Modeling the demands for wood by the inhabitants of Masada and for the Roman siege

https://doi.org/10.1016/j.jaridenv.2010.01.010Get rights and content

Abstract

Modeling the demands for wood, especially firewood, for the inhabitants of the unique desert fortress of Masada during the major period of its occupation (beginning about 150 B.C.E. and ending with its fall after the Roman siege in 73 C.E.) is based on the well-documented history of the site, of the number of inhabitants in each phase of occupation, and the current demand for firewood in traditional societies. The previously analyzed ancient botanical remains from Masada provide base-line data of the types of wood used. We have concluded that when the Roman siege began in C.E. 73, the vicinity of Masada would have been denuded of trees and shrubs as a result of ca. 225 years of occupation. Therefore, the Tamarix wood used to construct the upper parts of the Roman siege rampart was probably not local. The isotopic composition of the Tamarix beams probably indicates that they were imported from a different region, such as the more humid and cooler river banks east of the Dead Sea, rather than the result of climate change as previously proposed.

Introduction

Masada, a desert fortress with an impressive 2000 years old palace, was recently selected by UNESCO as a World Heritage Site. This unique, well-preserved ancient site, built mainly by Herod the Great, is located on the summit of a sheer cliff in the Judean Desert west of the Dead Sea (Fig. 1). Its natural location makes it an exceptionally well-defended site. After the destruction of Jerusalem by the Roman army in 70 C.E., the Roman 10th Legion laid siege to Masada in 72 C.E. It became the last fortress in Judea to be conquered by the Roman army. The 1st century C.E. historian Flavius Josephus (Flavius Josephus, 1960) recorded the siege and the most famous episode in the history of Masada – the mass suicide of its Jewish defenders following the breaching of the site walls, probably in the year 73 C.E.

A stone casemate wall about 1500 m long surrounds most of Masada's summit, and there are several groups of buildings on the plateau, the larger concentration of buildings including a palace located in the northern part. Masada was well supplied with water from a 40,000 m3-cistern system that was built by Herod the Great. The remnants of structures at the base of the cliff, built by the Romans for the siege, are probably the best preserved of their kind in the Roman Empire. There are stone foundations and walls of eight Roman camps, a 4500 m long surrounding wall and an impressive siege rampart that approaches the casemate wall from the west. Four of the Roman army camps were built in the lowland east of Masada and four on the higher, rocky, western and southern part. The major Roman military action was conducted from the west, and included the building of the siege rampart in the final attack (Flavius Josephus, 1960, Yadin, 1966).

Masada provides a special opportunity to study how human activity at an ancient site could have influenced desert vegetation since we have historical documentation (Flavius Josephus, 1960, 7: 8–9), extensive excavations at the site (Yadin, 1966), as well as detailed botanical (Liphschitz et al., 1981, Liphschitz and Lev-Yadun, 1989) and archaeological publications (e.g., Cotton and Geiger, 1989, Meshorer, 1989, Yadin and Naveh, 1989, Netzer, 1991a, Aviram et al., 1994, Aviram et al., 1999, Aviram et al., 2007, Foerster, 1995, Bar-Nathan, 2006). Moreover, the archaeological and environmental situation has not been seriously impacted by post-Roman period activities at the site, such as brief sojourns by Bedouins (Yadin, 1966).

The first model discussing the quantitative aspects of the use of wood by the Roman army during the siege of Masada and its possible environmental implications was proposed by Waisel (1986: 259–260). He, as well as other authors who have studied the palaeoecology of Masada (Liphschitz et al., 1981, Liphschitz and Lev-Yadun, 1989, Yakir et al., 1994) all assumed that when the Roman siege started the environs of Masada had not been damaged by previous human activity. Indeed, based on the isotopic signature of wood from the Roman siege rampart, Yakir et al. (1994) proposed that at this time the region experienced a more humid climate, based on the assumption that the wood was local. In contrast, on historical grounds, Gichon (2000) proposed that no local wood was left when the Roman siege of Masada begun. Thus, if the wood was not local, its isotopic signature does not reflect climate change, but rather a different, more humid origin. Since palaeoclimatic reconstructions for the late Holocene period in the Near East, based on palynology (e.g., Neumann et al., 2007), Dead Sea water levels (e.g., Migowski et al., 2006), oxygen isotope records from speleothems (e.g., Bar-Matthews et al., 1997), geological erosion and sedimentation (e.g., Migowski et al., 2006) have offered contradictory and unequivocal results (e.g., Staubwasser and Weiss, 2006, Faust and Ashkenazy, 2007), in order to evaluate the origin of wood at Masada we have adopted an alternative approach based on human and plant ecology.

