A pile, also known as a Roman pile, Gallo-Roman pile, or funerary pile, is a specific type of funerary monument in the archaeological vocabulary of France: elevated towers, typically square or rectangular in plan, with circular forms being less common. Their primary function was to serve as funerary structures within Roman Gaul.
Constructed between the 1st and 4th centuries AD, piles are found across a wide area of south-western Gaul, with a high concentration in the civitas of , around the present-day city of Auch, where they often bear a local name, such as tourasse or tourraque, derived from the Occitan word for "tower".
In the absence of in-depth site excavations, their function has long remained an enigma, but it is highly probable that they were funerary monuments dedicated to the memory of a local notable. Although they are not true mausoleums, since almost none have a burial chamber, they indicate the proximity of the burial itself, very often in an enclosure at the foot of the pile. Their ostentation also serves to demonstrate the wealth and power of the family of their patron, whose villa is frequently nearby.
Fourteen piles are protected as French historic monuments.
Monuments with a similar function also exist in other parts of the Roman Empire: Italy, where they may have originated, Hispania and Proconsular Africa. However, their morphology is often slightly different, and they are not called pile in the languages of the countries concerned.
Terminology The term pile, has come to be used in everyday French for any kind of massive construction resembling a pillar made up of superimposed structures, as noted by Camille Jullian in 1896. In Guyenne and Languedoc, the word tourrasse and, in Gascony, tourraque, are augmentations of the Occitan noun tour. The name peyrelongue (long stone), usually given to megaliths, is found in several places, with the forms pirelongue or pirelonge (Pirelonge tower in Saint-Romain-de-Benet, near Royan). In Saintonge, the noun fanal, which at first glance might suggest a signal tower, is in fact an evolution of the term fanum (temple). The toponym "Montjoie", often used to designate the site of a military memorial erected at a crossroads, is sometimes associated with a pile (the Gallo-Roman pile at Montjoie in Roquebrune, Gers). The pile d'Authon-Ébéon is sometimes called a "pyramid", a noun not systematically reserved for monuments with a geometrically pyramidal shape. In 1814, Alexandre Du Mège described the Lestelle pile, in Beauchalot, as an obelisk by analogy of form, without linking it to Egyptian culture, as he believed it was dedicated to Mercury.
Outside France, the more generic term mausoleum or mausoleum-tower is most often used, either because the monument's funerary character has remained evident through its good preservation, or because other functions have been attributed to it: milestone, commemorative monument, etc., which would have protected it from vandalism. An ancient mausolée, in the strict sense of the French term, is a "grandiose funerary monument [...] destined to receive the remains of a powerful person [...]"; this is not true for piles, the vast majority of which are not designed to serve as burial sites, and are more akin to a cenotaph.
Historiography and characteristics In 1898, Philippe Lauzun published his first inventory of piles in south-western France; he included all tower-shaped monuments, whatever their plan, whether or not they had niches or a burial chamber, and did not always conclude on their function, which was often unknown at the time.
Eight decades later, Pierre Audin undertook a study of the Cinq-Mars pile, but broadened the scope of his publication to include all turriform funerary monuments, in France and abroad, also considered "piles".
In the introduction to the book she published in 2016, devoted to the study of south-western piles, Pascale Clauss-Balty gives a much more restrictive definition of the pile, already outlined in 1993 by Pierre Sillières and Georges Soukiassian and based on the type most frequently encountered in south-western France. A pile is a massive, tower-shaped monument, square or rectangular in plan, with a niche in its upper part, originally housing a statue; this definition excludes monuments such as the Cinq-Mars pile or the Tour de Pirelonge (no niche), the Tourasse d'Aiguillon (circular plan) or the Tour de Mézolieux (hollow monument with a funerary chamber).
Generalities and layout The typical pile plan is therefore that of a solid tower – excavations carried out during clandestine digs or by "treasure seekers" in the 19th century never revealed the slightest interior cavity – quadrangular in shape, generally breaking down into several structures whose plan dimensions become smaller as they rise. In practice, pile plans sometimes deviate from this pattern, but the differences seem to be regional; a certain homogeneity in construction is discernible within the same geographical area. La Tourasse in Aiguillon (Lot-et-Garonne) is the only circular building still standing, but its status as a funerary monument, long debated, now seems attested.
The height of a pile is difficult to establish due to the deterioration of surviving monuments; but it varies widely, usually between 8 and 15 metres. However, that of Cinq-Mars (Indre-et-Loire), the tallest, reaches 29.40 metres and that of Clergué (Haute-Garonne), probably the smallest, does not exceed 6 to 7 metres.
Description
Constituent parts
The foundation, designed to anchor the monument in the ground, always protrudes well beyond the elevation and its depth, depending on the quality of the subsoil and the height of the pier, varies between 0.6 and 1.5 metres, except at Villelongue-d'Aude, where the pile is set directly on bedrock without the need for foundations. Generally devoid of facing, the foundations are made of cemented blocks.
