There is one other bit of history that intertwines with a traditional name. At the end of the 7th Century, Beja tribesmen, pagan nomads from northeastern Sudan and southern Egypt who began to accept Islam at about the same time, swept across the Eritrean plateau and forced the Aksumites to flee southward. Very likely it was they who destroyed the city we know as Matara. Interestingly, one name which the villagers in the region give to the ruins is Belo Kelo (sometimes transliterated as Belew Kelew). Belo or Belu is the name of one of the Beja clans, recalling their presence in the area.

If the Aksumite placenames are uncertain, when we move to the pre-Aksumite period the uncertainty is compounded by a paucity of evidence. A fair number of inscriptions in epigraphic South Arabian have been found, but most are fragmentary or very short. One place name which appears regularly is Da'amat(), which appears to be the name of the pre-Aksumite kingdom which extended from Yeha and Aksum in the west at least to KaskasË and Matara in the east, with Yeha apparently being the capital (the sketch map gives a view of the extent of Da'amat and some of the places associated with this pre-Aksumite kingdom). Little is known about the relationship between Da'amat and the Sabaean kingdoms in Yemen, but it is clear that there was a strong association, cultural if not political. The architecture of Yeha, for example, is so similar to that of Marib () in Yemen that one can envision the same architects and stone masons working on both sites, except that the proportions of the Yeha structures are even more massive than at Marib. The same language (or very closely related languages) appear to have been spoken in both places, if the inscriptions are any indication, and the religion and material cultures also show strong similarities. Yet there are very distinctive elements in the pre-Aksumite remains that set them apart from those on the other side of the Red Sea. Sorting out indiginous from imported cultural elements, and for that matter determining whether more aspects of culture were traded to the west or to the east, will require much more research in both areas.

Excavation and Interpretation

The excavation of Matara was conducted for 10 seasons during the 1960's through 1970 by Monsieur Francis Anfray under the auspices of the Ethiopian Institute of Archaeology. I was fortunate enough to have been able to participate in the final two seasons at the site.

As mentioned earlier, the archaeological site known as Matara actually contains remnants of at least two major levels of habitation, each with at least two subdivisions. The roots of the Aksumite city which comprises the uppermost levels, basically Christian and pre-Christian, cover another city of the pre-Aksumite Sabaean kingdom of Da'amat. The Sabaeans and others from South Arabia (present-day Yemen) built or shared in the building of this and numerous other cities on these highlands, contributing elements of their architecture, religion, culture, and language to the indigenous civilization, which half a millennium later gave rise to the Aksumite civilization. To date it is unknown whether the pre-Aksumite city at Matara died naturally and later was replaced by the Aksumite community, or if it was conquered and destroyed by invaders from Aksum during its expansion and consolidation of the northern plateau.

Below the Sabaean level is another which developed in the region over a period of several hundred years before the Sabaean influence appears. Distinguished by use of intensive agriculture, settlement in towns and cities, and a unique culture including its own distinctive pottery, this culture appears to have led to the pre-Aksumite civilization.

Excavation and Interpretation:
     Dating

Numerous stones inscribed with Sabaean script, reused as building material in the Aksumite buildings but clearly taken from the destroyed earlier structures, help to identify the people who built this older city of which so little remains visible at the surface (but which can be compared with remains from other pre-Aksumite sites, such as Yeha. While it is not possible to date with precision the buildings themselves, these inscriptions can be dated with some accuracy by the style of the letters, thus giving an approximate age (roughly 5th Century B.C.E.) to the lower foundations.

Other remains firm up this dating, as well as giving much additional information about the ancient city and its people, the most common and useful being a vast amount of potsherds and some well-preserved pottery vessels of many types which were found throughout the excavations. By the immense quantity of pottery found, one can deduce that the city stood for quite a long time, or was inhabited by a very large population, or perhaps both (as appears to be the case). Also, the kinds of pottery manufactured in Matara during the Aksumite and pre-Aksumite periods are specific to those periods, with very small amount of holdover of particular designs or techniques, suggesting rather significant cultural change (perhaps the arrival of a new population, or an extensive unoccupied hiatus) between the two periods. It also provides a useful indicator to help date particular constructions, for the pottery intimately associated with a certain structure can generally be assumed to be roughly contemporary with that structure. Pottery styles generally last for a relatively short number of years before being modified or replaced, and while pottery itself is extremely durable, pots tend to break after a few years' use at most, so potsherds provide perhaps the most valuable tool for dating.

Excavation and Interpretation:
     Sublevels

In addition to the two major cultural stages, there are within each level variations of a lesser extent, but sufficient to allow a division of each major period into two phases, and to make the deposits of these phases clearly recognizable from the others. Thus, in the Aksumite structures, exists one pottery phase which is datable to the pre-Christian period, centering about the 3rd Century, and another phase which is clearly later, with heavy use of the Cross for decoration, along with numerous inscriptions, mostly of Christian subjects, for which dating centers around the 6th Century, although earlier and later material is included.

