Architecture and Stratigraphy

H3 went through several stages of development during its period of use. The earliest occupation seems to have consisted of simple pits, often filled with blackened, burnt material. It may therefore have started life as a camp-site, the pits being used to cook or smoke food. In places, the pits became covered by an extensive scatter of edible shell, mostly turban shell (Lunella coronata). This species is still found nearby, though the immediate environment around H3 is now too saline and muddy to support it.

Eventually, people began to build stone structures out of the local beach-rock and red sandstone. Dry-stone walling survives in some cases to over half a metre in height. At least two separate multi-chambered stone structures have now been uncovered in the main excavation trench. On the western side there is a relatively simple structure, while a larger and more complicated cellular building  is found to the east.

The purpose of the chambers is sometimes evident. Some were clearly multi-functional. Chamber 1 seems to have been used as a living or eating room, judging from the food remains found inside, but was also used to store bitumen. We do not know if the medium-sized chambers were roofed, but if so, it would have been with perishable materials such as reeds or grass, or perhaps cloth, like a tent. Some of the smaller cells were probably used for storage, and were found to contain jars. 

Chamber 15 was used for flint-working and bead-making. It may have been an open area, unroofed but providing shelter from the wind.

The use of many of the chambers remains mysterious. Chamber 11 was carefully paved, presumably in order to maintain a clean, hard surface for some unknown process.

Towards the end of the use of the site some of the chambers were subdivided, and other small chambers were attached to the outside of existing, partially buried structures. The reason for this architectural elaboration remains mysterious, though careful analysis of the finds may eventually reveal if a change in the use of the site or the structures occurred.

Finally, the structures were abandoned, and no further stone structures were built, at least in the excavated portion of the site. The sand that then covered the buildings is extremely rich in lithics, and Ubaid pottery is also found. Some pits were found, dug into the sand which had buried one of the latest structures on the site (the paved chamber). It appears therefore that people continued to visit and use the site after some or all of the stone buildings had been abandoned and left to fill with sand. They made plentiful stone tools, but no longer built stone structures.

Although the scale and complexity of the stone structures at H3 is unprecedented in the region at this time, some comparisons can be made with other sites. The early fire-pit horizon is comparable to an Ubaid-related site in Qatar named al-Da'asa, where over 50 pits were excavated, many with signs of burning (de Cardi 1978). H3's architecture has parallels with a site called Shagra, also in Qatar, where a single small cellular stone building was found (Inizan 1988). The stone tools of this site belonged to the Arabian Neolithic tradition, like those of H3, but there was no pottery. It was interpreted as a fisherman's hut. Recently, similar architecture has been reported at Arabian Neolithic sites on the island of Marawah (MR11), and inland at Kharimat Khor al-Manahil, both in the UAE. At least four sites in Saudi Arabia are comparable to H3 in scale and depth of deposits. These are Abu Khamis, Dosariyah, Ain Qannas and Khursaniyah. The first three of these have been excavated, and although the material culture of these sites is closely comparable to that of H3, architectural similarities are not evident.