AGRICULTURAL LAND-MANAGEMENT METHODS AND IMPLEMENTS IN ANCIENT EREẒ ISRAEL

Ereẓ Israel is a small country with a topographically fragmented territory, each geographical region having a distinctive character of its own. These regions include: the coastal plain, the lowlands, the hilly country, the inland valleys, the north-south rift valley, and the arid and desert areas. The whole of the country, excluding the desert regions, has an area of a little over 10,000 sq. miles (26,000 sq. km.) and it has been estimated that of this only 2,500 sq. miles (6,544 sq. km.) are capable of cultivation. The country's semi-tropical climate, with a hot and dry season (between April and October) and a cold season with an unpredictable rainfall (between November and March), its varying altitudes, terrain, and soils, all imposed limitations on the type of land management which could be undertaken in the different parts of the country. There were very few springs of water, so dry farming was practiced in most parts. Agriculture in semi-arid regions, for instance the Negev, depended entirely on run-off irrigation. As a result, certain regions were suited for the cultivation of grain crops, others for fruits and vegetables, and some for animal husbandry. The writer of Deuteronomy was under no illusion regarding the restricted agricultural potential of the land. Speaking to the People of Israel before the entrance to Canaan, Moses said: "For the land, whither thou goest in to possess it, is not as the Land of Egypt, from whence ye came out, where thou sowest thy seed, and wateredst it with thy foot, as a garden of herbs. But the land, whither ye go to possess it, is a land of hills and valleys, and drinketh of the rain of heaven" (Deut. 11:10–11).

While the geographical fragmentation of the country would seem to encourage political and cultural regionalism, the diversity of the economic pursuits within the country resulted inevitably in heightened commercial interaction between people from one part of the country and the other, rather than isolationism. During the Bronze and Iron ages, more than 50 percent of the population in Ereẓ Israel were agriculturalists living in the countryside. The rest lived in towns or small cities and dealt primarily with administrative and commercial and industrial activities. Almost all farmers lived in small villages or hamlets, since the isolated farmstead was not known until the eighth century b.c.e. In the vicinity of these settlements they established fields and grew their crops and orchards. Numerous agricultural land-management techniques and implements were used over the millennia. By the Byzantine period (sixth to seventh centuries c.e.) crop yields appear to have reached levels that have only been approached in modern times.

Topographical

In ancient times the rural countryside was divided into vineyards and olive groves, arable fields, orchard plantations, vegetable gardens, and areas of public land given over to pasture and industrial activities (e.g., lime and charcoal burning, and stone quarrying). Topographically, the vineyards and olive groves were more suited to sloping ground or to the highlands. Vines are unable to grow with ease above 2750 ft. (900 m.) above sea level. Olives are also said to be difficult to grow above 2450 ft. (800 m.) or below 1225 ft. (400 m.). However, olive presses have been found on Mount Hermon at sites with elevations up to 3,300 ft. (1000 m), as at Kafr Dura. Olive trees could even be grown in areas with an annual rainfall of only c. 200–300 mm., as in northern Africa. Grain was best adapted to the plains, the broad valleys, and some of the internal valleys. Vegetables and certain types of fruit trees were grown in areas with access to permanent sources of water, springs, wells, or cisterns.

The location of many fields reflect traditional answers to the problems of the natural environment. Hence, field boundary walls were established along the same lines as those built thousands of years earlier, simply because they were the most topographically convenient. North-facing slopes were particularly favored for the establishment of new fields and this is because they were less exposed to the sun and their soils were able to retain moisture for longer. The size and shape of fields was usually affected by the steepness of the ground, with smaller and narrower fields on extremely steep slopes. The appearance of the field could also be dictated by the rockiness of the terrain. In Samaria and in the Modi'in foothills, for example, vines and olive trees were cultivated in "boxlike" pockets of deep soil scattered in certain areas of rocky outcrops. Whether or not the field was used for arable purposes or for fruit trees would very much depend on the type of soil available, the rockiness of the ground, aspect, drainage, and so forth. The position of natural sources of water used for irrigation, whether a spring or well, would have an effect on the location, shape and function of nearby fields.

Land Management

Agricultural fields and their crops are mentioned in a number of ancient written sources, principally the Bible and the early Jewish writings from the Roman period. Information about fields may also be found in some of the Classical sources, but these are usually not directly relevant to Ereẓ Israel. A smaller amount of information may also be derived from ancient inscriptions and various other epigraphic materials.

The Bible contains a wealth of information about agricultural practices and the landscapes of the country during the Iron Age ii period, although some of the passages may contain strands of information relating to earlier periods. Each town and village had its own surrounding territory of fields and common land. Most of the agricultural land was in private ownership and the family inheritance was referred to as the naḥalah or aḥuzah. Royal estates also existed and oẓarot ("storehouses") were built in the fields (i Chron. 27:25). Landmarks were set up between the various plots of land. The general term for cultivated land was sadeh (Lev. 27:16). The word was used to designate cultivated pieces of land next to the towns as well as open areas used for pasturage. The area of land which could be plowed with a pair or team of oxen during the course of a single day was referred to as the ẓemed sadeh (i Kings 19:19). A ma'anah was half of that area (i Sam. 14:4). The kerem referred to vineyards and olive groves (kerem zayit, Judg. 15:5). Mixed fruit trees were grown in the gan or ginnah, usually next to the houses (Song 5:1; 6:2; 14:12), or in the pardes (Song 14:13). Plantations of fruit trees (the mattah) were grown further afield and were sometimes irrigated (Ezek. 31:4).

Isaiah 5:1–8 has a description of terraced fields being prepared for a vineyard: the vegetation was uprooted ('zq), stones were cleared (sql) and then stone fences and an observation tower were built: "My well-beloved hath a vineyard in a very fruitful hill. And he fenced it, and gathered out the stones thereof, and planted it with the choicest vine, and built a tower in the midst of it, and also made a winepress therein: and he looked that it should bring forth grapes…" (Isa. 5:1–2). A subsequent verse (5:5) implies that the stone fence (gader) surrounding the terraced unit was surmounted by a thorny hedge. This was used to protect the grapes from animals and also from people walking along the mishol ha-keramim, the "path between the vineyards" (Num. 22:24). Some of the terraced slopes were not plowed but dug with hoes (Isa. 12:11). The heaps of stone visible in the terraced areas are referred to in Hosea 12:11 as the gallim ("heaps") "in the furrows of the fields" and in Micah 1:6 as the ai ("pile") in the field.

It has been suggested that one of the terms used in the Bible for terraces (especially for vineyards but also for orchards) is sadmot (ii Kings 23:4; Jer. 31:40; Deut. 32:32; Isa. 16:8). Hence, Habakkuk 3:17 should perhaps now be read as follows: "The fig tree does not blossom/ There is no produce on the vines/ The yield of the olive fails/ The shadmot ("terraces") do not produce food." The shdmt are also mentioned in two Late Bronze Age Ugaritic texts. The first, cta 23: 8–11, is rendered: "Let them fell him [the god Mot] on the terrace like a vine." The second, cta 2.1.43, is a damaged text and the context is not as clear. Another suggestion that has been made is that gbi (Jer. 39:10; 52:16) was an alternative word for "terrace" and that yogevim referred to the workers/owners of irrigated terraces. This seems unlikely and the reference is probably to the vats of winepresses located in vineyards (cf. King 1993, 159). Additional references to terraces in the Bible include the meromei sadeh mentioned in the premonarchic Song of Deborah (Judg. 5:18) and the sadeh teromot in David's Elegy for Saul and Jonathan (ii Sam. 1:21), both of which seem to refer to "built fields" on hillslopes. Terraces are more commonly referred to as madregot, and in Ezekiel 38:20 it is written: "…and the mountains shall be thrown down, and the madregot ('terraces') shall fall, and every wall shall fall to the ground." (cf. sing. madregah mentioned in Song of Songs 2:14).

The surfaces of fields were fertilized with manure, household rubbish, and ashes. Strict laws existed about fallowing land on the seventh year (Ex. 23:11). Roads are mentioned passing next to vineyards and fields of wheat (Deut. 19:14; Num. 22:30). Fields were sometimes expropriated and given by the ruler as presents to his supporters (i Sam. 8:14; 22:7). The process of buying a field was described in Jeremiah 32:44: "Men shall buy fields for money and subscribe evidences, and seal them, and take witnesses in the land of Benjamin, and in the places about Jerusalem…" Recent archaeological evidence indicates that during the Iron ii period, specifically from the eighth century b.c.e., there was an unprecedented expansion of agricultural territory with extensive terracing in the highlands. Terracing later spread into the Negev and Judean Deserts. Similarly, various biblical passages indicate a hunger for land during the course of the Divided Monarchy with the break-up of family inheritances and the creation of large estates owned by rich landlords. These landlords were cursed in Isaiah 5:8, where it is said that "woe unto them that join house to house, that lay field (sadeh) to field, till there be no place (left)…." Micah 2:2 spoke out against those who "…covet fields, and take them by violence…."

Fields were frequently mentioned in early Jewish sources from the Roman period. Fields were of different sizes, from the small plot known as the beit roba (approximately 336 sq. ft. 32 sq. m.) to the bet se'ah which had an area of 940 sq. yds. (784 sq. m.). The agricultural holding, the bet kor, was about 23.5 dunam in size, but most peasants probably had holdings of a much greater size than this. Boundaries for fields were sometimes indicated by a road, a path, a wadi – bed, an expanse of water, or even a water channel (Pe'ah 2:1–3). Much information is provided in these sources about the surface treatment of fields (plowing, manuring, and fallowing) and their yields. The depth of plowing achieved during this period was mentioned in Bava Kamma (2:5) as 3 tephaḥim (about 11 in., or 27 cm.) which is far greater than the maximum depth of 8 in. (20 cm.) known from recent traditional farming. This indicates that either a heavier form of plow was used or that the line of the furrows was plowed twice over to achieve the required depth. An area plowed during one day with a yoke of animals was known as bet hafarash. Fields with different types of soils were known by different names, for example sadeh madrin and sadeh kaskasin. Fields were even established on the summit of hills where the soil was so thin that "the oxen cannot pass over with the plow" and they had to be cultivated with mattocks (Pe'ah 2:2). The yield of a field was frequently estimated while the crop was still standing. The land of an orchard could belong to one person and the trees to another (see Pe'ah 3:4), a practice which still existed at the beginning of the 20^th century. A distinction was made between fields used for dry-farming (bet ba'al) and for irrigation (bet selaḥin).

