajan.⁵⁶ This is surely too great a quantity to represent pure accident or old remains, as maintained by Bloch. Neither Bloch nor Steinby have been able to date any more precisely the group of brickstamps attributable to the late Trajanic/early Hadrianic period (39 stamps, or 55 percent of the total of 70 found in situ).⁵⁷ It seems that part of this group was assigned to the early Hadrianic period primarily because of being found in the Pantheon, which by these scholars’ definition was built at this time – a classic example of circular reasoning.

The analysis of brickstamps from the Pantheon thus shows that it is very problematic to place the start of building as late as 118 or 119. It is this supposed date that made it necessary for Bloch to invent ad hoc hypotheses, and to explain the presence of so many Trajanic brickstamps by recourse to the notion of stockpiling by Rutilius Lupus, or the using up of surpluses from previous years. However, the present investigation suggests a more straightforward explanation – that the Pantheon started under the emperor Trajan. The simplest reason for the presence of brickstamps dated to the years 114–116 in both the rotunda proper and the intermediate block would be that they reflected the time the Pantheon was built.

As noted, many of the same types of brickstamps found in the Pantheon are found in presumably Trajanic buildings in Rome and Ostia. DeLaine’s demonstration that the bricks of this kind used at Ostia were not stockpiled but were, rather, brought to market in the normal fashion, and, what is more, in the late Trajanic period, makes it all the more likely that the construction of the Pantheon, too, was begun at this time.

7. What Happened with the Pantheon in the Period 110–118/119?

Let us leave the evidence of brickstamps and look at something else that puzzled Bloch: the interval between the destruction of the previous Pantheon by lightning, an event Jerome’s text dates to 110, and the initiation of rebuilding in 118 or 119. Bloch admits that such a long gap is peculiar.⁵⁸ However, he fails to realize that this period of inactivity, eight to nine years, is entirely his own creation. The fact is that his theory of a Hadrianic date for the Pantheon creates a void.⁵⁹ He imputed the delay to its status as an imperial project, presuming that work could not begin while the emperor Trajan was away, and which upon his death had to await the arrival in Rome of the new emperor Hadrian.⁶⁰

Bloch may also have thought that the rebuilding of the Pantheon could not start immediately after the fire of 110 because many other imperial constructions were being built or finished off around that time. Trajan’s major projects include Trajan’s Baths (for which the traditional date of inauguration is 109); Trajan’s Markets (supposedly finished in 110); Basilica Ulpia (concluded in 112) and Trajan’s Forum (substantially concluded in 112, although the works seems to have continued into the Hadrianic period), and the Atrium Vestae (finished in 113).⁶¹ Most of these projects were concluded by 112, and so why would the emperor wait to commence with the Pantheon? It might be thought that preparations for the Parthian wars represented an impediment, but Trajan did not leave the capital to lead the campaign until September or October 113.⁶² And in any case, it is not as if building projects were routinely suspended during military operations.

8. Heilmeyer’s Hypothesis

Ancient sources tell us that Apollodorus of Damascus was Trajan’s master architect and the designer of important structures, including Trajan’s Forum, Trajan’s Baths, and an audacious bridge over the River Danube.⁶³ Many scholars have endorsed this, and have gone on to propose Apollodorus as the author of other buildings constructed during the time of this emperor.⁶⁴ Some even argue that Apollodorus was the designer of the Pantheon.⁶⁵ One of them is Wolf-Dieter Heilmeyer, who in 1975 argued that the Pantheon was initiated during the reign of the emperor Trajan.⁶⁶ Heilmeyer’s hypothesis is based chiefly on stylistic evidence, especially the close affinities between the marble encrustation and the Corinthian capitals of the Pantheon and equivalent parts of Trajan’s Forum.⁶⁷ This interpretation has been challenged most forcefully by appeal to the evidence of brickstamps, as viewed in line with Bloch’s analysis. In an article dealing with the dating of the Large Baths of Hadrian’s Villa, A. C. G. Smith concluded that brickstamps support the conventional date of the Pantheon to the early years of Hadrian’s reign.⁶⁸ Yet if the evidence presented here is correct, the validity of such argumentation dissolves. Another objection to the involvement of Apollodorus has been seen to be his supposed execution at Hadrian’s behest, an event that Dio Cassius places in the earliest parts of his reign.⁶⁹ However, his narrative, as we have seen, is not entirely reliable, and in any case, a Trajanic start for the Pantheon in effect removes this objection, too.⁷⁰