Coins of various Aksumite rulers were found in the upper level, ranging roughly from the 4th to the 8th Century, but they could not be used for precise dating of any of the structures since they seem to be mixed homogeneously throughout the upper Aksumite phase. This may indicate that coins of the earlier monarchs remained in circulation along with those of the late emperors. Some jewelry, gold statuary, and other pieces of fine art were also found, perhaps buried for safety's sake before the city's destruction, and not rediscovered until the Anfray excavations.

In addition to the Aksumite wares, foreign goods in appreciable quantities were found among the ruins of the upper level. The most numerous imports were pottery, especially amphorae (tall conical terra cotta wine or oil jars) of which more than a score were found intact or substantially so, and innumerable sherds were found mixed with the local wares. As well as their primary functions as containers, these amphorae were put to some novel uses by the Aksumites. Often their bottoms were knocked off and, stacked one inside the other, they were used as clay drain pipes in the larger palaces. Occasionally they also served as coffins for small infants. Some imported Roman bowls and other pottery is found, along with fragments of Roman or Egyptian glass bottles and Egyptian stone vessels, a small Egyptian figurine (perhaps from Meroitic Kush), some bronze chandeliers and other items of church use from Syria, and a large cache of golden objects, some of local manufacture, but including 15 Roman gold coins. These imports underscore the heavy trade which is known to have taken place between the Aksumite Empire and the Mediterranean world.

Excavation and Interpretation:
     Aksumite City

A great deal of information could be deduced about the Aksumite city. The visitor, on first view of the site, was struck by the awesome dimensions of the palaces, majestic even in ruin. In typical Aksumite fashion, they generally are composed of a square central building, surrounded by courtyards and an encircling wall of secondary structures. Today, only the basement walls stand, but the buildings originally had two or in some cases even three stories above this basement level.

The three churches or church-like buildings are rectangular rather than square, but also are surrounded by secondary structures. One of them has a magnificent stone-lined tomb beneath it similar to those in the so-called "Tomb of King Kaleb" in Aksum. The other, which has a keyhole-shaped, walkthrough baptistry nearly identical to those found in other Aksumite churches, was dated to the 7th Century, and its floor plan is the basilica desgn used in the early Christian churches of Syria. Note that according to legend it was missionaries from Syria who established the Christian church in Aksum. Immediately below this church, of identical plan but somewhat smaller dimensions, are remains of a church of about the 5th Century, and below that are foundation remnants of what appears to have been a pagan temple of indeterminate age, although from what can be seen, it might bear some similarity to the foundation of the huge Grat Ba'al Gibri temple at Yeha.

In the residential quarter of the city was a confusing jumble of homes as well as commercial quarters of every age, jammed close to one another, poorly constructed and evidently of short duration. But they revealed much about the living conditions of the ordinary citizens of an Aksumite city, something which up to then was completely unknown.

Excavation and Interpretation:
     Pre-Aksumite City and Below

Of the pre-Aksumite level, not a great deal can be said. Little enough remains of it, and very little was exposed to view, since in all cases so far discovered Aksumite remains overlie the earlier foundations. Rather than destroy these upper remains to reach those below, excavation was carried out in the open areas between Aksumite walls. The pottery associated with the pre-Aksumite stone foundations is much like that of the Intermediate period excavated in Yeha, a major pre-Aksumite site in northern Tigray province in Ethiopia and capital of the Sabaean or Sabaean-influenced kingdom of Da'amat (some inscriptions identify kings who may have ruled kingdoms on both sides of the Red Sea), helping assign to this level a date between the 5th Century B.C.E. or earlier and the beginning of the Common Era.

Below the Sabaean pre-Aksumite level in turn lie remnants of a yet older habitation, thought to have been indigenous. While no construction was found during very limited excavation of this level, pottery of a somewhat different type appeared in great quantities, and the Yeha type was also found in the upper reaches of the phase. It was impossible to date this earliest phase except to say that it clearly is earlier than the 5th Century B.C.E. Current knowledge concerning Sabaean pottery in South Arabia at the time of these excavations was insufficient to allow comparisons to determine whether this ware is of Sabaean or indigenous design.

An excellent monograph on the latest information regarding the development of urban society in Eritrea and Ethiopia by Rodolfo Fattovich, an Italian archaeologist I worked with at Yeha in 1971 and 1972 and a leading authority on Aksumite and pre-Aksumite archaeology, can be downloaded here in PDF format. Acrobat Reader is required to view this file; click to get Acrobat Reader

Invaluable Legacy

In 10 years of work, one-third to one-half of the visible structures were excavated. The state of our knowledge of the city and the civilization remains fragmentary. This excavation, the first of its magnitude in any Aksumite or pre-Aksumite site, raised as many questions as it provided answers. But at least it provided a framework of comparison. As further sites are studied in detail, they can be viewed in reference to the knowledge gained from these pioneering studies. Thus the site of Matara is of inestimable value to the record of the development of the civilization of Eritrea and indeed that of Ethiopia. Any damage to this site or its contents thus robs the region, in fact the whole world, of a priceless and irreplaceable legacy.

All photos, graphics and text copyright © 1969-2005, Skip Dahlgren, unless otherwise credited.