In this paper we examine the issue of wood supply for human use in Masada during its major period of occupation (non-continuous and spread over some 225 years) and for the short but intensive period of activity during the Roman siege. To undertake this, we have developed a model of wood exploitation with reference to historical accounts, modern ethnography on firewood use, and biogeography of trees and shrubs in the region. Since we do not wish to over- or under-estimate the extent of environmental damage in the past, our study uses conservative figures (i.e., the lower range) to estimate the per-capita wood consumption and the number of inhabitants. This conservative approach was adopted primarily, because an overestimation of the potential damage to the local woody vegetation by the site's occupants could result in a faulty conclusion, namely that the wood used by the Roman army during the siege was brought from another place when the source was in fact local.

Section snippets

The geographical setting

Masada is located on top of a boat-shaped plateau, ca. 650 m long and about 220 m wide at its widest part. At an elevation of about 50 m above mean average sea level, the plateau stretches some 450 m above the surface of the Dead Sea, about 2.5 km west of its present shore. The current shore is located somewhat east of the shore which prevailed 50 years ago because of the over 10 m drop in the water level of the Dead Sea during the last 50 years. The elevated plateau of Masada is surrounded by steep

Current woody vegetation in the region of Masada

The present-day vegetation near Masada is dominated by the Saharo-Arabian element with a Sudanian contribution in humid habitats such as En-Gedi (Zohary, 1962). Our botanical survey in the region conducted along the roads in the upland and lowland, and in off-road sectors of the territory as part of the data collection for this essay, indicated that today, after half a century of nature conservation, there are only several hundred Acacia trees growing along the wadis within a distance of 1-h

Chronology and estimated population size of Masada

In order to calculate the amount of wood used by the people of Masada and the Roman military, we have to estimate the number of people in each occupational phase at the site (Table 1). As usual with ancient periods, we have very little actual data. Given the general accuracy of his descriptions, we accept the numbers provided by Flavius Josephus (1960) for the period of the Roman siege as sufficiently accurate.

The chronological frame relevant to this study (Table 1) starts with the founding of

Demands for wood and its possible origin in each phase of occupation

In the first phase of occupation at Masada (150 B.C.E. to 40 B.C.E. = 110 years), and according to Waisel's proposed tree density of 90 per km2 the annual firewood consumption for 100 people using 300 kg each, is 30 ton (60 trees) (assuming that they occupied Masada only in the cooler months, or no more than six months each year). If shrubs provided 25% of this fuel, only 22.5 of the total 30 ton of this firewood originated from 45 trees. This number of trees would have occurred in an area of ca. 0.5 km

Acknowledgements

We thank Nancy R. Bain, Susan Lev-Yadun and Liora Kolska Horwitz for their comments.

References (60)

  • S. Ashkenazi

    Acacia Trees in the Negev and the Arava, Israel. A Review Following Reported Large-Scale Mortality

    (1995)
  • Atlas of Israel

    Survey of Israel

    (1970)
  • S. Avitsur

    Man and His Work. Historical Atlas of Tools & Workshops in the Holy Land

    (1976)
  • C. Bailey et al.

    Bedouin plant utilization in Sinai and the Negev

    Economic Botany

    (1981)
  • R. Bar-Nathan

    The pottery of Masada

  • U. Baruch

    The late Holocene vegetational history of Lake Kinneret (Sea of Galilee), Israel

    Paléorient

    (1986)
  • U. Baruch

    Palynological evidence of human impact on the vegetation as recorded in Late Holocene lake sediments in Israel

  • H.M. Cotton et al.

    The Latin and Greek Documents

  • A. Fahn et al.

    Wood Anatomy and Identification of Trees and Shrubs from Israel and Adjacent Regions

    (1986)
  • Catchment Analysis. Essays on Prehistoric Resource Space

    Anthropology UCLA

    (1980)
  • Flavius Josephus

    The Great Roman–Jewish War: AD 66–70. The William Whiston Translation

    (1960)
  • G. Foerster

    Art and architecture

  • M. Gichon

    The siege of Masada. Les Légiones de Rome sous le Haut-Empire

    Collection du Centre d'Études Romaines et Gallo-Romaines Nouvelle série

    (2000)
  • D. Gill

    A natural spur at Masada

    Nature

    (1993)
  • R. Gophna et al.

    Man's impact on the natural vegetation in the Central Coastal Plain of Israel during the Chalcolithic and the Bronze Ages

    Tel Aviv

    (1986/7)
  • W.S. Hanson

    The organisation of Roman military timber-supply

    Britannia

    (1978)
  • C. Hawkes

    The Roman siege of Masada

    Antiquity

    (1929)
  • Cited by (0)

    View full text