The primary function of a plinth is to elevate the main structure, often referred to as the pier or upper storeys, enhancing its visibility. Typically, the plinth's facing material matches the upper storeys, but its design does not dictate the overall layout. Additionally, plinths are generally less decorated compared to the upper portions of the structure.
The podium is the first ostentatious floor of the building, and its importance is reflected in the quality of its cladding and the decorations (pilasters, cornices) it is sometimes provided with, at least for the largest piles. It is also on the main face of the podium that the inscription plaque bearing the dedication of pile is affixed.
The aedicula is the most important element of the stack. This is where, if present, the niche is installed to house a representation of the deceased whose memory is to be honored. The construction of the aedicula incorporates the same decorative elements as the podium. The niche is often barrel-vaulted, with a flat or rounded base, and its width is generally greater than its depth: it is not necessary to have a deep niche, since the statue it contains is to be installed at the entrance, in full view.
The question of the roof remains unresolved, as none of the piles have been preserved intact up to this level. Depending on the layout of the aedicula, any shape can be envisaged to allow water to run off properly: gable, hip or even pyramid, cone, with a mortar-sealed tile roof – the use of a wooden framework is difficult to envisage for reasons of longevity – or stone slabs held together by metal studs– this is the option chosen for the Pirelonge tower.
Masonry and decoration
The masonry consists of a core of opus caementicium (uncalibrated stone embedded in mortar) covered by a facing of opus vittatum (small, regularly cut limestone rubble), sometimes in large blocks (Saintonge piles) or bricks (Cinq-Mars piles). For the Roques pier, the use of Garonne pebbles for the foundations and core and brick for the facing is an adaptation to local building materials and techniques, also used in the nearby Gallo-Roman rampart of Toulouse, probably built at the same time.
For rubble stone facings, pilasters are generally carved into the facing stones, but stucco elements, which have disappeared over time, may also be used to complete the decoration. Brightly colored plaster was sometimes used to decorate the aedicula and its niche: vestiges of this can still be seen, for example, on the Roman pile at Luzenac. At Tourreilles, the pile is decorated with a play of different-colored rubble stones. The brick, limestone, and terracotta panels on the Cinq-Mars pile have no known equivalent, although a geometric motif in baked clay is thought to have been used on the lost pile at Chagnon.
Environment and layout
Often the pile is accompanied by a square or rectangular funerary enclosure at its foot; the pile may be in the center of the enclosure, against one of its walls or even outside. This layout is undoubtedly more common than research shows: excavations are sometimes incomplete or old and poorly documented, and these enclosures go all the more unnoticed as some must have been made of perishable materials (palisades or hedges). At Mirande (piles de Betbèze), numerous burials were found in and around the enclosure, confirming the pile's role as a funerary monument. On this site, and no doubt on others, the pile is not isolated, but part of a larger necropolis whose boundaries are not defined.
In several cases, as at Ordan-Larroque, the discovery of a villa near the pile leads us to infer that it was the family of the villa's deceased owner who decided to erect the monument. Here again, insufficient data may lead us to underestimate the importance of the link between pile and habitat.
The presence of an ancient road, along which the pile is built, is reported at more than fifteen sites, and in these cases, the pile niche faces the road. This can be explained by the desire to show the pile to as many people as possible, including travellers passing along the thoroughfare. The latter may be a major thoroughfare (the route from Tolosa/Toulouse to Aquae Tarbellicae/Dax for piles in Haute-Garonne, or the route from Limonum/Poitiers to Burdigala/Bordeaux for piles in Saintonge) or a less-frequented road (the Clergué pile or the Chèvre d'Or tower).
Beyond these considerations, the emphasis on maximizing visibility makes it hard to establish a uniform design for piles. Their location, whether on a hill or in a valley, and their orientation relative to cardinal points seem less important than their overall prominence in the landscape. Similarly, proximity to the marked graves or associated dwellings appears less crucial than visibility.
Function Piles have long been described as "mysterious", as knowledge of their function has been lost because their exterior decoration has often disappeared. The first excavations, carried out in the 19th century, were often incomplete and conducted using techniques that could be improved. In 1896, in the introduction to his report on excavations at the Chagnon site (Charente-Maritime), Camille Jullian raised the "pile question", reviewing the various hypotheses on the function of these monuments.
Bollards One hypothesis suggests piles functioned as milestones due to their presence along some Roman roads. However, even considering potential lost structures, their overall number seems insufficient for a systematic milestone system. Additionally, their large size appears excessive for a simple marker. Furthermore, some piles are located away from major roads. Undoubtedly, piles likely served a secondary function as visual landmarks within the landscape, akin to other prominent structures or trees.