Further parallels between the Pantheon and buildings thought to be associated with Apollodorus are explored elsewhere in this volume by Giangiacomo Martines, Gene Waddell, and Mark Wilson Jones. For myself, the most sustained similarities with the Pantheon are found in the hemicycles in the Baths of Trajan, whether in terms of spatial conception or in details of the composition and decoration (see Figs. 5.2, 5.7). Also comparable with the Pantheon is the articulation of the walls with alternating triangular and rounded pediments, and the coffering of one of the hemicycles that are still standing. The diameter of the largest hemicycles is considerable, and if we imagine putting two of them together, the result may have looked strikingly similar to the Pantheon.

9. Conclusion

Given the sparse information that we possess about Roman architects and their activities, it is surely problematic to name with any confidence the creator of the Pantheon. It is easier perhaps to say who it was not. Dated brickstamps give scant support for the claim that the emperor Hadrian was either its patron or its designer. There is no reason not to take the Trajanic brickstamps found in the Pantheon at face value. We know that it was reconstructed entirely from foundation to dome following the fire of 110, and it now seems probable that the planning of the project started soon afterwards. The site could have been cleared, the materials ordered, and works under way by around 114. Consequently, any influence that Hadrian may have had on the Pantheon must have been limited.⁷¹The construction of the rotunda may in fact have commenced in the last years of Trajan’s reign, in which case it is, of course, to this period that we should assign its conception and design. Being designed and partially built before 118, the monument really belongs to Trajan’s reign, even if it were completed by his successor.

This interpretation of the facts also illustrates the illogicality of the sometimes almost surgically clear-cut presentation of Roman buildings according to the sequence of emperors, and the implied role they might have played in creating successive styles. Relatively abrupt changes or “architectural revolutions,” as proposed by some scholars, are also to some degree a result of the modern need for characterization, rather than a description of what actually happened.⁷² In reality, the changes that did occur in Roman architecture happened relatively gradually, in response to the varied activities of many different individuals or schools of architects (at times working simultaneously in different directions), rather than being promoted and propelled by one dominant creative force.

The questioning of the dating of the Pantheon presented here will, it is hoped, lead us to question not just our understanding of this specific building but also what we understand by the widespread notion of “Hadrianic architecture.”⁷³ Entrenched “facts,” such as the date of the Pantheon and its connection to Hadrian, clearly need reevaluation. Going back over and revising the conclusions of earlier research may appear to be a negative kind of activity, but whenever they are too easily taken for granted it becomes imperative.

A more detailed account is published in Lise Hetland, “Dating the Pantheon,” Journal of Roman Archaeology 20, pp. 95–112.

1 William L. MacDonald, The Pantheon: Design, Meaning and Progeny, Cambridge, Mass., 1976, repr. 2002, p. 13.

2 In a volume summarizing recent conservation work carried out at the Pantheon under the direction of Giovanni Belardi, parts of the present building are still attributed to the Augustan period, although no real arguments are presented to support this supposition (Giovanni Belardi, Il Pantheon: storia, tecnica, e restauro, Viterbo 2006).

3 This inscription is usually understood to signify that the Pantheon was built (or dedicated) when Agrippa was consul for the third time, in 27 BC, yet some coins use the legend “M. AGRIPPA. L.F. COS. III” long afterwards. Consequently, the dedication may also be read as “built by M. Agrippa, son of Lucius, consul three times.” See David L. Vagi, Coinage and History of the Roman Empire, vol. 2, Chicago 1999, pp. 233–234; Ilaria Romeo, Ingenuus Leo: L’immagine di Agrippa, Rome 1998, pp. 19–45.

