An Unexpected Journey-The French Expedition of Charles Fossey at Hatra (Iraq)

This paper deals with the overlooked French Expedition of the famous architect Charles Fossey at the end of the 19th century at Hatra (Northern Iraq) using information of unpublished documents from the Archive Nationale de France. The collected data shed new light on the understanding of Small Shrine 2, one of the fourteen small shrines built inside the districts of the city.


INTRODUCTION
The archaeological site of Hatra, al-Hadr in Arabic, is located at about 80 kms from Mosul (Iraq) in northern Mesopotamia. During the 2 nd and 3 rd century AD, the centre became the capital of an important buffer state placed between the Roman and Parthian Empires (Hauser 2000;Venco Ricciardi 2008;Hauser 2009;Foietta 2018: 141-150). The settlement, founded during the Post-Assyrian period (end of 5 th -4 th century BC) 1 , flourished during the first two centuries of the Common Era. The city was delimited by a double pseudo-circular curtain wall defining a large built area of about 300 ha. The complex urban layout was framed by a tortuous street network, demarcating districts, where dwelling areas, small shrines, and monumental funerary buildings are recognizable. In the centre of the city is the Temenos, built in ashlar blocks, where the most important temples were located. 2 Numerous archaeological, historical, and epigraphic questions have been answered, thanks to different archaeological expeditions and explora-tions carried out during the 19 th and 20 th centuries, from the first preliminary expeditions by English explorers, to the contemporary archaeological excavations of the Iraqi, Polish, and Italian archaeologists. 3 This paper focuses on the overlooked French Expedition to Hatra, directed by Architect Charles Fossey at the end of the 19 th century, which was the first archaeological excavation at the site, furnishing new suggestions about Small Shrine 2, built in the city layout.

BEFORE THE FRENCH ARCHAEOLOGICAL EXPEDITION: THE ENGLISH AND FRENCH EXPLORERS AT HATRA
In May1836, John Ross, an English surgeon working in Baghdad, left the capital with the purpose of reaching Hatra. It was not a well-planned undertaking, for several reasons. It would be detailed for the first time a few years later, along with a subsequent trip, in a long article published on the Geographical Journal in 1839 (Ross 1839: 443-470). During that first visit, which lasted only a few hours, the small group was forced to flee from the site due to an attack from the Shammar, a nomadic local tribe that controlled the area ruled by the Pasha of Mosul (Ross 1839: 443).
Their second attempt, the following year, was better organised by involving local chiefs of the Shammar tribes to prevent reprisals from their groups as well as attacks from the other big local tribe, the belligerent Aneizah. This allowed Ross to visit Hatra safely for several hours on 15 May 1837 (Ross 1839: 460, 463). On that occasion, Ross drew the first simplified plan of the site, reporting for the first time the Sasanian circumvallation wall (Hauser, Tucker 2009;Hauser 2013), the main pseudo-circular wall, the inner wadi, at the time registered as a channel, several funerary buildings in the western part of the city, mounds for the residential areas in the eastern part of the city, and the central Temenos, containing some buildings (Ross 1839: 467-470) (Fig. 1). The paper wrongly reported a diameter of approximately 2 miles, together with other incorrect data about the main curtain wall, such as the ones regarding the regular distance between the defensive towers and the homogeneity of the building technique for the fortifications (Ross 1839: 467). Nevertheless, the information reported have been of great importance and usefulness to scholars throughout the years. Ross also identified the two funerary buildings N and E outside the curtain wall, in front of the city gates (Ross 1839: 467-468). Such ruins were still clearly visible during the German Expedition, directed by W. Andrae, and were accurately documented in his fundamental volume published in 1912. Ross also mistakenly reported a strict division of the city in a western, residential part and an eastern part assigned to the necropolis, which was rectified by scholars in later studies (See the plan of Ross 1839: 470).