There were two kinds of gardens in Ereẓ Israel during antiquity: the horticultural garden and the decorative garden. There is no evidence that a combination of the two ever existed. Vegetable gardens were usually located near a source of water, a spring or a cistern. Important archaeological evidence for irrigated garden plots has come to light during the investigation of Byzantine monasteries in the Judean Desert and elsewhere. Decorative gardens and groves of trees probably existed adjacent to temples and palaces from very early times, but direct archaeological evidence is still lacking. For the later periods, remains of decorative gardens have been discovered at the Herodian palace complex at Jericho. Plants and bushes were planted in ceramic pots with holes in their sides. Similar pots were found in the first-century gardens at Pompeii and in the second-century gardens at Hadrian's villa at Tivoli. Planting pots with holes in their sides are mentioned in Demai 5:10 and in connection with a vineyard in Kilayim 7:8. Cucumbers are said to have been grown in planting pots made of cattle-dung and unfired clay. The latter also had round holes in their sides, the size of which is debated in Ukẓin 2:9–10, "How great should be the hole?" the answer being, "such that a small root can come out."

Very few ancient epigraphic finds have been made concerning agricultural land management in Ereẓ Israel. Most of the inscriptions and documents which have been found deal with matters of economic administration and only very indirectly with work in agricultural fields. However, some evidence does exist and a number of examples will be given. The Gezer calendar is probably one of the earliest and most interesting of the documents preserved relating to agricultural activities. It lists information about the activities which needed to be undertaken at different points during the agricultural year. The text was clearly written by an inexperienced hand, perhaps by an apprentice scribe, and can be dated to the end of the 10^th century b.c.e. Information on fields exist in a number of other Iron Age inscriptions. For example, the Tel Siran inscription from Jordan, dated to c. 600 b.c.e. which mentions a vineyard (krm) and gardens (gnt).

A larger amount of material exists for the Roman and Byzantine periods. Interesting information on agricultural plots in the lower Dead Sea region has emerged from the Babatha archive, deposited in c. 132 c.e. in a cave in the Judean Desert. The largest of the plots owned by Babatha measured 20 bet seahs or 12.5 dunam. One of the Greek papyri (bb 21) deals with the sale of a date crop and indicates the names of three orchards in the area of Maoza from which the dates came: the Pherora orchard, the Nikarkos orchard, and the Molkhaios orchard. Another document (bb 16) is the land registration for four other groves of palm trees owned by Babatha at Maoza. Two of them are described as extending down towards the Dead Sea. For each grove an identifying name was given, the size of the grove, the taxes paid on it and the names of lands or features abutting the groves.

An important batch of non-literary papyri dating from the sixth and seventh centuries c.e. came to light during the Colt expedition to Auja Hafir (Nessana) in the central Negev. They include a vast amount of information on the agricultural lands belonging to the settlement, on sown land, vineyards, gardens, reservoirs, water channels, rights to water, and data on crops of wheat, barley, aracus, olives, and dates.

Terracing

Very little is known about the earliest forms of agricultural terracing in the highlands of Ereẓ Israel. It seems reasonable to assume that the technology of creating flat areas on hillsides by building walls and leveling fills was invented by various rural groups acting in cooperation at a local level, in different parts of the Levant and at different times. Hence, various centers of origin for terrace construction may have existed, with Ereẓ Israel being one of them. Incipient forms of terracing, such as soil held in place by logs of wood, by rows of wooden stakes, or piled rocks, would be very difficult to detect in the archaeological record. It was probably recognized early on that obstructions placed across a stream channel would eventually help towards stopping the movement of eroded soils and would induce a process of alluviation. Early slope terracing may have taken place initially in the lower parts of hills with newer terraces later being built further up the slopes. Another suggestion which has been made is that the natural steplike appearance of many of the slopes in the highlands, with thin layers of chalky marl interposed between limestone or dolomite strata, may have prompted man's first attempts at terracing. However, no evidence supports the assumption made by some investigators that the earliest terraces with stone walls must have been crudely executed, low in height, and built on relatively slight slopes. Indeed, the earliest terrace found at Sataf, which is of Early Bronze Age date (late fourth millennium b.c.e.), was relatively well constructed and was built on a very steep slope. It has also been suggested that the origins of terracing should be sought in the marginal semi-arid regions of the Near East. The suggestion is that the "channel-bottom, weir terrace" type was possibly the earliest form of terrace. However, terracing in the Negev and in the Judean Desert cannot be shown to be older than the Iron Age ii (seventh century b.c.e.), even though flood farming itself existed in the Negev from as early as the Chalcolithic or Early Bronze Age. Studies of New World terracing have also shown that the earliest terraced sites must have been in the less arid areas first.

Another important point which needs to be taken into account is that the idea of creating leveled areas on hillslopes for agricultural purposes is not dissimilar from the basic technology of architectural terracing or slope stabilization. At settlement sites in Ereẓ Israel, architectural terracing can be traced back to as early as the Natufian period. A system of four architectural terraces supporting 13 hut dwellings, are known from the Natufian site of Nahal Oren. These terraces were 80 ft. (24 m.) in length and 6–17 ft. (2–5 m.) in breadth, and their retaining walls were built of field stones. At many Early Bronze Age sites, architectural terracing supported houses and other structures on the slopes of hills. Examples of eb architectural terracing are known from Tel Yarmut, Tell el-'Umeiri, and sites in the Wadi el-Hasa. Architectural terracing continued to be used throughout the rest of the Bronze Age. At Jerusalem, a remarkable series of architectural terraces were unearthed by Kenyon on the east slopes of the City of David, probably dating from the very beginning of the Iron Age. Since the technology of architectural terracing in Ereẓ Israel can be traced back to late prehistoric times, it is possible to assume that terracing for agricultural purposes likewise had a similar antiquity in the hilly areas of the country.

Archaeological evidence indicates that terracing was introduced into the highlands of Ereẓ Israel at the beginning of the Early Bronze Age. It is not surprising that the earliest known use of terracing in the highlands should coincide with the introduction of plow agriculture in that area. However, terracing was clearly only practiced on a limited scale during the Early Bronze Age and as late as the eighth century b.c.e. No evidence supports the theory that the early Israelites (or Proto-Israelites) were responsible for inventing or introducing terracing into the highlands during the early Iron Age. They simply made use of an existing technology without any special adaptations or innovations. This refutes the stand taken by some scholars who have suggested, without any supporting evidence, that terracing prior to the early Iron Age was "unsystematic." Archaeological work has shown that the major expansion of terracing in the highlands took place at a number of times over a period spanning some 1600 years, from the Iron Age ii (eighth century b.c.e.) to the *Abbasid period, with cycles of contraction operating in the landscape at intervals of between 250 and 350 years. A decline in the use of the terraced areas appears to have set in around 750 c.e. and apart from some signs of renewed terracing activities during the *Mamluk and *Ottoman periods, especially in areas of irrigated terraces, this decline continued until the present century.

Demarcation of Agricultural Territory

The demarcation of agricultural territory was quite frequently referred to in biblical passages. The boundary, or territory, was referred to as gevulah. In Deuteronomy 19:14 the injunction is that "thou shalt not remove thy neighbour's gevul (landmark), which they of old time have set in thine inheritance…" (see also Prov. 22:28). The practice of tampering with landmarks, especially with those in the "fields of the fatherless" – abandoned fields – was apparently prevalent during the Iron Age, and if caught the perpetrator was cursed and punished (Deut. 27:17; Job 24:2; Prov. 23:10). The physical appearance of these "landmarks" is unknown but they were probably permanent and immovable. A boundary stone at the edge of a field of grain is depicted in a New Kingdom tomb painting from Egypt. These boundary markers were often swept away by the inundations of the Nile and had to be replaced by reference to documented cadastration. This problem did not exist in Ereẓ Israel, but markers could be moved during disputes between neighboring farmers. There is no evidence, however, to suggest that the exact locations of the markers were documented in any way. Josephus, writing in the late first centuryc.e., warned: "Let it not be permitted to displace boundary marks, whether of your own land or of the land of others with whom you are at peace; beware of uprooting as it were a stone by God's decree laid firm before eternity." (Ant. iv, 8:18).

Very little archaeological evidence exists for the demarcation of territory in Ereẓ Israel during ancient times. Hence, there are difficulties in defining the extent of cultivation and pastureland belonging to a settlement at any given period. But some evidence does exist. The lands of *Gezer, for example, were marked out during the late second or early first century b.c.e. with inscribed boundary stones. Some of these markers also refer to the adjoining estates of Alkios, Alexa, and Archelaus. Boundary stones are known to have been set up to delimit taxable properties between cities, towns, and villages in the Roman provinces. Inscribed boundary stones delimiting agricultural territories, dating from the time of Diocletian (late third century), have been found in the Hauran, the western Golan, and in the Huleh Valley.

Less information exists on the use of stones or markers to demarcate the ownership over lands of a specific farm or individual plots of land. The more permanent type probably took the form of prominent natural rocks, ancient trees, large piles of stones, posts, fences, or walls (see Or. 16:5). Boundary markers in the form of monoliths were found along roads bordering areas of fields in Samaria and Modi'in and dated to the Hellenistic and Roman periods. In Bava Batra (4:8) it is stated that "if a man sold a field he has sold also the [boundary] stones that are necessary to it."

Temporary markers made of piles of stones were the easiest to create and they can be seen all over areas of ancient and traditional field systems. In the highlands near Jerusalem such landmarks were known in Arabic as rassem or hejar et-tuhm. The latter term was also used for the pile of stones located at both ends of a cultivation strip in the plains. One commentator of the 19^th century noted that the lands of the Sharon were marked out with small lumps of stone. Mark Twain in his description of agriculture in the Lebanon (1869) wrote that he saw "rude piles of stones standing near the roadside at intervals, and recognized the custom of marking boundaries which obtained in Jacob's time. There were no walls, no fences, no hedges – nothing to secure a man's possessions but these random heaps of stones." Wilson in 1906 reports that to mark off two plots a double furrow was driven between them and piles of stones were set up at short intervals within the furrow. These piles were usually quite small.