4 Dio Cassius, 53.27.1–2 (cf. trans. as Dio Cassius: Roman History, by Earnest Cary and Herbert Foster, Cambridge 1917).

5 Dio Cassius, 66.24.

6 One source records that “Many public buildings were erected in Domitian’s reign: ... and the Pantheon” (Chronograph of 354, for which see http://www.tertullian.org/fathers/chronography_of_354_00_eintro.htm; and Jerome’s Chronicles in Rudolf Helm, ed.,Eusebius Werke. Siebenter Band, Die Chronik des Hieronymus, Leipzig 1956; Abr. 2105, for which see also Malcolm Drew Donalson, A Translation of Jerome’s Chronicon with Historical Commentary, Lewiston 1996).

7 Paulus Orosius, “Pantheum Romae fulmine concrematum,” in Historiae adversum paganos, ed. C. Zangemeister, Leipzig 1889, 7.12.5; cf. Jerome in Helm 1956, p. 195; Eusebius of Caesarea, Chronicon, ed. Alfred Schoene, Berlin 1866, p. 219; Herbert Bloch, “I bolli laterizi e la storia edilizia romana,” Bullettino della Commissione Archeologica Comunale di Roma 64, 1937–1938, pp. 1–353; p. 116.

8 Dio Cassius, 69.7.

9 Dio Cassius may have used sources such as an autobiography by Hadrian and a work by Marius Maximus, also long since lost, but the Roman consul was no eyewitness (Ronald Syme, Emperors and Biography, Oxford 1971, pp. 113–117, 128 ff.; RonaldSyme, ed., Historia Augusta Papers, Oxford 1983, pp. 16, 30 ff; Barry Baldwin, “Dio Cassius on the Period AD 96–180: Some Problematic Passages,” Athenaeum 63, 1985, pp. 195–197. Book 69 of Dio Cassius’s Roman History on the life of Hadrian is known to us only in an epitomized version; it was written in the eleventh century by the Byzantine jurist, later monk, eventually the patriarch of ancient Trapezus, Ioannes Xiphilinos, for emperor Michael VII Dukas (1071–1078). We do not know how much is really Dio’s own work, and how much is that of Xiphilinos. Cf. Fergus Millar, A Study of Dio Cassius, Oxford 1966; P. A. Brunt, “On Historical Fragments and Epitomes,” Classical Quarterly 30, 1980, pp. 488–492.

10 Scriptores Historiae Augustae, Hadrian, 19.9–10: Cum opera ubique infinita fecisset, numquam ipse nisi in Traiani patris templo nomen suum scripsit. Romae instauravit Pantheum, Saepta, Basilicam Neptuni, sacras aedes plurimas, Forum Augusti, Lavacrum Agrippa; eaque omnia propriis auctorum nominibus consecravit.

11 Scriptores Historiae Augustae, Antoninus Pius, 8.2. On the interpretation of terms such as instaurare as used on inscriptions, see E. Thomas and E. Witschel, “Claim and Reality of Roman Rebuilding Inscriptions from the Latin West,” Papers of the British School at Rome 60, 1992, pp. 135–177; cf. G. Fagan, “Reliability of Roman Rebuilding Inscriptions,” Papers of the British School at Rome 64, 1996, pp. 81–93. Dio’s belief that the Pantheon was built by Agrippa could explain why he refers to the interventions by Hadrian and Antoninus Pius as restorations.

12 The brickstamps were incorporated in the Corpus Inscriptionum Latinarum (CIL), volume XV. Brickstamps CIL XV 424 a 1 (one example not in situ) and CIL XV 617.1 (one example not in situ) have been dated to the Antonine period by Steinby (E. M. Steinby, “La cronologia delle figliane doliare urbane dalla fine dell’età repubblicana fino all’inizio del III sec.,” Bullettino di archeologia cristiana 84, 1977, pp. 60 and 78).

13 The inscription reads: IMP CAES L SEPTIMIVS SEVERVS PIVS PERTINAX AVG ARABICVS ADIABENICVS PARTHICVS MAXIMVS PONTIF MAX TRIB POTEST X IMP XI COS III P P PROCOS ET IMP CAES M AVRELIVS ANTONINVS PIVS AVG TRIB POTESTAT V COS PROCOS PANTHEVM VETVSTATE CORRVPTVM CVM OMNI CVLTV RESTITVERVNT. The brickstamps nos. CIL XV 155 and 157 found in the intermediate block have been dated to the Severan period (193–211), and CIL XV no. 602, found in the dome, is similarly assigned to 198–211, see Steinby 1977, CIL XV 155–157, pp. 37–38; CIL XV 602, p. 92.