Ross' analysis focused on the Temenos central complex, particularly on the Great Iwans and Twin Iwans, suggesting they served as a palace or a temple, the latter function being confirmed by the systematic Iraqi investigations and cleaning of the site undertaken in the 1950s 4 (Fig.2). The English author provided the general measures of the ruins, reporting that there were less blocks and debris in the eastern part than in the western one, which featured important constructions, such as the Great Iwans and the Twin Iwans (Ross 1839: 468-470).
Ross describes the iwans from south to north. He started by analysing the small south iwan and preceded to the small north iwan, as they are called, from the Great Northern Iwan. He indicates the approximate measures of the rooms and describes the architectural decoration, focusing on the 'flat' masques on the blocks and the sculpted voussoirs in situ (Ross 1839: 468-469) 5 . The architectural decoration was described at length by explorers and archaeologists of the time; the historical photographs by G. Bell and W. Andrae of the voussoirs and decorated ashlars from the Great and Twin Iwans are well known and famous to all the scientific community, along with their interesting drawings. 6 Strangely enough, Ross does not report the presence of statues on the ground at the site, although they had been mentioned by members of the local tribes.
In the same year of Ross' trip, the site was briefly visited by H.B. Lynch, English Lieutenant, whose expedition had the purpose of drawing a trigonometric map from the Armenian mountains to Baghdad. In his short report however, the author did not mention the ruins of Hatra, even if he places the site on the map (Lynch 1839).
During the spring of 1840, a group composed by E.L. Mitford, A.H. Layard, Mr Rassám, and W. Ainsworth, together with an expert local guide and a convoy sent by the Pasha of Mosul, visited North Mesopotamia. A first report of this expedition was published by W. Ainsworth in 1841 on the Journal of Royal Geographical Society and republished using the same data in Travels and Researches in Asia Minor, Mesopotamia, Chaldea and Armenia II (Ainsworth 1841). It provided a more detailed description of the ruins at the site and rectified Ross' division between a necropolis and dwelling area. Also, the tortuous inner wadi was correctly attributed and no longer considered as a channel. A long part of the paper is devoted to the study of the ruins of the central Temenos, with the publication of some drawings by A.H. Layard of decorated blocks with masques from the Iwans and a sketch of the elaborated lintel placed between the Square Temple and the South Iwan (Fig. 3). A façade of one of the funerary stone buildings of the city, which is difficult to identify, is also represented. 7 Also, quoted here for the first time are the Zengid inscriptions engraved on the Great Iwans mentioning the names of the Atabeg of Mosul (Ainsworth 1841: 14;Andrae 1912: 108;Aggoula 1991: X;Parapetti, Venco Ricciardi 2013: 220).
Layard's report was published only several years later (1891) in Transactions: Vol. VII. New Series. His description of the city mostly coincides with the accounts given by Ainsworth and Ross, while also focusing on water resources at the site (Layard 1891: 64-65). The graphic documentation drawn on tracing paper by Layard is extremely interesting. His plan of the Temenos identifies many of the important architectural complexes, which would be uncovered only in the later works by Andrae and Iraqi archaeologists (Layard 1891) (Figs 4-5). In 1897, Ch. Jacquerel, Ingenieur des Ponts et Chaussées at Mosul, published a short paper in French after a visit to the site, under the title 'Les Ruines de Hatra' on the Revue Archéologique. Jacquerel reported some significant information about the urban layout and the fortification, identifying on the field the most important elements of the curtainwalls and the line of Sasanian circumvallation. 8 Jacquerel also accurately described the central ruins of the Temenos, publishing a plan of the Great Iwans complex and of the Twin Iwans, the latter represented as a single room due to the rubble found inside, just as in Ross' previous drawings. The French scholar also sketched some drawings of the voussoirs, door frames, and lintels, including the one situated between the Square Temple ( Fig. 6) and the Great South Iwan, as well as the inscription H246. 9

THE FRENCH EXPEDITION OF CHARLES FOSSEY -A FORGOTTEN EXPEDITION
In 1898, after submitting a first request to Constantinople for permission to excavate at Çatal Tepe in Turkey, the French architect Fossey, was authorized to start working at al-Hadr (Hatra), in Mosul district. Fossey's expedition proposal came at a particular time, that is when France was falling behind other European nations, especially Germany, in the field of excavations and archaeological research in the Ottoman Empire territories (Chevalier 2002: 63-112). Fossey submitted his request, after the failed attempt of creating a first 'Mission de Costantinople' -suggested by A. Joubin and J. Cambon, consul at Constantinople (Chevalier 2002: 83-89) -and when the 'École française d'Athene', directed by T. Homolle had just advanced a similar proposal (Chevalier 2002: 89-90 (Aggoula 1991: 123;Beyer 1998: 77). under the direction of the Byzantinist Charles Diehls. Fossey had previously studied at the School of Athens for three years, starting from 1894, and had spent one year at the 'Institut Français du Caire', building contacts that turned out to be useful to obtain the permission documents from the Sacred Door (Nougayrol 1946: 25).