Care has to be taken not to identify all stone piles seen in the fields as markers delimiting ownership. Many of the larger piles in the fields are clearance heaps of surplus stones (see Shev. 2:3; 4:1). Others covered the ruins of ancient structures and this was referred to in Oho. 15:7. The remains of numerous structures were found beneath piles of stones during archaeological surveys around Jerusalem. Some of these may have been memorial heaps, for example the large stone pile (gal avanim) mentioned as having been erected over the place where Achan was stoned and set on fire (Josh. 7:26). A group of large stone piles investigated in southwestern Jerusalem, dating from the Iron ii, may have been memorials of this sort, or perhaps they were connected with harvest rites since agricultural installations, including a wine press, are known in their immediate vicinity. Many piles of stones on the top of hills were perhaps connected with peasant rituals, and Rabbi *Akiba (second century c.e.) warned that "where-soever thou findest a high mountain or a lofty hill and a green tree, know that an idol is there!" (Av. Zar. 3:5). Aborted fetuses were sometimes buried within cairns (Oho. 16:12). At the beginning of this century, stone cairns (rujm) were being set up at sites where murders had taken place. A stone heap of this kind was known in Arabic as meshad, i.e., a "witness."

All individual fields were originally marked out in one way or another, even if only in a very rudimentary fashion. It was only then that the clearing and plowing of the land could begin. All agricultural work carried out with plows or with hoes produce scarps or banks of soil along the edges of the plots being prepared for cultivation. Loose stones from the surface of the field, which are a hindrance to plowing procedures, would have been thrown to the edges of the plots, thus creating boundary lines of low piled stones. In areas where there was very little competition for the land, this way of marking out the plot was clearly sufficient. Alternatively the plot of land could be marked out with a ditch (which had to be at least 3 ft. (0.93 m.) deep and about 1 ft. (0.37 m.) wide according to Kilayim 4:1). Ridged or dug-out boundaries of this sort may be detectable on the ground but field boundaries made out of perishable materials, such as piled logs of wood or reeds, will usually not be archaeologically evident.

In areas where there was pressure on the land or where there was a possibility of disputes between neighboring farmers, a proper system of field boundaries became necessary. Such fences tended to be built of stone where rocky outcrops were available to the farmer. In areas where stone was not freely available, especially in the plains and in the broad valleys, the field boundary was probably defined by a furrow in the ground, a wooden fence, a row of fruit trees (see Or. 1:1), or a thorny hedge. Archaeological evidence for such boundaries is rarely found. Stone fences, however, survive quite well and traces can still be seen in many parts of the country, as well as in the arid areas of the Negev and Judean Deserts.

Thorny hedges are referred to in various biblical passages. In the Parable of Jotham there is mention of plots of fruit trees edged by the atad bush which, if unattended, would easily grow wild (Judg. 9:15). It has been suggested that this was the thorny bush (Lycium europaeum) which still grows in the Kisuphim area where it originally was used during the Byzantine period as a hedge around fields. The mesukat hedek (identified as Solanum incanum L.) is another type of thorny hedge, sometimes mentioned in association with stone walls (Isa. 5:5; Prov. 15:19; Hos. 2:6). Hedges surrounding fields were also mentioned in the Mishnah (bk 3:2). The hedge was referred to as the hasav in tb, bb 55a. The thorny bush Sarcopeteria spinosum is presently used by traditional farmers as a hedge around cultivated fields. The practice of surrounding fields with cactus bushes (Arabic sabreh) dates back only to the Ottoman period when the cactus was first introduced into Ereẓ Israel. In the 19^th century, rectangular fields were cleared within the woodland of northern Golan, leaving rows of oak trees standing as boundaries between the various plots.

Stone fences were usually built to a height of about 3ft. (1 m. and with an average width of over 2 ft. (0.75 m.). This wall (known in Arabic as jedar) was built without the use of mortar. It has been estimated that a modern builder could construct a dry-stone wall around 15 ft. (4.5 to 5 m.) in length, 1.5 ft. (0.50 m.) thick and 4 ft. (1.4 m.) high (using 115 cu. ft. – 3.3 cu. m.) of stone) in the space of one working day. Principally the walls served to keep animals (dogs, foxes, and jackals) away from the crops. Frequently the tops of such walls were reinforced with a thorny bush (netsh in Arabic; usually this was the Poterium spinosum in the central highlands). The bush would project a few inches beyond the upper edge of the wall and was kept in place with the weight of a few stones. Repairs to these walls were always made in time for the period leading up to the vintage. In the Bible, the stone fence around a vineyard is referred to as the gader avanim (Prov. 24: 31; cf. Isa. 5:5; Ps. 62:4). Special builders known as goderim were employed for the task of building stone fences (cf. Ezek. 22:30).

The stone fence was also known as a gader among Jewish farmers in the Roman and Byzantine periods (Matt. 21:33; Pe'ah 2:3; Oho. 17:3). The hayis apparently referred to the secondary partition wall in the field systems and the gader to the main enclosure wall. The boundary path between two plots of land was referred to as the mesar (tj, Kid. 1:5). It was apparently not customary to build stone fences in the flat lands of wide valleys (bb 1:2). The minimum measurements that this fence had to have according to the Mishnah, was a height of 3 ft. (0.93 m.) and a width of 1 ft. (0.37 m.) (Kil. 2:8). If it was less than this height then it could be regarded as a "quarry" and its stones dismantled and taken away. However, its foundations (to a height of about 4 in. (10 cm.) always had to be left intact (see Shev. 3:6). The boundary markers were set up outside the line of the fence so that if the wall collapsed the farmer still could claim ownership over the place and the fallen stones. However, in the case of two separately owned plots of land, the boundary stones were set up on either side and if the wall collapsed then the stones belonged to both of the farmers (bb 1:2). Ditches were sometimes dug along the foot of the fences (bb 7:4) and in a vineyard a border 7.35 ft. (2.24 m.) wide was left uncultivated between the fence and the vines (Kil. 6:1).

The use of stone fences around fields has been found dating back to the period of the Iron Age ii. The fact that they have not yet been found does not mean, however, that stone fences were not built during earlier periods as well. Iron Age stone fences have been documented at sites in the Judean Hills, but fewer in Samaria. A possibility to consider is that stone fences around fields may originally have been derived from the type of stone walls used to surround animal pens, for example those from the Iron Age i which are known from Mount Ebal and Giloh.

Roads can be most useful in outlining the borders of agricultural lands or in separating one group of fields from another, particularly in defining the extent of fields which were not in any other way demarcated with stone fences. Three types of roads are known from the rural countryside of Ereẓ Israel: highways, regional roads, and local rural roads. A border of about 25 ft. (8 m.) was left uncultivated on either side of the highway, which was of no prescribed width (bk 6:4). The local roads were the access routes which linked the farms and villages with their fields and areas of pasture. They also gave farmers access to the regional roads leading to the market settlements. Produce would have been conveyed to market using beasts of burden. Wheeled transport was rarely used as roads were too narrow and stony. Most local roads extended between sites with prominent water sources (such as springs) and served as "corridors" between blocks of fields. Cisterns and wells are frequently found alongside both types of roads. High walls built of stone were erected on either side of roads, to separate the public space from the agricultural land. They prevented animals from entering unattended fields and damaging crops, especially the flocks of sheep and goats which were shepherded along these roads from the animal pens next to the settlement to the area given over to pasture which was usually located some distance away. The walls also served to discourage passing travelers from entering the fields and picking fruit during the harvest seasons.

Natural "gates," or private roads, led from the local roads into the field systems. Local roads were either for public use or were under private ownership and a clear distinction was made between the two during Roman times (see Pe'ah 2:1 and Kil. 4:7). A private road (derekh ha-yaḥid) was prescribed as having a width of about 6.5 ft. (2 m.) and a public road (derekh ha-rabbim) a width of about 25 ft (8 m.) (bb 6:7). A road ending at a cistern, lime kiln, cave, or wine press was usually a private road (Toh. 6:6). In some cases blocks of fields may be observed laid out within a pre-existing network of local roads. When such roads can be dated, they can then provide a terminus ante quem date for the field systems. Datable roads when encroached on by field systems will provide a terminus post quem date. Bowen (1961) wrote that the "integration of fields with any trackway leading into a settlement seems the best assurance of contemporary relationship, or at least in some phase, though the individual fields might of course have suffered considerable subsequent alteration and not necessary be characteristic of the period of the settlement."

Shape and Size of Fields

Agricultural fields in Ereẓ Israel were usually very small and approximately rectangular. There were, of course, exceptions to the rule as well as regional and local variations. The fields in the flat areas of the plains and broad valleys were on the whole proportionally larger and more rectangular than the fields located on sloping ground. Fields in extremely rocky terrain were even smaller in size. The shape of an individual field on sloping ground was quite often determined by topographical and lithological factors. The edges of the fields were either defined by natural features (such as a wadi, a gully, rocky outcrops, etc.) or by man-made boundaries (such as stone markers, stone fences, or terrace walls). The size of the fields also depended on the type of farming technology available to the farmer. Plowed land was always larger than land cultivated only with hoes. This was also true of the digging and clearing equipment which were essential in the highlands for clearing stones, breaking up the ground, and flattening the field surfaces. Different sized fields are to be found in the areas around the villages. Some of this may be the result of the fragmentation of land owing to the division of property at different times. The fields associated with farms, however, appear to have been much more uniform in size and were sometimes part of integral systems that were quite well defined. Compared to the fields used for dry farming, irrigated plots of land tended to be smaller and extremely regulated, with flatter surfaces and well-defined boundaries.

Fields also tended to be of rectangular shape because they were easier to measure for the purpose of estimating surface area or the quantity of crops grown in them. Measuring lines (kavei ha-middah) or cords (ḥevelim) are mentioned in a number of biblical passages referring to the measuring of lands and fields for the purpose of division (ii Sam. 8:2; Isa. 34:17; Jer. 31:39; Micah 2:4–5; Amos 7:17; Ezek. 47:3; Ps. 16:6). The "line" or "cord" had knots at specified intervals along its length. The measurement of the distance between the knots of the biblical cord is unknown. Perhaps it was the length of the kaneh ("measuring reed") mentioned in Ezekiel 40, which was used for measuring small areas and expanses. A tax assessor who is seen measuring standing crops with a knotted cord to determine the yield was skillfully depicted in a painting on the wall of a New Kingdom tomb in Egypt. The knotted cord is shown extended horizontally across the top of the crops and held between the assessor and his assistant. A roll of additional cord is shown slung across the assessor's left shoulder.