14 The sizes of Roman bricks were different from bricks most commonly used today, being larger, square shaped and only some 2.5 cm to 4.5 cm thick (see Chapter Four).

15 Only the last brick of each batch, or one in so many, was stamped to ease the counting process, according to J. W. P. Campbell, Brick in World History, Cambridge 2003, p. 48.

16 It is important to acknowledge that there are different and even opposite approaches to dating Roman buildings by means of brickstamps. Their very usefulness has also been questioned; see Esther Boise Van Deman, The Atrium Vestae, Washington 1909, and “Methods of Determining the Date of Roman Concrete Monuments,” American Journal of Archaeology 16, 1912, pp. 417–421. (For responses, see Bloch 1937–1938, p. 12; Steinby 1977, esp. p. 17).

17 D. Manacorda, “I diversi significati dei bolli laterizi – appunti e riflessioni,” in La brique antique et medieval: production et commercialisation d’un material, ed. Patrick Boucheron, Henri Broise, and Yvon Thébert (Collection de l’Ecole Française de Rome 272), Rome 2000, pp. 127–159.

18 Manacorda 2000. The latest nonurban dated brickstamps were made in Todi in AD 93.

19 The first known dated brickstamp from Rome, CIL XV 18, was made in AD 110.

20 Heinrich Dressel and Herbert Bloch list some 289 variants of brickstamps mentioning Apronianus and Paetinus, the consuls in AD 123 (Heinrich Dressel, Inscriptiones urbis Romae latinae, Berlin 1891; Herbert Bloch, I bolli laterizi e la storia edilizia romana. Contributi all’archeologia e alla storia romana (1937–1938), Rome 1947–1948 [supplement to Vol. XV, 1 of the Corpus Inscriptionum Latinarum; reprinted by Harvard Studies of Classical Philology LVI–LVII (1947), LVIII–LIX (1948)]). The issue of the brickstamps from the year 123 is very intricate, and existing explanations may not have grasped its entire complexity.

21 The principle of consular dates was understood by Marini and earlier scholars. He argued, for example, that a brickstamp with the name of one of the consuls for the year 142, found in situ in the Baths of Diocletian, represented material reused from an older building, instead of evidence for a hitherto unknown consul from the time of Diocletian (Bloch 1937–1938, p. 4). Although compiled between 1789 and 1799, Marini’s manuscript was kept in the Vatican Library until it was published by Gian Battista de Rossi in 1884 (in Rome) with the title Iscrizioni antiche doliari, and with the inclusion of important notes by Dressel.

22 Rodolfo Lanciani, Pagan and Christian Rome, Boston 1892, p. 158 (information on Falconieri); Francesco Piranesi, Seconda parte de’templij antichi che contiene il celebre Pantheon, Rome 1790, Tav. 28 and 29.

23 Carlo Fea, L’integrità del Pantheon rivendicata a Marco Agrippa, Rome 1807, p. 26. Fea highlighted one brickstamp he believed to be Trajanic (CIL XV 315), found during his investigations of the building in 1804. Being convinced that the core of the building was Agrippan/Augustan, he discharged the problem without any discussion; Fea 1807, pp. 27–28.

24 R. Phené Spiers, “Monsieur Chedanne’s Drawings of the Pantheon,” Journal of the Royal Institute of British Architects 2, 1895, pp. 175–182; cf. William C. Loerke, “Georges Chedanne and the Pantheon: A Beaux Arts Contribution to the History of Roman Architecture,” Modulus, 1982, pp. 40–55.

25 Lanciani 1892, p. 151, writes “... this is not a real discovery, but confirmation of facts already known” (from Fea’s investigation in 1804).