Having already spent part of the money at Mosul for his two-month stay without the excavation permit (irade), Fossey decided to reach Hatra immediately after its release at the beginning of January. The field report and the story of his travel are preserved in a folder of unpublished documents stored at the Archives Nationales de France -Paris (F17/2963). 10 Without having viewed such documents, B. Aggoula mentioned Fossey's expedition in his book on Hatra inscriptions, suggesting that the French archeologist did not even visit the site directly.
In Fossey's folder the information reported in a letter he sent to the French Ministry dated 30 April 1899 (Fig. 7) is of great relevance (Appendix 1), as are the enclosed photographs, which remain the first pictures ever taken at the site (Figs 8-14). 11 The letter, which Fossey wrote to obtain new funds for a second mission that would focus mainly on the Temenos area, reports two field trenches. However, Fossey's expedition did not last long, as the attacks of local tribes forced Fossey and his team to seek refuge inside the Temenos for several days.
The exact location of the French trenches became known thanks to the data published in 1912 in W. Andrae's volume, where the areas explored more or less ten years earlier are mentioned ( Fig. 15) (Andrae 1912: 61, 70). A first trench (a) reaches across a landfill located west of Small Shrines 3 and 4, in the western part of the city. The area is placed precisely inside urban area AU16 and in squares l11 or l14 of Andrae's grid (Foietta 2018: 310-313;Andrae 1912: 61).
The second trench (b) was dug in correspondence with the Small Shrine 2, named Gebäude T by Andrae, and with a partially excavated house added westward (Andrae 1912: 70). This is where the French expedition focused its efforts; the same area that would later be explored in depth by an Iraqi expedition in the 1950s (Safar 1952: 37-52). The archive folder does not include any plans of the explored rooms, nor a more detailed documentation about the trenches and works.
The French Expedition also explored and surveyed the Temenos area, probably without carrying out any soundings. The tent camp for the workers and for Fossey's staff was certainly pitched in this area where, according to one of the pictures, the most remarkable pieces from this expedition were identified: a relief depicting an attendant or a god, a dog on a pedestal, and a standard, now preserved at the Istanbul Archaeological Museum (Fig. 13). 12

THE DISCOVERY FINDINGS
Fossey's letter does not mention any object from the landfill area (a) (Fig. 15); here a large quantity of pottery sherds and organic layers was probably discovered, as successively reported by W. Andrae for similar areas or for another area excavated by J. Ibrahim north of the Temenos (Urban area 10 -AU10) (Ibrahim 1997(Ibrahim -1998Foietta 2018: 209).
From the area of Small Shrine 2 and the house built westward comes a Hatran inscription, identified with H28, engraved on two blocks (56 x 23 cm e 57 x 32 cm) and probably belonging to the base of a statue (Aggoula 1991: 22-23;Beyer 1998: 35) (Fig. 11). Interestingly, the inscription reports the name of Sanatruq II (AD 200-241), the last king of Hatra, and his wife and son, the last one with the title pšgrb', usually translated as 'crownprince' (Gnoli 2002: 79). It is probably a dedication from the statue of Sanatruq II's wife, btsmy' (Marcato 2018: 45), 'Daughter of Samya', placed inside or in front of Small Shrine 2, as customary for several small shrines in the city, or in the nearby west dwelling area. The French expedition left the inscription at the site due to the weight of the blocks and it was later found and published by the Iraqi Archaeological Expedition that cleared the area near the shrine (Safar 1953: 15).