In Roman Ereẓ Israel, land which was to be sold had first to be "measured by the line" (bb 7:2–3; see also the methods of measuring differently shaped fields as explained by Columella v, 1,13–2, 10). It is reported that the farmers of Beth Namer "used to reap their crops by measuring-line and leave pe'ah (gleanings) from every furrow" (Pe'ah 4:5). The division of agricultural land by surveying is mentioned in one of the Nessana papyri (P. Ness. 58) dating from the late seventh century. Measuring cords are also mentioned in a passage from Pirkei de-Rabbi Eliezer, dated to between the eighth and early ninth century, which deals with the Muslim conquest of Ereẓ Israel in the seventh century c.e. It says that the conquerors will "measure the land with ropes, and make the cemetery into a dunghill where the flock rests, and they will measure them and from them unto the tops of the mountains …." The knotted cord during the Roman and Byzantine periods was probably marked out with a measuring rod of a standard size used in Ereẓ Israel (cf. Kel. 7:6), perhaps identical to the orgua or kalamos measures. An iron measuring rod, five cubits in length and dating to the seventh century, has been discovered during excavations at an ecclesiastical farm at Shelomi in Western Galilee. This rod may have been a kalamos measure of 11 spithamai, 8.5 ft. (257.4 cm.) Long cords for the purpose of measuring village lands for taxation purposes were still used by the Ottomans in the 19^th century. In Southern Ereẓ Israel, lands were divided up with a measuring line (ḥavaleh) before plowing. This line was made of goat's hair and was a little more than one centimeter thick.

The earliest archaeological evidence for the general appearance of the layout of field systems in Ereẓ Israel dates from the Iron Age ii. This may have been the period during which widespread field systems were first laid out as blocks of rectangular-shaped plots on high ground and along the valleys. In some areas, it would appear that these blocks of fields developed piecemeal, but in the highlands and in the semi-arid regions a number of these systems appear to have been planned in advance. However, it is only during the Roman and Byzantine periods that fields were laid out in a specific pattern in different parts of Ereẓ Israel and Syria. Some even follow Roman land-partition principals.

The size of the individual field used for arable purposes during antiquity was usually the area of land which could be plowed with a yoke of oxen during the course of one day's work. According to Bava Batra (1:6) the practical size of sown ground in dry farming was an area of 11,000 sq. ft. (1,176 sq. m.) (but according to R. Judah – 2^nd century c.e. – this could be reduced to as little as 5,500 sq. ft. (588 sq. m.). The size of plots used as vegetable gardens and vineyards were even smaller. The size of plots were apparently progressively reduced in the course of the third and fourth centuries. In Samaria, at Khirbet Buraq most individual plots were between 1 and 7.5 dunam in size, with the largest not exceeding 15 dunam. The size of individual plots is in fact not helpful in an attempt to reconstruct the area of smallholdings, since most families in the Roman period would have owned more than several plots and these could very easily have been scattered. Furthermore, a family could also have controlled areas of pasture and woodland outside the field system.

Various suggestions have been put forward regarding the size of the lands cultivated by the individual peasant small-holder in Roman and Byzantine Ereẓ Israel. As a result of his archaeological surveys in Samaria, Dar has estimated the size of the typical family holding around a number of Byzantine village sites at between 25 and 45 dunam (note that four dunam are the equivalent to one acre). At Khirbet Buraq the holdings were estimated at 25 dunam each and around Qarawat bene Hassan at between 39.7 and 45.6 dunam in size. These figures do not take into account the possibility of the scattered ownership of land. At sites in the Galilee, holdings averaged between 6 and 11 dunam and around Nablus between 15 and 18 dunam (compare these figures with the average of 16 dunam for a block of enclosed fields noted by some scholars for the Galilee).

In a study of the economy of Roman Ereẓ Israel, Safrai proposed that an average family of four individuals, practicing subsistence agriculture, would have required lands covering an area of approximately 11 dunam (or 13.7 dunam for a family of five). However, with the burden of taxes and the need for outside purchases a minimum of 20 dunam seemed to him to be a far more realistic figure. Safrai's figures are based on calculations regarding the possible harvest yields of plots used for the cultivation of wheat, olives, figs, grapes, and legumes. However, such yields could have varied quite substantially from one part of the country to another, depending on the type of soils available, location and precipitation levels.

Broshi in another study has suggested that a family of five needed an area of 50 dunam for subsistence, a figure which seems quite reasonable compared to estimates of the sizes of traditional holdings during recent times. This also fits well with the estimated minimum requirement of 20 iugera (= 50 dunam) for a farm in Roman Italy which cultivated land with the plow and kept animals. However, Broshi does not believe that manuring was carried out in the sown areas and so in antiquity the smaller areas did not produce yields which were any greater than those of traditional agriculture of the 19^th century.

A number of ancient sources exist dealing with the question of the size of the peasant holding in antiquity (note the following equivalents are used here: 25 dunam = 2.5 hectare = 10 iugera = 20 plethra). In his Historia Ecclesiastica (iii, 20:1), Eusebius mentions Judas's nephews cultivating an area of 39 plethra (= 48.7 dunam). Thus each farmer would have had a holding of about 24 dunam. Talmudic sources also contribute much information, but more about the quantity of yields and the measurement of individual plots of land than about the size of the individual holding. Additional figures are available regarding the sizes of holdings in other ancient Mediterranean countries and Egypt. Apparently the smallest and largest allocation of land in Roman Italy was from 2 iugera (= 2.5 dunam) to 200 iugera (= 5,000 dunam) or more. Since 2 iugera was clearly insufficient to support a family even on a subsistence level, this allotment must have been additional to areas used for pasturage. In 133 b.c.e., allotments for the poor in Italy measured between 10–30 iugera (= 25–75 dunam). In Greece, in the Metaponto area, plots varied between 48 and 530 dunam, with 50 percent of these totaling about 138 dunam in size. In Egypt, the average property size varied during the mid-fourth century: 76 iugera (190 dunam) at Hermopolis and 37 iugera (92.5 dunam) at Antinoöpolis).

The above figures make it quite clear that there are many difficulties in estimating the size of the average small peasant holding in antiquity, prior to the Abassid period when there was a general decline in the agricultural productivity of the land lasting until modern times. However, the size of a holding will have differed considerably from one type of environment to another, with larger holdings in the more rugged environments having poorer soils, and smaller allotments in the flat fertile plains and internal valleys where there was much more pressure on the land. At times when there was a serious fragmentation of holdings, with lands being absorbed by larger estates, such as during the 3^rd–4^th centuries c.e., there would have been pressure on privately owned ancestral holdings and these would naturally become smaller with individual fields being sold off piecemeal. The size of the holding also depended on the type of agricultural regime being practiced and whether or not it included the replenishment of soil fertility with proper manuring and fallowing procedures. The lack of such procedures could well have restricted the size of the areas being cultivated. Finally, the minimum size of a subsistence holding will differ considerably between one which had to support a family and one which supported an extended family. Furthermore, a peasant farmer could have held more than one holding. The attempts made by Dar and Safrai to divide up the fields surrounding ancient villages into holdings of theoretical sizes, seems to be fundamentally flawed when one considers the evidence regarding the lands owned by Rabbi Yohanan in the 3^rd century, whose fields were not concentrated at one location at all but scattered at different locations in the countryside between Tiberias and Sepphoris.

However, the existence of such problems should not mean that scholars should abandon the attempt to delimit the minimum size of peasant holdings in Ereẓ Israel during the different periods. The maximum cultivable land in Ereẓ Israel in 1931 amounted to 6.54 million dunam which, subdivided by the population number of 600,000 for that general period, provides an average of 11 dunam of land per inhabitant. This figure can be compared favorably with Broshi's suggested 10 dunam of land per individual in antiquity. However, two facts need to be taken into consideration regarding the 1931 figure: first, not all the population at that time were involved in agriculture and, secondly, a large percentage of the available cultivable land was not necessarily being cultivated. For example, records show that up to 55 percent of the hills around Nablus-Tulkarm had neglected terraces in the Mandate period. Hence, a more realistic minimum of 5 dunam per individual seems likely. A family of five would therefore have had a holding of approximately 25 dunam or more. Hence, the suggested size of a peasant holding in Ereẓ Israel during the time preceding the Abassid period, i.e., before 750 c.e., probably varied considerably between 25 and 50 dunam.

It is interesting to compare these figures with information known about the size of the average smallholding in different parts of the Near East during more recent times. An average of 50 dunam per peasant family has been estimated according to figures taken from an Ottoman census of 1909, for the lands in the three sanjaqs of Jerusalem, Nablus and Acre; 85 percent of landowners from Hebron cultivated areas ranging between 1–20 dunam. The sizes of average holdings in three present-day agricultural villages in the Zarqa River basin in Jordan range between 19 and 68 dunam. In Syria, however, a peasant holding in an area of dry farming was much larger, averaging 50–70 dunam. At Episkopi in Cyprus, the village was estimated to have had a total of 10,000 dunam of land and a maximum population of 700 individuals; thus a holding for a family of five would have averaged at about 71.5 dunam. At Ashvan in Anatolia the total landholding per family averaged 122 dunam but 40 dunam of this was used for pasturage.

The systematic study of field systems during archaeological work, combined with a critical examination of evidence gathered from a variety of sources, provides very important information on the history of rural land management in different parts of Ereẓ Israel. One of the major problems in establishing a typology of field systems for Ereẓ Israel is that the best evidence comes from hilly areas and from the marginal semi-arid areas, and it is questionable whether these patterns were also the same in areas of flat land. While the tenurial organization of land cannot be worked out solely from the archaeological remains of the fields themselves, the archaeological work is able to provide historians with empirical data on the appearance and extent of ancient field systems of different periods. In some field systems the evidence points to continuity, with the position of boundaries being respected and retained from period to period. In some cases substantial changes have occurred and this may have been the result of the dispossession of lands, new property relationships, or new forms of exploitation.