26 Luca Beltrami, Il Pantheon: La struttura organica della cupola e del sottostante tamburo, le fondazioni della rotonda, dell’ avancorpo, e del portico, avanzi degli edifici anteriori alle costruzioni adrianee. Relazione delle indagini eseguite dal R. Ministero della Pubblica Istruzione negli anni 1892–93, coi rilievi e disegni dell’ architetto Pier Olinto Armanini, Milan 1898.

27 Bloch (1937–1938, p. 7) acknowledges his great debt to Dressel’s work (1891), but points out that his treatment of undated brickstamps is erratic. Having identified brickstamps with the consular date 123, Dressel assumed that all of the bricks made by the same brickmakers should be ca. 123, ignoring the possibility that their activities could have extended over several years.

28 Bloch dated many important public buildings from the reigns of Domitian, Trajan, Hadrian, and Antoninus Pius, along with two buildings from the third and fourth centuries.

29 Dressel 1891; Bloch 1937–1938.

30 Steinby 1977, pp. 7–113; Janet DeLaine, “Building Activity in Ostia in the second century AD,” Acta Instituti Romani Finlandiae 26, 2002, pp. 41–102.

31 Bloch 1937–1938, p. 117.

32 Apart from Dressel’s publication of 1891, Bloch’s work was also based on the study by Guey, which was published the year before, 1936; J. Guey, “Devrai-on dire: Le Panthéon de Septime Sévère? A propos des estampilles sur briques recueillies dans ce monument, notamment en 1930 ou en 1931 et depuis,” Mélanges d’Archéologie et d’Histoire (Ecole Française de Rome) 53, 1936, pp. 198–249.

33 Bloch 1937–1938; Bloch 1947–1948; Bloch, “The Serapeum of Ostia and the Brick Stamps of 123 AD,” American Journal of Archaeology 63, 1959, pp. 225–240; Steinby 1977; DeLaine 2002. All of these authors concord on this dating.

34 For a complete list of brickstamps found in situ in the Pantheon, divided into categories, see Hetland 2007.

35 Apart from the 4 dated brickstamps, the following are assigned to the Trajanic period: CIL XV 314–315 (8); CIL XV 377 (1); CIL XV 693 (3); CIL XV 811 d–f (3); CIL XV 1008 (1); CIL XV 1106 (3). The combined total amounts to 19 brickstamps.

36 Bloch 1937–1938, pp. 112–117.

37 Bloch 1937–1938, pp. 14–19; pp. 316–320.

38 Bloch 1937–1938, pp. 87–102.

39 Inscriptiones Latinae selectae, ed. Hermann Dessau, Berlin 1906, 8658, 3, as reported by Bloch 1937–1938, p. 317.

40 Bloch 1937–1938, pp. 317–318.

41 Bloch 1937–1938, pp. 316–320; Bloch 1959.

42 Bloch 1959.

43 Vitruvius 2.3.1–4. Ever since Boëthius’s mention, Vitruvius’s recommendation has been used to support Bloch’s hypothesis; see Axel Böethius, “La datazione dei mattoni,” Eranos 39, 1941, pp. 152–156; Bloch 1959, pp. 225–240; A. C. G Smith, “The Date of the Grandi Terme of Hadrian’s Villa at Tivoli,” Papers of the British School at Rome 46, 1978, pp. 73–93.

44 Vitruvius 2.3.2.

45 T. Helen, Organisation of Roman Brick Production in the First and Second Century AD, Helsinki 1975, pp. 16–18.

46 See Hetland 2007 for details on the manual production of bricks made of clay from the Tiber valley.

47 Bloch 1959, pp. 225–240. In fact, 15 (or 41%) of the total of 36 brickstamps found in situ in the Serapeum were made in the years 125 and 126.

48 Bloch 1959, pp. 225–240, esp. p. 234: “It seems most remarkable, that in 125, and assuredly in 126, the supply of bricks manufactured before 123 to be used in the walls were virtually exhausted.”