A statuette of an enthroned woman on a base with an inscription fragment, a bust of a naked male, a small statue of a warrior god with a sceptre (Fig. 12), and another fragment were also collected in the area of Small Shrine 2.
On the basis of the iconography, the statuette of the woman (h. 29 cm), whose head is not preserved, could be most probably be interpreted as a goddess on a throne, with a dog sculpted to her right and another crouching animal to her left (Fig. 16). She wears a tunic with sleeves at the elbow and a long garment with an elaborate drapery. A long girdle tied at the waist softly falls on her the knees. Her neck and breasts are adorned with two interesting necklaces: a short torques with small pendants and a long necklace with loop-in-loop chain with a round central medallion. She holds a palm branch, or a feather, in her left hand, and a small oblong object (stylus or distaff) in the other hand. The analysis and study of the sculpture has been proposed by L. Dirven, finding convincing comparisons inside and outside Hatra (Dirven 2013b). According to preliminary works by H.J.W. Drijvers and H. Ingholt, the goddess should depict Atargatis, but an identification with Allat or another local goddess of the Hatrean pantheon is far more likely, as suggested by Dirven (Dirven 2013b: 148-153). The statue is now preserved at the Archaeological Museum of Istanbul (inv. no. 3831), following the partition law of antiquities, which established that 2/3 of the findings belonged to the Ottoman Empire, while the remaining 1/3 could be brought to the foreign state (Chevalier 2002: 497-500). As stated by Dirven, the museum acquired the statue, a few years after its discovery (1908).
Given the image preserved in Fossey's folder (Fig. 13), the famous relief with dog (55 x 45cm), attendant/god and standard (samya) was probably discovered inside the Temenos area, possibly close to the Great Iwans. 13 The object, now preserved at the Archaeological Museum of Istanbul (inv. no. 3829) (Homès-Fredericq 1963: 56), was also studied in detail by Dirven in the same paper examining the enthroned goddess (Dirven 2013b: 151-152) (Fig.  17). On the right of the relief, a dog is engraved in left profile on a plinth. A keeper or a god with a spear (?) in the right hand and a sword on the left one is represented on the left part of the relief. Between the two described elements, an elaborated standard (samya) is sculpted. The dog on the pedestal is usually interpreted as the representation of a statue of the god Nergal, considering also the title Nergal klb', attested several times in Hatran inscriptions, and translated differently as 'Nergal the dog', 'Nergal of the dog' or 'Nergal keeper of the dog' (Dirven 2013b: 151, footnote 43).
The naked bust of a man, as indicated by Fossey, was probably brought to France and its precise location at the moment is unknown, while the small headless deity, probably in marmar (gypse), was taken to Istanbul as per the partition law (Fig. 12).

SHEDDING A NEW LIGHT FROM AN ARCHIVE RESEARCH. SMALL SHRINE 2: COMBINING DATA FROM THE FRENCH AND IRAQI EXCAVATIONS
An Iraqi Archaeological Expedition, directed by F. Safar, excavated the Small Shrine 2 and the western house built close to it during the 1950s (Safar 1952: 37-52) (Fig. 18a, b). The shrine is located south of the Temenos along the so-called south circumvallation, where the Small Shrines 1 and 8 were also built. 14 The shrine is similar in shape to other religious complexes situated in the dwelling area, with a wide antecella (16.2 x 5.9 m) and a narrow square cella (5.1 x 5.0 m), T-shaped shrine (Jakubiak 2014: 72-74) (Figs 19-20). Two square basements flanked the entrance were located, and were probably the plinths for the statues or the guardian lions, as in the case of Small Shrine 1.
Along the walls of the cella there are no basements, which in other cases sustained statues of gods, kings or lords, priests, and nobles. Two rooms were built on both sides of the cella during a second phase of the shrine.