Dating fields is still an acute problem for archaeologists. One scholar wrote in 1984 in a pessimistic vein that in his opinion "it is very rarely possible to attribute an ancient agricultural field or system of old terraces to a specific period." Research in the 1990s and early 2000s shows that this is no longer the case and methods of landscape archaeology enable scholars to disentangle field systems belonging to several periods. The first step, of course, is to evaluate the degree of association between a system of fields and adjacent villages and farms, and the network of communications (paths, roads, etc.). Field boundaries and their relationships with earlier features, from dolmens to earlier fields, must be examined. Some of these earlier features may have been respected or dismantled. The extent of a field system also needs to be determined. It is not always an easy task to establish where one system stops and another starts. Attempts should also be made to examine patterns of field boundaries and the stages of construction of stone fences defined through structural analysis. Some field systems are of an aggregate pattern and others may have a planned parcelation arrangement with a more rectilinear design. Some systems are responsive to the local terrain, others are not. Ancient boundaries are frequently "fossilized" in modern boundaries. A comprehensive study subsequently needs to be made of sherd scatters located next to the settlements, around cisterns, wine and oil presses, alongside the edges of fields and across the field surfaces. Many of these sherd scatters may turn out to be the result of ancient manuring practices. This will provide the surveyor with a general idea as to the date of the fields which are being examined. The excavation of test pits within the settlement and next to the various features seen within the area of the fields, can serve as a "control" over the results from the survey. The overall results will show whether or not there has been cultural continuity or discontinuity within a given landscape of agricultural fields.

Recent work has shown that a variety of ancient land-management systems were operating in different parts of Ereẓ Israel in antiquity. These reflect the many different ways that the rural population responded to environmental and climatic constraints. The quality of these responses depended on the level of social and economic organization achieved by the different farming communities and on the level of farming technologies available to them. The tactics of village communities differed considerably from those of individual farmers, and this had a clear effect on the general layout of fields in a landscape. The earliest known evidence for the imposition of a regular and widespread cultivation pattern in Ereẓ Israel can be dated to the Iron ii period. The parcelation of land was consolidated during the Hellenistic to Roman periods. Thereafter, areas of field systems expanded and in time, particularly during the Byzantine and Umayyad periods, became much more complex and fragmented than before.

Today, the shape and character of fields in Ereẓ Israel have changed dramatically. Mechanized equipment now allows for the cultivation of extremely large tracts in areas of flat land. Terraces can be cut into the slopes of hills with bulldozers and soil can be transported by truck from one place to another. The use of artificial fertilizers has meant that fields no longer need to lie fallow and yields can be constantly high. A wide variety of weeds, however, are thus being banished to the margins of fields and complete wildlife habitats are being destroyed. Large areas of fields in the traditional landscape have already been eradicated. Their boundaries and fences have now been erased forever and can only be seen in aerial photographs. The morphology of the agricultural landscape is no longer the same and the growing of cash crops has replaced the economic variability of ancient times. While it is sad to observe the rapidly disappearing traditional landscapes, harvests in Israel, the West Bank, and Jordan are now much greater and more plentiful than they ever were in antiquity and this can only be to the advantage of the local farmers of today.

Manuring Procedures

In ancient Ereẓ Israel, an enormous amount of effort was dedicated to the proper organic fertilization of fields and periodic fallowing. The archaeological evidence for ancient manuring is represented by the very large quantities of sherds which can be found scattered in the fields all over the country. In many cases they represent frequent farming procedures undertaken to improve the fertility of the soil, especially in areas where frequent cropping was undertaken with very little fallowing. The fertilizers and manure had to have the right composition of potassium, phosphorus, lime, magnesium and nitrogen, allowing for the proper growth of the crops in the fields.

Fertilizers and manure of all sorts were extremely valued in antiquity as soil-improving agents. The most frequently-used manure was the dung of cattle, sheep and goats which was collected from the mangers and animal pens and distributed in the fields with household rubbish. The term domen was used in the Bible when referring to manure left in the field as an organic fertilizer (ii Kings 9:37; Ps. 83:11; Jer. 9:21). When animal dung was mixed wet with straw, it was known as madmena (Isa. 25:10). Household rubbish and sherds were probably added to the dung while wet and this would have helped break down the dung before being scattered in the fields. Human excrement was not apparently used and was clearly forbidden by Jewish custom, unlike in China where human faeces, diluted and fermented, was one of the principal manuring agents used in the fields. However, ashes (efer: cf. Ezek. 28:18) and burnt ashes from animal sacrifices (deshen: cf. Isa. 34:7) were probably used. This was certainly the case during the Roman period when ashes of a sacrificed animal were "sold to gardeners as manure" (Yoma 5:6). A dove-raising industry existed in Ereẓ Israel in the Hellenistic and Roman periods and is represented by the discovery of numerous columbaria caves and structures. The dove droppings from these columbaria were highly sought after especially for fertilizing valuable garden plants. The rule was that columbaria had to be located at a distance of 25 m. from the farm or village (bb 2:5). Organic waste materials such as the cakes resulting from the pressing of olives, were pulverized and then added to the compost heaps which were then later scattered on the orchards and gardens. In Roman times, an importance was attributed to fertilization procedures with organic materials and compost produced from organic waste (bk 3:3; Shev. 2:14, etc.). The importance of fertilizer is shown by the comparison of dung to precious stones (Sanh. 59b). Blood (Yoma 5:6), ash, and fine sand (Tosef., Shab. 8:9) were all used. Sheep droppings were applied by enclosing the flock in a temporary fold ("sahar"). The enclosure would be set up for some time and then be moved from place to place in the field (Tosef., Shev. 2:15, 19).

Manure was used in orchards, gardens, and sown fields. The best time to manure the fields was after the first rains in time for the winter crops and in the spring for the summer crops. The rains softened the ground as well as the manure itself. In irrigated areas, manuring was continued throughout the year. Manure was distributed to the fields in special baskets (known in Latin as sirpeas; cf. Kel. 19:10; 24:9) which were loaded on beasts of burden. It has been estimated that 235 cu. ft. (6.7 cu. m.) of manure were used per dunam of land. The manure was first deposited in heaps in the field and was then scattered by a process of plowing or hoeing (cf. Shev. 3:1–4). Seeds from weeds could sometimes get into the soil of a field through animal dung and this was something which concerned Jewish farmers during the Roman period (Kil. 5:7). A field that had undergone manuring was known as a bet ha-zevalim. Movable enclosures were sometimes set up in the fields to help localize the dung of grazing animals. In dry soil, manures could effect the movement of water between the surface of the ground and the subsoil, and so manures had to be only lightly covered. But care had to be taken to make sure that the manure was dug in properly, since unrotted manure could burn crops (White 1970: 129). Manure was heaped around individual trees and then dug in. In Luke 13:8 reference is made to the manuring of a fig tree so that it should bear fruit.

Broshi has suggested that manure was mainly used in orchards and gardens during the Roman and Byzantine periods and not at all in the fields that were used for the growing of grain. The claim has also been made that because of pressure on the land only one third of it was actually allowed to lie fallow during that period, but this would have been contrary to the Jewish rulings of the time (bm 9:7). The true extent of the manuring practices at that time is clearly demonstrated by the enormous distribution of sherd scatters in areas used for the growing of grain across the country. Manuring at a distance of more than a mile (1.5 to 2 km.) from the village would probably have been uneconomical without the use of carts.

The phenomenon of sherd scatters representing manuring regimes is known from ancient fields investigated in many of the Mediterranean lands, from Spain to Greece, and as well as in other parts of the world. Various methods have now been developed for the study of the distribution of sherd scatters within agricultural lands and around settlements. In Ereẓ Israel, organic household, and courtyard rubbish was gathered and used as a fertilizer for the fields, and this rubbish frequently included broken pottery. Household refuse was sometimes mixed with animal manure to make it easier to scatter in the fields. In Samaria, sherds have been found in the fields around Qarawat bene Hassan, up to 2–3 km. from the village center, as well as around Khirbet Buraq and in other fields. Sherd distributions have also been studied in the fields around agricultural settlements in the Golan. Soil was sometimes taken from the ruins of ancient sites and scattered on the fields as a fertilizing agent. Archaeologically, this is known only from sites with plots used for irrigation purposes, as at Sataf west of Jerusalem.

In the Ottoman period sown fields were very rarely manured and in many areas this eventually resulted in the substantial exhaustion of the land during the 19^th century and the early part of the 20^th century. Manure was mainly kept for use in vegetable gardens and some orchards, or dried and used for fuel. However, Baldensperger in 1907 does mention the manuring not just of orchards but also of sown ground. Karmon and Shmueli have also pointed out that unlike other parts of the country, the regular manuring of terraced plots was being carried out around Hebron until the 1970s. Reifenberg in 1947 complained that organic manure was often very badly handled in Ereẓ Israel: "the heaps being exposed not only to the intense sun but also to the torrential winter rains. The dung is very often left for weeks in small heaps on the field instead of being plowed in as quickly as possible."

Irrigation

Although Ereẓ Israel was described as possessing "brooks of water, of foundations and depths, springing forth in valleys and hills" (Deut. 8:7), the reality is that there were very few permanent sources of water that could be used for agricultural purposes, namely a number of perennial rivers, and springs and wells, and mostly what local farmers could rely on was winter rainwater (Deut. 11.10–11); otherwise their lands would remain dry (forlorn like "the garden that hath no water," Isa. 1:30). The few permanent sources of water that did exist were therefore exploited to the fullest and the danger of drought was always present (Deut. 11:17). Water rights and ownership of access to water led to serious disputes in antiquity (e.g., the disputes with the Philistines regarding wells of water, Gen. 21:25; 26:15–22). Rabbinic legislation covered many aspects of the distribution of water among those who shared in water rights. "Water turns" were assigned (mk 11b) to those entitled to use the supply. Some individuals were compelled to buy or lease seasonal rights (tj, mk 1:2, 80b). Many biblical passages make use of the well, spring, and river as symbols of abundance and security (Isa. 58:11; Jer. 31:11). Orchards, too, grow better when partially irrigated. Hence the comparison made between the person who puts his trust in God to "a tree planted by streams of water" (Jer. 14:8; Ps. 1:3).