49 Bloch 1937–1938, p. 113: “Come per tutti gli edifici del gruppo, M. Rutilio Lupo ha fornito mattoni anche per la costruzione del Pantheon: essi sono tutti datati. ... In altri termine: è rappresentato ogni anno della serie datata, ciò è identico a quanto abbiamo osservato nel ‘quartier des docks’ in Ostia (cfr. anche il Palazzo Imperiale in Porto), senza che si possa in nessuno dei casi desumere qualche cosa sull’andamento dei lavori, anzi non abbiamo nè per il Pantheon nè per gli horrea in Ostia il minimo indizio che la costruzione sia stata iniziata prima dell’avvento al trono di Adriano.”

50 DeLaine suggests that Bloch was aware of this fact, despite his contrary arguments. (DeLaine 2002, esp. pp. 42–43; see also p. 78). The brickstamps found in and around the building are from the years 114–117, which also may be an indication that the works were finished by 118.

51 Bloch 1937–1938, pp. 100–102.

52 Bloch 1937–1938, p. 114: “Con ciò è accertata la contemporaneità del Pantheon e delle costruzioni trattate sopra [i.e.. Il ‘Quartier des Docks’ (Ostia), La Casa dei Triclinii (Ostia), Il Portico di Claudio (Portus), Il Palazzo Imperiale (Portus)].”

53 Bloch 1937–1938, pp. 14–19; pp. 112–117.

54 In the last years of the reign of Trajan, Anteros Severianus apparently started using a new matrix, CIL XV 811 a–c. Bloch (1937–1938, pp. 112–113) emphasizes that it is not possible to see an immediate and absolute change; he does suggest, though, that by Hadrian’s reign, all of Anteros Severianus’s workers were using only this matrix. However, it is impossible to exclude the possibility that some of these types were produced contemporaneously, and so it is best to assign them a broad date range, late Trajanic/early Hadrianic.

55 Bloch 1937–1938, p. 112: “... tuttavia egli disponeva ancora, quando il Pantheon fu costruito, di vecchie rimanenze munite dei timbri 811 d e f : 4 es. (3 in situ + 1). 811 d, f appartengono, come abbiamo dimostrato, al primo e al principio del secondo decennio del secondo secolo d. Cr.” English translation: “... after all he [Anteros Severianus] still disposed, when the Pantheon was constructed, of old remains with stamp types 811 d e f : 4 (3 in situ + 1). 811 d, f belong, as we have demonstrated, to the first and the beginning of the second decade of the second century.”

56 Bloch 1937–1938, pp. 112–117.

57 Often it seems that it is Bloch’s and Steinby’s assumption of a Hadrianic date for the Pantheon that guides the date assigned for undated brickstamps; for more details, see Hetland 2007.

58 Bloch 1937–1938, p. 116: “È strano che dopo la catastrofe nel 110 il monumento restò come era, salvo forse insignificanti riparazioni.”

59 This was brilliantly pinpointed by Wolf-Dieter Heilmeyer, “Apollodorus von Damaskus – der Architekt des Pantheon,” Jahrbuch des Deutschen Archäologischen Instituts 90, 1975, pp. 316–347, esp. p. 328.

60 Bloch 1937–1938, p. 116. “L’attuale costruzione non può essere anteriore al 117, e se si considera che si tratta di un monumento imperiale ... non può essere dubbia la conclusione che il totale rifacimento si iniziò solo dopo che l’imperatore era venuto a Roma, cioè dopo il principio del luglio 118, nella seconda metà del 118 o nel 119, ossia il principio della costruzione cade ancora nell epoca di Apollodoro di Damasco, che forse non fu estraneo al grandioso progetto.”