These rooms were probably part of the domestic complex built to the west. According to Safar, the domestic area should be interpreted as part of the palace of Sanatruq II, the last king of Hatra, even if he did not provide further data to support this idea (Safar 1951).
Several goddesses statues have been found inside the shrine: -a goddess on a round bas-relief resting on a carved crescent moon (Safar, Mustafa 1974: no. 193) (Fig. 21); -a fragmentary female statue. A tunic covering the lower part of her body drapes over her shoulder, a necklace with a medallion decorates her naked breasts and a partially broken crescent moon extends from her shoulders. A diadem on her head (Safar, Mustafa 1974: no. 194) (Fig. 22); -a relief with three standing goddesses wearing long tunics. The goddess on the right shows a halo and a moon crescent on her upper back (Safar, Mustafa 1974: no. 192) (Fig. 23).
A burner was also discovered in the antecella (h. 90 cm, w. 35 cm, t. 30 cm). On its main face is carved a bearded male figure, who has been identified with Nergal for the similarity of the iconographic features with the famous 'Cerberus relief ' from Small Shrine 1, or better a Zaqiqu, given the presence of the inscription H13 on it, mentioning this particular god (Fig. 24). 15 According to the archaeologists who excavated the shrine, Small Shrine 2 was dedicated to Atargatis for the iconography of the female goddess (Safar, Mustafa 1974: 204) (Fig. 22), while according to K. Jakubiak, it was dedicated possibly to Nanaya, Atargatis, or to another local goddess (Jakubiak 2013: 93-94;Jakubiak 2014: 72-74).
The findings uncovered by the French Expedition, whose location is impossible to attest precisely, must be added to the statues found by the Iraqi Archaeological Expedition. The presence of the enthroned goddess (Fig. 16) confirms the hypothesis that the entire shrine could be devoted to an important goddess, who was worshipped alongside other 'secondary' deities, as was a common practice in smaller shrines at Hatra. The recovery of the small warrior deity statue with a sceptre must be considered in this sense. The small statue could be identified with a local god, quite similar to others recovered in other minor shrines. The naked male bust could be identified with a fragment of a Herakles/Nergal, although only a direct survey of the statue would settle the issue. 16 The left arm seems to be in fact at a right angle as to hold something, perhaps a cup or the apples of the Hesperides, similarly to other statues of Heracles/Nergal at the site (Fig. 12).
The inscription (H28) mentioning the mother of Sanatruq II's son, 'bad Samya, could come both from the area of the small shrine, given the numerous dedications of statues discovered inside the small shrines, or from the domestic complex, reinforcing Safar's suggestion that this was the palace of Sanatruq II, although the dimensions and features of the rooms do not seem to support this assumption. It is interesting in any case that the statue of Sanatruq II's wife and mother of the crown-prince was not discovered during the French and Iraqi Excavations. For this reason, it is possible to suggest that the statue was most likely located in an area of the house that had not been excavated yet, or that the buildings (shrine and house) were sacked during the final siege of Hatra (AD 240/1), or (more plausible) that the 'queen' statue was incomplete and still had to be placed on its basement. Perhaps the dramatic historical events of the last years of Hatra prevent its placement on the basement?
The study of the finds discovered inside Small Shrine 2, or in its proximity, by cross-checking both published and unpublished information from different archaeological expeditions, allows 'fresh' interpretations that are based on the existing photographs, notes and archive documents, without the need of new field research. This method is economically sustainable and is indeed relevant because it could be further improved and rapidly extended to other religious contexts of the city. In fact, the Iraqi State Board of Antiquities and Heritage archive contains a considerable amount of unpublished information concerning the Small Shrines excavated in the Fifties, which, alongside data preserved in other archives of archaeological expeditions, would represent a fundamental source for the future to better understand such religious buildings, where kings, aristocrats and priests made their offerings to the deities of the local pantheon.
Je dus donc interrompre les travaux et suivre au retour. L'inscription araméenne restait sur le chantier, faute de ressources pour la transporter i parmi les autres objets. Je choisi le torse que je rapportai en France et que je tiens à la disposition du Musé du Louvre.