Irrigation methods were many and diverse. In those parts rich in rivers and springs, as in the Jordan and Ḥuleh valleys, water flowed into the fields by gravitation and was directed through channels dug with shovels or pressed down by foot (cf. Deut. 11:10). Trees were similarly irrigated (Ezek. 17:7; Song 5:13): a shallow pit dug around a tree was filled with water, which would sink deep into the ground and moisten the roots (Kal. R. 3:52, 4). However, mechanical means were sometimes needed to lift water into the fields. The most simple of containers was the deli ("pitcher") of earthenware or metal attached to a rope or chain (Shab. 15:2; Kel. 14:3). A similar utensil was the ḥavit or jug, which was also used for water drawing (Mak, 2:1). A larger device for lifting water was the kilon of mishnaic times (Makhsh.. 4:9, etc.), which was also frequently depicted in ancient Egyptian and Babylonian representations; it has survived up to the present time in the traditional agricultural practices of Egypt (known there in Arabic as the shaduf). It consists of a vertical pillar on which a long horizontal bar is placed. To the one extremity of the bar, a jug is attached by rope or pole, while the other extremity is weighted down by a stone to balance and facilitate the raising of the full container. Herodotus described the Egyptian use of this device (Gr. κηλωνήιον, "keloneion"; cf. History 1:139). The bucket wheel was used at wells for irrigation and was turned by a horse, mule, or ox. Water was conveyed to the fields by conduits on high stone walls, sometimes arched. Examples of waterwheels used to raise water from rivers and wells of the "Persian" type with ceramic jars attached to them (antila), are known in Ereẓ Israel dating back to the Roman period. The antila is essentially a vertical wheel to which earthenware pitchers or wooden containers are attached. An animal rotates a shaft whose attached, wooden-toothed gear in turn engages the wheel to which the drawing pails are tied. According to the halakhah, water drawn by an antila invalidates a mikveh, since the waters are separated from their source during the process. The Tosefta (Makhsh. 3:4; Mik. 4:2) designates another vessel, the "kevulin" or "keḥulin" whose use does not invalidate the mikveh, since the water is not detached from its source during the operation. This keḥulin (Lat. cochlea, "snail") consists of an Archimedean screw in a pipe. The screw, turned by an animal, forces the water to rise. Strabo (17:807) reported that water for the Roman camps was drawn from the Nile by means of κοχλίαι. The galgal or gullah described in Ecclesiastes (12:6), with a wheel turning an axle to which a pail or two were attached by rope or chain, may have been an earlier type of antila device. Such a device was frequently mentioned as μηχανή ἁντλοῦσα (i.e., "drawing machine") in Hellenistic Egyptian papyri. Evidence of its use has been found in the south of Israel in a Byzantine Greek inscription that reads: "This excellence too of μηχανή, the glorious father Helarion invented." The horizontal waterwheel may have been developed in the Upper Galilee during the second century b.c.e., but clear archaeological evidence for this is still lacking. The large upright wheel operated by water power (na'urah in Arabic) was first developed in Syria, the most famous examples being those at Hama. It was later introduced into Ereẓ Israel, where it was used to draw water from some of the perennial rivers and is referred to in the halakhah as a device with "self-drawn water from the sea or river" (Tosef., Mik. 4:2, so in correct texts). The Jerusalem Talmud designates this instrument by the name "agatargatkaya" (tj, mk 1:2, 80b), which is related to the Greek term καταρράκτης, "cataract" or "waterfall." It refers to the cataract effect created by the water-driven wheel. In addition to these water wheels, underground Qanat systems were constructed to exploit the shallow groundwater in the arid basins of the Arabah and the lower Jordan Valley. Examples of these systems have been surveyed by Porath, who has suggested, on the basis of his fieldwork, that the knowledge of qanat construction may have been introduced into the region from Iran either during the late fourth century b.c.e. or (and this seems much more likely) during the Umayyad period in the seventh century c.e. However, the use of these chain wells did not last beyond the Abassid period.

The most important source of water in the hilly regions are the springs, and an estimated 800 exist in Ereẓ Israel. In ancient times the lands adjacent to the springs were mainly used for irrigation purposes. Only a handful of these springs, namely the Dan, Yarkon, and Na'aman springs, have a flow of more than one cubic meter of water per second. A little more than 40 produce a volume of between 100,000 and 1,000 liters per second. The rest produce much smaller amounts of water, sometimes as little as one liter per minute. Irrigated plots of land tended to be located fairly close to routes giving ease of access for those conveying their produce (such as vegetables) to the markets or back to the village or farm. Irrigated plots tended to be fairly close to the settlement for security reasons. Most springs in the hills of Ereẓ Israel had various installations attached to them, including flow tunnels and large water-storage pools. The irrigated fields were located below the pools and water was conveyed to the fields along channels by a process of gravitation (it would "leave higher ground and go to the lower," Ta'an. 7a). Each terrace area received water once every cycle of specified number of days, according to the size of the terraces owned. Spring-irrigated lands have been investigated during archaeological surveys in Samaria and in the Judean Hills. Spring-irrigated system of terraces have been investigated at Ain Yael, Sataf, and Suba.

Artificially watered or irrigated fields were frequently mentioned in the Mishnah. Irrigated land was known as bet ha-selahin to distinguish it from the areas of rain-fed land known as bet ba'al. The smallest irrigated allotment was between 350 and 700 sq. ft (32.5 and 65 sq. m.) (bb). It is interesting to compare this with the allotment of 1 dunam of irrigated terraced land which every family farms in present-day Battir. Title over irrigated lands was given to anyone who could prove three years undisputed possession (bb 3:1). The surface of the irrigated field was divided up into square plots of about 5 × 5 sq. ft. (0.56 × 0.56 m.), separated one from the other by an earthen border (gubal) about 4 in. (10 cm) in height (Kil. 3:1). Irrigated plots were heavily manured and the frequent manuring of beds of cucumbers and other vegetables was mentioned in Shevi'it 2:2. Soil taken from ancient ruined sites was frequently used to top-up and fertilize irrigated fields. The furrow or channel which brought water to these plots had a width and depth of about 10 cm wide and 10 cm deep (Kil. 3:2). Channels led from tree to tree (mk 1:3). In large areas of irrigated lands, small plastered tanks were frequently built at the end of the main irrigation channels. Because of evaporation rates, these small tanks were the first to be filled with water: "The cistern nearest to a water-channel is filled first – in the interests of peace" (Git. 5:8). It was only then that the plots between the spring and the tanks could be directly irrigated from the main channel. Springs in areas of irrigated fields were known to suddenly dry up (bm 9:2). A distinction was made in Mo'ed Katan 1:1 between old springs and "newly flowing springs."

Irrigation plots were frequently established next to aqueducts leading from the springs. Plots of land irrigated in this fashion existed next to the aqueduct which led water to the settlement of Na'aran in the Golan. An interesting Greek inscription of Justinianic date was found near the principal aqueduct leading to Jerusalem. It prohibited the use of the land on either side of it, up to 15ft. (4.6 m.), for cultivation purposes. This was to prevent water from being siphoning off for irrigation. This prohibition reminds one of the ban in Roman Ereẓ Israel regarding the cultivation of plots of land on either side of highways.

Cistern irrigation was also practiced in Ereẓ Israel and dates back at least to the Iron Age (e.g., Isa. 27:3). Vines watered by irrigation are mentioned in a field with furrows (aru got) in Ezekiel 17:5–7. Cistern-irrigated plots are mentioned in Jewish sources of the Roman period (see mk 1:1;) and in Pe'ah 5:3 a field was irrigated "with a pitcher." Cistern-irrigated plots of land have been investigated by Dar at a number of sites in Samaria (1986a, 200–2). A cistern in an orchard is mentioned in one of the Nessana papyri (Kraemer 1958).

Traditional irrigation in Ereẓ Israel clearly perpetuates ancient practices as Dalman was able to show in his publications. Many of the Arabic toponyms in Ereẓ Israel are linked to the word ain ("spring") indicating the importance of the spring even when it was located a few kilometers away from the settlement. Descriptions of traditional spring-irrigated plots and watering procedures were frequently published by 19^th and early 20^th century travelers. Water was divided up for irrigation either by degrees or by hours. General research on irrigation has been carried out by Avitsur and at springs in the terraced Judean Hills by Ron. The irrigated areas investigated by Ron comprised only 0.6 percent of the total of the terraced areas west of Jerusalem. This percentage is probably true for all of the terraced areas of the hills of Ereẓ Israel.

Installations and Implements

Associated with each system of ancient agricultural fields, in addition to the villages and farms, are the remains of solitary structures, towers, caves, cisterns, columbaria, wine and oil presses, threshing floors, and animal pens. Pools for steeping flax and fig-drying installations are also mentioned in the sources. Many of these features were linked by roads and paths. Near them were lime kilns and quarries.

The existence of rural towers in areas of terraced fields has been frequently commented on by travelers and scholars since the 19^th century. A detailed study of these structures was carried out by Ron in the terraced zones of the Judean Hills. Approximately 50 percent of these stone towers were associated with vineyards and 39 percent with orchards. In general terms, the layout of the tower resembled the organized space of the traditional village house, with the lower area serving for storage and the upper part for habitation. Towers were usually located at a distance of 500 meters and up to a kilometer from the village or according to Wilson writing in 1906 up to 6 or 8 km. from the village. The interior was perfect for the cool storage of agricultural produce, since its temperature was 8 to 13 degrees cooler than outside it. Water cisterns and wine presses were frequently found next to the towers. Ron believes that these towers indicate private family ownership of land with cultivation taking place in small plots, thus reflecting a traditional subsistence economy. A detailed study of towers associated with fields in Samaria dating from Hellenistic and Roman times was made by Dar. Similar structures have been examined in terraced areas around Jerusalem, some of them dating back to the Iron Age ii. Additional tower-like structures in areas of terracing, also dating from the Iron Age ii, have been examined in the Tell el-'Umeiri area in Jordan, to the southwest of Amman. This kind of structure should perhaps be identified with the stone-tower (migdal) mentioned in Isaiah 5:2. Additional structures located in fields are mentioned in the Mishnah, including "cone-shaped huts, watch booths, and summer huts" (Ma'as. 3:7; cf. Kil. 5:3). The sale of a field also meant the sale of the watchman's hut and its stones, unless it was made of perishable materials and "if it was not fastened down with clay" (bb 4:9). The differences between these ancient structures and the more recent ones are not as outstanding as Dar believes. Dar argues that these structures differ architecturally, with the ancient examples built of hewn stone without built stairs and with the recent examples built of surplus stone cleared from the fields. The ancient towers, however, could easily have been supplied with wooden ladders and this would have given them a much better defensive edge than having built stairs. Furthermore, surveys around Jerusalem have shown that many of the traditional towers were in fact built of stone supplied from quarries and not cleared from fields at all. The similarities between the ancient and traditional towers are much greater: both sets of towers were located in fields at a distance from the settlement, they had cool interiors suitable for storing agricultural produce and they could have been used as watchtowers during the harvest seasons. In antiquity, the social standing of the farmer was probably indicated by the size and appearance of the tower. Even some of the more recent towers can be quite imposing. Extremely flimsy structures could also be erected in the fields during harvest times, such as the traditional structures in the fields near Beth Shean (Dalman 1932, Pls. 12–13). The Bible refers to temporary structures of this sort; the melunah in an irrigated plot of cucumbers and the sukkah in a vineyard (Isa. 1:8; 4:6; 24:20).