61 For the date of Trajan’s Baths, see J. C. Anderson, Jr., “The Date of the Thermae Traiani and the Topography of the Oppius Mons,” American Journal of Archaeology 89, 1985, pp. 499–509 (who argues for an earlier, Domitianic, initiation of the works); G. Caruso and R. Volpe, s.v. “Thermae Traiani,” in Steinby 1995–1999, vol. 5 1999, pp. 67–69. For Trajan’s Markets, see Lynne Lancaster, “The Date of Trajan’s Markets: An Assessment in Light of Some Unpublished Brick Stamps,” Papers of the British School at Rome 63, 1995, pp. 25–44; E. Bianchi, “I bolli laterizi dei Mercati Traiani,” Bullettino di archeologia cristiana 104, 2003, pp. 329–352; L. Ungaro, s.v. “Mercati di Traiano,” in Steinby 1995–1999, vol. 3, Rome 1997, pp. 241–245. For Trajan’s Forum, see James Packer, s.v. “Trajan’s Forum,” in Steinby 1995–1999, vol. 5, Rome 1999, pp. 348–356; E. Bianchi, “I bolli laterizi del Foro di Traiano: il catalogo del Bloch e i rinvenimenti delle campagne di scavo 1991–1997 e 1998–2000,” Bullettino della Commissione Archeologica Comunale di Roma 102, 2001, pp. 82–120. For Atrium Vestae, see R. P. Scott, s.v. “Atrium Vestae,” in Steinby 1995–1999, vol. 1, Rome 1995, pp. 138–142.

62 Anthony R. Birley, Hadrian: The Restless Emperor, London 1997, p. 191.

63 Dio Cassius (69.4.1) states that Apollodorus built the forum, the odium, and the gymnasium for Trajan in Rome. The bridge over the Danube (the ancient Ister) was described by Procopius as being built by Apollodorus (Procopius, Buildings, IV.6.12–13). Dio Cassius (68.13) also describes the building of the bridge, but he does not attribute it to Apollodorus. The Historia Augusta (XIX.13) states that Apollodorus assisted the emperor Hadrian with the constructing and the moving of the colossal statute of the Moon (situated on the western side of the Flavian Amphitheatre, the Colosseum).

64 Several scholars credit Apollodorus with Trajan’s Forum (including Trajan’s Column), Trajan’s Markets, Trajan’s Baths, and the bridge over the Danube; see inter alia William L. MacDonald, The Architecture of the Roman Empire, vol. 1, An Introductory Study, London 1965, pp. 133–134; John Ward-Perkins, Roman Imperial Architecture 2, London 1981, p. 9; Mark Wilson Jones, Principles of Roman Architecture, New Haven 2000, pp. 21–24 (who, however, leaves the Markets as an unresolved attribution and gives further references to others who have given Apollodorus a more extensive portfolio).

65 Bloch 1937–1938, p. 116; Heilmeyer 1975, pp. 316–347; Alessandro Viscogliosi, “ Il Pantheon e Apollodorus di Damasco,” in Tra Damasco e Roma: L’architettura di Apollodoro nella cultura classica, ed. Festa Farina et al., Rome 2001, pp. 156–183;Wilson Jones 2000, pp. 192–193, 212; Gerd Heene, Baustelle Pantheon: Planung, Konstruction, Logistik, Düsseldorf 2004.

66 Heilmeyer 1975, esp. p. 328. In a 1994 article, Lothar Haselberger endorses Heilmeyer’s suggestion and argues for a Trajanic dating; L. Haselberger, “Ein Giebelriss der Vorhalle des Pantheon. Die Werkrisse vor dem Augustusmausoleum,” Mitteilungen des Deutschen Archäologischen Instituts, Römische Abteilung 101, 1994, pp. 279–309, esp. pp. 296–298.

67 Heilmeyer 1975. Stylistic similarities between the architectural decoration of the late Trajanic and early Hadrianic periods have also been linked to Apollodorus by Strong (D. E. Strong, “Late Hadrianic Architecture Ornament in Rome,” Papers of the British School at Rome 21, 1953, pp. 118–151).

68 Smith 1978, pp. 75–93. Smith accepted Bloch’s theory that bricks were cured for a period of years before they were used (Bloch 1937–1938 and 1959, pp. 225–240; Smith 1978, pp. 75–93). He also believed that the Pantheon was built contemporaneously with the earliest parts of Hadrian’s Villa (Smith 1978, p. 77), which he dates to sometime after 117. Smith’s hypothesis found support from Mary T. Boatwright in her work on buildings in Rome connected with Hadrian, who called Heilmeyer’s Trajanic dating of the Pantheon “quite implausible” (Mary T. Boatwright, Hadrian and the City of Rome, Princeton 1987, p. 13).