Fenced-off animal pens are frequently found adjacent to field systems, since arable farmers in Ereẓ Israel usually also kept stock. These were grazed on the common lands and on arable fields when fallow. At harvest the crops were protected by tethering animals or keeping them within fenced enclosures. The study of animal pens is instructive in regard to the relationship in antiquity between livestock to arable activity in a given landscape.

The excavation of rural sites, such as farms and villages, has brought to light the remains of agricultural installations, farming equipment, and food-processing equipment. These provide information on the level of technology and farming methods available to the farmers at different periods. Indirectly they also provide information on what was grown in some of the surrounding fields. Agricultural installations may include wine and oil presses, and large flour mills. The dating of a specific type of installation within a settlement can help towards dating similar installations found associated with the fields outside a settlement. Important studies of wine and oil presses have been undertaken in recent years by various scholars, notably by Frankel. Examples of farming equipment may include digging sticks, plows, mattocks, hoes, sickles, pruning knives, and so forth. These may be compared with traditional farming equipment used in Ereẓ Israel (see, for instance, the work of Avitzur and more recently by Ayalon). Examples of food-processing equipment may include pestles and mortars, querns, rotary mills, and so forth.

[Shimon Gibson (2^nd ed.)]

Plowing

The light plow (or ard) was one of the most important farming implements used in the fields of Ereẓ Israel for the breaking up of ground and preparing it for cultivation. The introduction of this implement into Ereẓ Israel at the end of the fourth millennium bce. probably had a revolutionary effect on the appearance of the earliest fields and the size of the average plot probably increased considerably (the earliest evidence for plowing in the Near East are the plow marks dating to the fourth millennium b.c.e. which were found at Susa A, and the Uruk pictograms of plows which date from the late fourth millennium b.c.e.).

Traditional plowing helps to understand the ancient procedures. Proper plowing ensured a successful harvest. Before plowing was commenced, especially in a mountainous area, the plots were cleared of stone with the use of a hoe or mattock (in a process known in Arabic as naqb). The manner in which the plot of land was plowed has been described by a number of different authorities, among them Wilson, Turkowski, and Avitzur. This operation had to be undertaken properly to ensure a successful harvest, otherwise the crop could be ruined (Job 4:8). Continuous plowing of the entire field was never attempted. Instead, the area was divided into separate plots (each plot known in Arabic as a ma'anah) usually one-third or one-fourth of a faddan and with a length of between 90 and 120 ft. (27 and 36 m.). According to Wilson, a furrow (Arabic tilm) was first run on the ground "and others plowed parallel to this, until a piece of ground of that length and about half the breadth is finished; then a similar piece is plowed next; and so on until the whole is completed." Not only the type of plow but also the type of draft animal used for the plowing could affect the overall depth of the resulting furrows. The maximum depth of plow in modern times was 8 in. (20 cm.) according to Dalman. In antiquity the plowpoint penetrated the soil to a depth of three handbreadths (Gen. R. 31; bb 12:2), i.e., 27 cm., and this helped to maintain the overall fertility of the land (tj, Ta'an. 4:8, 68b).

The relatively small area plowed by the farmer per day in antiquity has been estimated by Feliks (see *Agriculture) to be about 1,170 sq. m. (cf. Tosef., bm 11:9), which is about one third of what is usually covered using traditional Arab methods of plowing. The prophet Isaiah noted that the soil had to be plowed twice before it was sown: the first time to expose the soil to the penetration of rainwater, and the second to level the ground for the planting (Isa. 28:24). The Mishnah, however, enumerates four separate plowings: one in the summer, after the harvest, in the form of broad lines or plots that were set far apart (referred to as the "furrows of pati'aḥ" in the language of the Mishnah; Kil. 2:6). The second plowing took place after the first rains, and then, too, the furrows were not placed close together. Spaces were left intentionally between the furrows to prevent torrential downpours from eroding the soil. In this way each field would have the appearance of alternate furrows and ridges ("gedudim" in the Bible). Rain leveled the ground and was soaked into it. The Psalmist (45:11) prayed for a good year in which rain would soak into the soil and level the "gedudim" in the ground. Following this, the main deep plowing was commenced in which the furrows were placed close together (the "shiddud" of the Bible) and in this way the soil was ready for sowing. Finally, the fourth plowing was undertaken in order to cover the seed itself (Tosef., Kil. 1:15).

To preserve the fertility of the soil, a rotation system was practiced in which land was sown and left to fallow alternately. In the fallow year, the field was plowed five to seven times to rid it of noxious weeds and to restore its fertility. These plowings were called "tiyyuv" or "nir" (Tosef., Shev. 3:10; Men. 85a). A well-plowed field attracted attention owing to its cleanness (Avot 3:7). It was not easy for the farmer to adapt himself to this cycle, since it meant he was only able to plant his field during three years out of seven, the seventh (shevi'it) being the Sabbatical year (Mekh. de-Kaspa 20), but he knew that this was the only way to ensure continuous, abundant harvests. Only artificially-fertilized fields may be plowed year after year with fairly good results.

The plow used in antiquity was not essentially different from that used in traditional Arab farming, except that the earlier implement was sturdier and was capable of penetrating deeper into the soil. The main parts of the plow were made of wood with the plowshare (biblical "et"; talmudic "yated" or "kankan") made of metal (bronze in earlier times; later of iron). Numerous examples of such implements have been uncovered in archaeological excavations in Israel. The metal part was funnel-shaped, ending in a sharp point. The plowshare was attached to the sharp wooden tailpiece ("ḥerev") which in turn was joined to the "knee" ("borekh") and was tied to the handle, a long pole ("yaẓul") attached to the yoke (see Kelim 21:2).

The plowman depressed the handle with one hand. In the other he held a long staff or goad ("malmad ha-bakar"), one end of which held a nail and was used to goad the oxen and hold them in line, while the other end was shaped like a shovel and served to clean the plow ("maḥareshet" in the Bible; "ḥarḥur" in talmudic literature; see i Sam. 13:21; Kelim 13:3). The oxen were tied to the plow by the yoke, which was a pole (the biblical "motaḥ") placed on the neck of the ox or cow. To yoke a pair of oxen, an additional pole was drawn under their necks while pegs ("simyonim") joined the two poles together and thereby enclosed the heads of the oxen in frames. A broken pole could not be repaired, and the animal would have to be released. Hence breaking the yoke symbolized liberation (Jer. 2:20, etc.). Generally the yoke was made of wood; only in exceptional cases was it made of metal, so accordingly an "iron yoke" represented abnormal or tyrannical oppression (Deut. 28:48). A single ox was tied to his yoke by ropes ("moserot," "aguddot motah"). The snapping of these bonds, too, became a metaphor for liberation (Jer. 2:20; Isa. 68:6).

A sturdy strain of oxen capable of bearing a double-poled yoke was used for plowing. Evidence points to the Zebu oxen, capable of sustained exertion, as being in common use in Talmudic times. The Torah regards the ox and the donkey as plowing animals (Deut. 22:10), and Isaiah mentions the use of donkeys for tilling the soil (30:24). Rabbinic literature, however, names only the ox and the cow as plowing animals.

Sowing and Planting

The main crops were the winter grains, which were planted at various times prior to the rains, and especially following the first rain. Usually the farmer spaced his sowing activities at intervals over the winter season as a protective measure (cf. Eccles. 11:6), for if he planted all his crops together, a single adverse natural phenomenon could ruin them all at one blow. The normal planting season lasted from Tishri (October) to the end of Tevet (December; Tosef., Ta'an. 1:7). In some instances, planting also took place in Shevat (January), as was the practice with the barley for the Omer offering (Men. 8:2).

A distinguishing feature of local agriculture in Roman times was the small quantity of seed sown per unit of land, approximately 4–8 kg. of grain per 1,000 sq. m. (Tosef., Kil. 1:16). This is much less than the average amount planted in traditional Arab farming. This probably explains the meaning of the saying: "Thou shalt carry out much seed into the field and shalt gather little in" (Deut. 28:38). The yield of an Arab farmer is 3–4 times the amount of seed, while the Jewish farmer during Roman times reaped 30 or 40 fold, sometimes even obtaining a harvest of "a hundred measures" (bm 105b; Pes. 87b; see also Matt. 13:8; Varro, Rerum Rusticarum Libri, 1:44). These high yields were a result of rational and intensive methods of cultivation.

Hoeing and Weeding

The usual cycle in crop cultivation was plowing, planting, hoeing (tj, Shek. 68:4), with the last activity designed to remove noxious weeds. The main implement involved, the "ma'ader," is somewhat different from the modern hoe. It consisted of two sticks tied together by a cord (Tosef., Kelim, bb 1:8) to form an acute angle. At the end of the shorter stick, a metal "tooth" was inserted (Kelim 13:2). Such hoes frequently appear in Egyptian drawings. For deeper plowing and hoeing, the kardom (Peah, 6:4; Kelim 29:7; cf. ibid., 13:3) was used. An implement resembling the modern hoe, the magrefah, served for moving earth or fruit (Gen. R. 16; Kelim 13:4). The ma'ader was used for digging in mountain areas which could not be successfully plowed (Isa. 7:25; Peah 2:2). The process of deep digging to prepare the earth for saplings was known as "izzuk" (lsa. 5:2; cf. Sif. Deut. 355).

Harvesting

No special importance was ascribed to summer planting in ancient times. Kaẓir applied to winter crops. There were two such harvests: first, the early ripening of the grain, the various types of *barley; later, the *wheat and rice-wheat harvest (Ex. 9:31). At Passover, the Omer of barley was offered (Lev. 23:10; cf. ii Sam. 21:9; ii Kings 4:42). The Gezer calendar also lists the barley as the first of the cereal harvests. Before the Omer was offered, the new season's grain was not to be eaten (Lev. 23:14). Only in the Jericho Valley, was harvesting permitted before Passover (Men.10:8). Seven weeks later, the wheat harvest began with the offering of "the two loaves of bread" (Lev. 23:17).

At harvest time, the climate in Israel is hot and dry. More than once the reaper was felled by sunstroke (cf. ii Kings 4:18–20; Judith 8:2–3). He rose early to take advantage of the cool, morning hours (Prov. 10:6), and he had to work quickly to avoid plunder, pests and the scattering of the grains. He sought, in addition to his family, to employ hired hands. The division of labor is depicted in the Book of Ruth. A supervisor would watch the workers. Girls were occupied with gleaning and also in making sheaves. The owner supplied part of the food to his workers, namely bread dipped in vinegar. Only the wicked who exploited their workers failed to provide food (Job 24:10–11).

Even though the work was backbreaking, it was performed to the accompaniment of joyous shouts (Ps. 126:5–6; Isa. 9:2). The poor, who gathered their gifts, also contributed to the festive atmosphere. Sometimes, joy would be absent and especially when the land was afflicted by drought (Deut. 11:17, etc.) or when an enemy had attacked the reapers and pillaged the harvest (Isa. 16:7).

Various implements were used during the harvesting procedures. The "Ḥermesh" (Deut. 16:9) and "maggal" (Jer. 2:16) are two such implements mentioned in the Bible. The latter is the usual term appearing in rabbinic literature, and it is almost certain that the two names signify the same object, the sickle. The scythe, the long handle of which was grasped with both hands, was not known in ancient Ereẓ Israel. The sickle had a short handle, in which there was inserted a curved blade with its short teeth bent backward. (Ḥul. 7:2). Archaeological excavations in Israel have uncovered flint, bone, bronze and iron sickles. Some sickles extend back in time to the Natufian and Neolithic periods. When harvesting, the reaper grasped the stalks in his left hand. In his right he held the sickle, which he would "send out" (Joel 4:14) and pull back, so severing the stalk (Isa. 17:5; Ps. 129:7). When his left hand was full, he would lay the grain on the ground in united bundles ("ẓavitim"; Ruth 2:16) or else tie them together with a straw ("kerikhiot"; Men. 10:9). The small heaps (ẓevatim or kerikhot) laid along the harvested rows would be gathered up by the sheaf-binder and held in his bosom ("Ḥoẓen"; Ps. 129:7). They were then put together in larger bundles, which had no fixed size, fluctuating between three and 30 pints (½ and 17 liters) of kernels (Pe'ah 6:6). The bundles were left lying on the ground or else tied together in sheaves (Gen. 37:7–8; bm 22a). The next step, which took place once the grain was dry, consisted of collecting the grain in a large stack ("gadish"). An alternative practice was to heap the grain in various types of stacks to hasten or retard the drying process as desired (Pe'ah 5:8). This stacking was the final destination of the grain prior to its being transported to the goren or threshing floor.

The goren stood close to the city or village (i Kings 22:10). To prevent the chaff from being blown into houses, the rabbis ruled that the threshing floor should not be nearer than 50 cubits (i.e., approximately 25 metres) from the city (bb 2:8). The location could neither be on high ground nor exposed to strong winds which would scatter the grains during winnowing (Ruth R. 5). Usually, the threshing floor was in a broad public place. It was surrounded by a fence of thornbushes (Sot. 13a). Trampling (Jer. 51:35) and sprinkling water hardened and leveled the floor. Then the grain was brought and spread out in a circle. During the threshing, the straw and ears were pounded to separate the kernels from the husks. This could be accomplished in several ways. A wooden board, about 0.70 m. wide and 1.20 m. or more long, was used. Its underside was set with stones, mainly of basalt (sometimes of flint), so that when dragged by a pair of oxen, the board would separate the grain. The device, the morag, is still used in traditional Arab farming (Arabic "norg"). Since this tool was only adopted much later in Greece and Rome, some scholars have suggested that it might have been an invention of the farmer in Ereẓ Israel. This assumption may explain the exclamation of Isaiah (end of ch. 28): "This also cometh from the Lord of hosts; wonderful is His counsel and great His wisdom," for in the preceding verses the prophet enumerates the threshing tools then in use: the ḥaruẓ, i.e., the morag ḥaruẓ (ibid. 41:15), the normal morag with saw-like strips of iron in addition to the stones set in it. Isaiah also referred to the "ofan agalah" and "galgal agalah" ("cartwheels") which were stone wheels or iron discs sharpened like saws, examples of which are still extant among the Arab farmers of today. The last type of tool was introduced into Hellenistic Egypt and Rome. The chronicler of Roman agriculture called it "plostellum poenicum" (Varro, Rerum Rusticarum Libri, 1:52). In contrast to mechanical means, another threshing method was the running of several oxen tied together ("revekah"; Tosef., Par. 2:3) over the grains. Seemingly other animals were also used in this type of threshing, for the rabbis interpreted the Pentateuchal prohibition against muzzling an ox while threshing (Deut. 25:4) to include other animals as well (bk 5:7). The heavy implements described above were suitable for wheat and legumes. More delicate grains were normally threshed with a stick (Isa. ibid.) as were smaller quantities of wheat (Judg. 6:11; Ruth 2:17). While the threshing was in progress, additional quantities of grain would constantly be thrown in the path of the threshers by means of a wooden-pronged implement resembling a pitchfork ("eter"; Tosef., Uk. 1:5).

Winnowing

Threshing separated the three components of the grain: kernels, chopped straw, and chaff. Winnowing, which consisted of throwing the threshed substances into the wind, caused the lighter elements to be carried away while the heavier kernels fell into a heap. The implement used for this process, themizreh (Isa. 30:24), resembled a pitchfork with broad prongs. Following this the kernels would be thrown up by means of a shovel-like implement, the raḥat (cf. Tanh. to Isa. ibid.). Once this operation was over, the stack was considered to be complete. The farmer would measure its size (Haggai 2:16) and stand guard over it until it was transferred to the barn (Ruth 3:7).

The chopped straw left over from the winnowing was kept as livestock feed or compost, or was used as an additive in mortar. The chaff was useless except for making fire (Gen. R. 83:3). Yet even after the final winnowing, the kernel heaps still retained waste matter. The grains would then be shaken horizontally in a sieve ("kevarah"), a round device, to whose bottom a fiber net was attached. The heavier waste would fall through the threads, and the lighter material gathered on top of the kernels. The top waste would constantly be removed, until only the clean kernels remained within the sieve (cf. Amos 9:9; Ma'as 15:6). The kernels were then milled or crushed, and further cleaned with the aid of sieves with perforations of various sizes, depending on the required size of the finished product (Avot 5:15).

[Jehuda Feliks /

Shimon Gibson (2^nd ed.)]

add. bibliography:

S. Avitzur, Man and His Work: Historical Atlas of Tools and Workshops in the Holy Land (1976); E. Ayalon, Review, in: Israel Exploration Journal, 55 (2005), 116–20; O. Borowski, Agriculture in Iron Age Israel (1987); M. Broshi, "The Diet of Palestine in the Roman Period – Introductory Notes," in: The Israel Museum Journal, 5 (1986), 41–56; M. Broshi, "Agriculture and Economy in Roman Palestine: Seven Notes on Babatha's Archive,"in: Israel Exploration Journal, 42 (1992), 230–40; G. Dalman, Arbeit und Sitte in Palastina, vols. 1–7 (1928–42); S. Dar, Landscape and Pattern: An Archaeological Survey of Samaria, 800 b.c.e.–636 c.e.bar International Series 308 (1986); C. Dauphin, "Man Makes His Landscape," in: Bulletin of the Anglo-Israel Archaeological Society, 11 (1991–92), 22–28; J. Feliks, Agriculture in Eretz-Israel in the Period of the Bible and Talmud (1990); R. Frankel, Wine and Oil Production in Antiquity in Israel and Other Mediterranean Countries (1999); R. Frankel, S. Avitsur, and E. Ayalon, History and Technology of Olive Oil in the Holy Land (1994); C.H.J. de Geus, "The Importance of Archaeological Research into the Palestinian Agricultural Terraces, with an Excursus on the Hebrew Word gbi," in: Palestine Exploration Quarterly, 107 (1975), 65–74; S. Gibson and G. Edelstein, "Investigating Jerusalem's Rural Landscape," in: Levant, 17 (1985), 139–55; idem, "Agricultural Terraces and Settlement Expansion in the Highlands of Early Iron Age Palestine: Is there Any Correlation Between the Two?" in: A. Mazar (ed.), Studies in the Archaeology of the Iron Age in Israel and Jordan (2001); D.C. Hopkins, The Highlands of Canaan: Agricultural Life in the Early Iron Age (1985); A. Kasher, A. Oppenheimer, and U. Rappaport (eds.), Man and Land in Eretz-Israel in Antiquity (1986); A.M. Maier, S.Dar, and Z. Safrai, The Rural Landscape of Ancient Israel. (2003); Z.Y.D. Ron, "Agricultural Terraces in the Judaean Mountains," iej, 16 (1966), 33–49, 111–22; idem, "Development and Management of Irrigation Systems in Mountain Regions of the Holy Land," in: Transactions of the Institute of British Geographers, n.s. 10 (1985), 149–69; Z. Safrai, The Economy of Roman Palestine. (1994); D. Sperber, Roman Palestine 200–400: The Land. Crisis and Change in Agrarian Society as Reflected in Rabbinic Sources. (1978); L.E. Stager, "The Archaeology of the Family in Ancient Israel," in: basor, 260 (1985), 1–35; G. Stanhill, "The Fellah's Farm: An Autarkic Agro System," Agro–Ecosystems, 4 (1978), 438.