69 It is listed under the first year of Hadrian’s reign, in 117 (Dio Cassius, 69.4).

70 Heene (2004) also follows Heilmeyer’s late Trajanic time scheme for the Pantheon, though without any explanation. Heene offers new thoughts as to how the construction of the Pantheon proceeded that contrast with the intriguing (if not very feasible) rope-technique solution put forward by Taylor (Rabun Taylor, Roman Builders, Cambridge 2003); cf. MarkWilson Jones, “Review of R. Taylor’s Book: Roman Builders,” Journal of Roman Archaeology 16, 2003, pp. 557–560.

71 The idea that the emperor Hadrian was involved in, perhaps even responsible for the design of, the Pantheon is often mentioned in the scholarly literature: F. E. Brown, “Hadrianic Architecture,” Essays in Memory of Karl Lehmann, ed. L. F. Sandler, New York 1964, pp. 55–58; Kjeld de Fine Licht, The Rotunda in Rome. A Study of Hadrian’s Pantheon, Copenhagen 1968; MacDonald 1976, pp. 11–12; H. Stierlin, Hadrien et l’architecture romaine, Fribourg 1984; Boatwright 1987, pp. 30–31; P. Gros, “Hadrien architecte. Bilan des recherches récentes,” Hadrien empereur et architecte, ed. M. Mosser and H. Lavagne, Paris 1999, pp. 33–53, esp. pp. 48–57; E. Salza Prina Ricotti, “Adriano – architetto, ingegnere e urbanista,” Adriano architettura e progetto, Milan 2000, pp. 41–44; Ricotti, Villa Adriano: il sogno di un imperatore, Rome 2001; D. Danziger and N. Purcell, Hadrian’s Empire: When Rome Ruled the World, London 2005, pp. 17–18.

72 The term “the Roman Architectural Revolution” was first coined by MacDonald (1965, pp. 41–46), and later adapted by another influential scholar, Ward-Perkins (1981, pp. 97–120).

73 The expression “Hadrianic architecture” was first used by Italian scholars such as Promis and Rivoira, see C. Promis, “Gli architetti e l’architettura presso i Romani,” Reale academia delle scienze di Torino: Memorie 28, 1873, pp. 177–180; Giovanni T. Rivoira, “Di Adriano architetto e dei monumenti Adreanei,” Rendiconti della Accademia dei Lincei 18, fasc. 3, 1909; Rivoira, Roman Architecture and Its Principles of Construction Under the Empire, with an Appendix on the Evolution of the Dome Up to the 17^thCentury, New York 1972, p. 118. It was taken up by Toynbee, Snijder, and Strong; see J. M. C. Toynbee, The Hadrianic School, a Chapter in the History of Greek Art, London 1934, esp. p. iii; Geerto A. S. Snijder, “Kaiser Hadrian und der Tempel der Venus und Roma,” Jahrbuch des Deutschen Archäologischen Instituts 55, 1940, pp. 1–11; D. E. Strong, “Late Hadrianic Architecture Ornament in Rome,” Papers of the British School at Rome 21, 1953, pp. 118–151, and later amplified by Brown (1964). Since then, the idea of “Hadrianic” architecture has received numerous endorsements, of which the following is a selection: MacDonald 1965, pp. 94–121, pp. 129–137; MacDonald 1976; William L. MacDonald, “Hadrianic Circles,” Journal of the Society of Architectural Historians43, 1993, pp. 394–408; C. F. Giuliani, “Volte e cupole a doppia calotta,” Mitteilungen des Deutschen Archäologischen Instituts, Römische Abteilung 82, 1975, pp. 329–342; Boatwright 1987, p. 8; D. M. Jacobson, “Hadrianic Architecture and Geometry,” American Journal of Archaeology 90, no. 1, 1986, pp. 69–89, esp. pp. 57–71; Wilson Jones 2000, esp. pp. 93–100; Anna Maria Reggiani, “Villa Adriana. Riflessioni per la conoscenza di un unicum,” Adriano: architettura e progetto, Rome 2000, pp. 3–8, esp. p. 7; E. Salza Prina Ricotti, Villa Adriano: il sogno di un imperatore, Rome 2001, esp. pp. 21–22.